Skin neoplasms characterization through standardized medical thermography

Faculdade de Engenharia da Universidade do Porto Skin Neoplasms Characterization Through Standardized Medical Thermography Liliana Vasconcelos Esteves Dissertação realizada no âmbito do Mestrado Integrado em Bioengenharia Ramo de Engenharia Biomédica Orientador: Prof. Dr. Ricardo Ângelo Rosa Vardasca, Ph.D. AMBCS September 2014 ii © Liliana Vasconcelos Esteves, 2014 iii "Success requires no explanations. Failure permits no alibis." Napoleon Hill iv v Abstract The systematic process of skin cells regeneration can be affected of several factors causing an abnormal grow of new cells. The excess of cells may form a mass of tissue called neoplasm. The neoplasm may be benign or malignant. Benign neoplasms rarely require treatment, just periodic checkups are recommended to control and to ensure that they do not evolve into malignancy. Within the malignant neoplasms, the most relevant are melanoma, basal cell carcinoma and squamous cell carcinoma which have effective treatments. Contrary, malignant melanoma can only be remised with an aggressive treatment in an early phase. According World Health Organization (WHO), the incidence of this type of cancer in the European Union (27 countries) are 12.2 persons per 1M inhabitants, the mortality is approximately 2.2 persons, and the prevalence at 1st, 2nd and 3rd years are 12 %, 34.1 % and 53.9 % respectively. Thus, it is necessary to find new cost-effective tools that can contribute to early diagnosis or treatment assessment. Medical thermography is a noninvasive, non-ionizing, non-contact and reliable medical imaging method, which allows monitoring the temperature distribution on the skin surface. This information is directly related to the physiology, enabling a real-time monitoring, especially in peripheral microvascular and autonomous nervous systems. This study aims to use the medical thermography for physiological characterization of skin neoplasms. It is a retrospective and explorative study consisting of a control group and a group of malignancy holders. Skin surface temperature was characterized according to the dermatomes location of the autonomous nervous system and main blood vessels location of the microvascular system. The relationship of these characteristics with the type of recurrent skin neoplasms was assessed. It was possible to characterize the temperature of skin neoplasms, facilitating early diagnosis, targeting adequate treatments and a substantial reduction of associated costs. Keywords: Medical thermography; Monitoring; Skin cancer; Skin surface temperature; Skin neoplasms. vi vii Resumo O processo sistemático de regeneração das células da pele pode ser afetado por vários fatores causando um crescimento anormal de novas células. Esse excesso pode formar uma massa de tecido designada por neoplasia (benignas ou malignas). As benignas raramente necessitam de tratamento, apenas são recomendados exames periódicos para controlo, de forma a garantir que não evoluem para malignas. Das neoplasias malignas destacam-se os melanomas, os carcinomas basocelulares e os espinho-celulares. Os carcinomas basocelulares e espinocelular são os mais comuns, tendo um tratamento bastante eficaz. Pelo contrário, o melanoma apenas pode entrar em remissão, com um diagnóstico precoce e tratamento agressivo. A incidência deste tipo de neoplasia na União Europeia – 27 países (UE27) é de 12.2 pessoas por 100000 habitantes, a mortalidade cerca de 2.8 pessoas e a prevalência no 1º, 2º e 3º ano é de 12%, 34.1% e 53.9% respetivamente. Assim sendo, é necessário encontrar novas ferramentas que contribuam para um diagnóstico precoce ou avaliação de tratamento e que sejam financeiramente atrativas. A termografia médica é um método de imagem clínica não invasivo, não ionizante e preciso, que permite monitorizar a distribuição da temperatura à superfície da pele. Esta informação está diretamente relacionada com a fisiologia, permitindo a sua monitorização em tempo real, nomeadamente dos sistemas microvascular periférico e nervoso autónomo. Esta investigação teve como finalidade utilizar esta técnica para avaliação fisiológica, caracterizando as lesões neoplásicas da pele. Trata-se de um estudo retrospetivo e explorativo constituído por um grupo de controlo e indivíduos detentores de neoplasias. Usando a termografia de forma padronizada, caracterizou-se a temperatura à superfície da pele de acordo com a localização dos dermatomas do sistema nervoso autónomo, e com a localização dos vasos sanguíneos principais do sistema microvascular de cada indivíduo. Avaliou-se também a possível relação destas características com o tipo de neoplasia recorrente nos indivíduos. Foi possível caracterizar a temperatura das neoplasias, facilitando o diagnóstico atempado, tratamento direcionado e uma redução substancial dos custos associados a estas patologias. Palavras-chave: Cancro de pele , Monitorização, Neoplasias de pele, Termografia médica, Temperatura superficial da pele. viii ix Acknowledgments To my parents and brother for all the encouragement and support in all my choices and decisions. To my best friend and boyfriend, João Ribeiro, for all patience, help, care, love and strength in these difficult times. With you I can go further and topple giants, being more than I had ever imagined. To my friends, in special Ana Marques, for their friendship and for being always by my side, giving me words of comfort and encouragement. To my supervisor Ricardo Vardasca and, the head of the Laboratory of Automation and Instrumentation in Biomedicine at FEUP, Prof. Joaquim Gabriel for the teaching and knowledge transmitted. To Dr. Marco Rebelo and Matilde Ribeiro for all of the support and for making it possible to put this study in practice at IPO-Porto. To all the volunteers of Porto University and IPO-Porto who showed willingness to take part in this study. xvi Table 23: Results of p-value, U, W and Z values of U Mann Whitney test between moments for thermal symmetry. .................................................................................... 51 Table 24: Obtained results for each characteristic to median temperatures of aright and lefts sides and ΔTs for diagnosed group. ................................................................. 53 Table 25: Results follow-up group characteristic to right and lefts sides and ΔT. ............... 53 Table 26: U Mann Whitney test results in terms of p-value and ranking values for diagnosed group. .................................................................................................... 54 Table 27: Kruskal-wallis test results in terms of p-value for diagnosed group. .................. 54 Table 28: U Mann Whitney test results. ................................................................. 54 Table 29: Kruskal-wallis test results in terms of p-value. ............................................ 54 Table 30: Prevalence of malignancy by age classes. .................................................. 55 Table 31: Prevalence of malignancy in gender. ........................................................ 55 Table 32: Prevalence of malignancy by BMI classes. .................................................. 55 Table 33: Prevalence of malignancy by sun sensibility classes. ..................................... 55 Table 34: Results of U Mann-Whitney test. ............................................................. 57 xvii Acronyms and Symbols ANS Autonomous Nervous System BMI Body Mass Index FEUP Faculdade de Engenharia da Universidade do Porto IPO-Porto Instituto Português de Oncologia do Porto ITI Intraoperative Thermal Imaging MS Microvascular System ROIs Regions of Interest TS Thermal symmetry ΔT Temperature difference between the left and the right side of skin neoplasms U U-Mann Whitney W Wilcoxon W Z Wilcoxon Z Chapter 1. 1.1 Introduction Skin neoplasms are an abnormal mass of tissue that results from an abnormal growth, or cell division. There are two types of skin neoplasms, benign and malignant, also called skin cancers. A benign neoplasm is localized, manifests itself by growing above the skin surface, does not invade adjacent tissues, does not spread to other parts of the body, and can easily be removed, rarely puts the patient’s life at risk. On the other hand, the malignant neoplasm presents an uncontrolled growth of abnormal skin cells. This situation occurs when unrepaired DNA damaged skin cells (most often caused by ultraviolet radiation from sunshine or tanning beds), triggers mutations, or genetic defects, which lead the skin cells to multiply rapidly. The main types of skin malignant neoplasms are: basal cell carcinoma and squamous cell carcinoma (non-melanoma skin cancer) and melanoma [1]. Table 1 shows the countries that have a higher incidence of skin cancer: Australia, New Zeeland, Switzerland, Denmark and Norway [2]. Table 1: Melanoma incidence rate in the 5 countries with more pathology prevalence [2]. Country Skin cancer incidence rate (1 to 100000) Australia 36.7 New Zeeland 36.2 Switzerland 25.8 Denmark 24.1 Norway 25.3 Portugal 8.2 UE27 13 The incidence of both, non-melanoma and melanoma skin cancer, has been increasing over the past decades. Currently, between 2 and 3 million non-melanoma skin cancers and 132 thousands melanoma skin cancers occur globally each year. Basal cell carcinoma is the most common form of skin cancer and the squamous cell carcinoma is the second. Melanoma is generally the most serious form of malignant neoplasms because it tends to spread (metastasize) throughout the body quickly. According to the American Cancer Society, one person dies of melanoma approximately every hour [5]. 2 In Portugal, there are about 700 new cases of melanoma annually [3]. The incidence of this type of cancer in the European Union of 27 countries (EU27) is of 13 persons per 100000 people, the mortality is approximately 2.2 persons, and the prevalence at 1st year is 12%, in the 2nd year is 34.1% and finally, in the 3rd year is 53.9% respectively [3, 4]. This year an estimated of 76100 new cases of invasive melanoma will be diagnosed in the United States and 9710 people will die of melanoma [5]. In United States, the cost of skin cancer treatment varies, depending on the size, type, depth and location of the lesions, and, most important, the stage in which it was detected. One study published in the Archives of Dermatology found average treatment costs for skin cancers to range from 1732 to 56059 United States Dollars (USD) depending on the stage of the disease [6]. The costs associated with melanomas treatment exceed USD 740 million annually [7]. In 2009, the average monthly cost per patient with melanoma was approximately about USD 2194 during the first four months of treatment. For people aged 65 or older, the annual cost was around USD 390 million [8, 9]. Australia is the country in the world with the highest incidence of melanoma patients. Every year, 10430 new cases are identified. The costs associated with cancer in this country are of 3.8 billion Australian Dollars (AUD) per year in direct costs of health care. Worrying is the fact that only 1% (AUD 30 million) are spent on treatments of melanomas [8]. Nonmelanoma skin cancers have cost to the Australian health system more than AUD 500 million in 2010 and are expected to cost AUD 700 million by 2015. Medical Thermography is an imaging technique that allows mapping the thermal radiation emitted by the surface of the human body skin, which under some conditions may reflect abnormal or normal situations of human physiology. Thermography is technique a noninvasive, non-contact, non-ionizing and therefore safe for the patients. This technique is passive, and relies on the registration of the thermal radiation emitted by the surface of the skin from various parts of the body through a set of appropriate sensors that create an image of the surface temperature. The capture protocol recommended for medical thermography defines the environment in which an examination should be performed; the ambient temperature should be between 18 and 24 °C, relative humidity be less than 50%, the air flow should be less than 2 m/s and there should be absence of light incident upon the subject [10– 12]. 1.2. Aim This work aims to investigate the use of medical thermography for physiological assessment, characterization and monitoring skin neoplasms in several areas of the body, on IPO patients. 3 Two different studies will be made. The first has the objective to characterize the temperature of skin surface in asymptomatic individuals, based on dermatomes distribution of the Autonomous Nervous System (ANS) and in the perforated blood vessels of the Microvascular System (MVS), and their relationships with individual’s characteristics. In this group a wind stimulus (fan) will be applied during a period of two minutes, in order to evaluate the physiological response of the face. Thus, it was possible to perform a characterization study of the skin surface temperatures of the two systems (ANS and MVS). The second study, will be focused on the characterization of the skin surface temperature in patients with various skin neoplasms (benign or malign) and individuals who have already undergone in treatment. It will also be evaluate the relationship between individual’s characteristics and skin neoplasms type. In addition, a comparison case study will be performed with basal cell carcinoma individuals and a control group with similar BMI and gender. The results of this research will enable improvements in early diagnosis of skin neoplasms, targeting adequate therapies, and leading to a reduction of these pathologies associated costs. 1.3. Dissertation Outline This document is organized in six sections. The first introduces the topic and characterization of the problem and its context. The literature review describes additional information found in the literature, which is relevant for the present work. The third section, outline the materials and methods used during this research. The fourth section presents the results for each experiment. In the fifth section, there is a discussion and analyzes of the significance of the obtained results. Finally, the conclusion addresses the final considerations and main achievements, as well as the future work. The structure of the dissertation and the links between different parts of this document can be seen in figure 1. 4.2 Individuals Characteristics and Skin Temperature 5.2 Individuals Characteristics and Skin Temperature 4.3 Individuals Characteristics and Malignancy 5.3 Individuals Characteristics and Malignancy Figure 1: Dissertation Outline. Chapter 2. Literature review As a start, it is important to understand the anatomy of the skin, nerves, arteries and veins, the role of skin in thermoregulation, and the neoplasms associated with this part of the body. It is also necessary to understand their risk factors, treatment and diagnosis. Skin Neoplasm is a lump or growth that is formed from abnormal cells. Thermology is the science that studies temperature and its exchange process. It covers all thermal energy transfer processes such as: heating or cooling. Thus, this encompasses how heat can be exchanged between objects, as well as the features that this process involves. Within this chapter and together with these concepts, a brief introduction of thermography and its physical principles, history and applications in medicine are addressed. The alternative techniques to thermography are also outlined. 2.1 Human skin The skin is the outer layer of the human body, and his biggest organ, with over 2 m2. This extensive area has external environment interaction like sun exposure, resulting on a excessive water loss and insulation [10]. It is composed of a complex structure system: dermis, epidermis and hypodermis [13–15]. The skin possesses, per cm2, about 650 sweat glands, 20 blood vessels, over a thousand hair follicles, half a million melanocytes, more than a thousand nerve endings, and also sebaceous glands [10]. The skin protects the body from heat, injury and infection, as well as damage caused by UV radiation from the sun. It stores water and fat, which participates in homeostasis, and plays an important role in thermoregulation [16–18]. The epidermis is the outer layer of the skin, composed mostly of squamous cells on the top and basal cells and scattered melanocytes on the bottom. Melanocytes are responsible for melanin production, the main pigment responsible for skin colour. When the human body is exposed to UV radiation, melanocytes acquire pigmentation, becoming darker [10, 11, 17, 19]. The dermis is the layer under the epidermis, thicker than the latter. The main components of this layer are collagen and elastin fibers. The dermis contains various types of 7 cells, such as blood vessels, lymphatic vessels and glands. Some of these glands are responsible for the production of sweat, which allows the human body to cool down. Others produce sebum, which helps keep the body dry. Both reach the surface of the skin through little orifices named pores [10, 13, 17, 19]. Figure 2 shows the different layers of skin, namely epidermis and dermis. Figure 2: Different layers of the skin, dermis, epidermis and constituents cells [13]. 2.1.1 The role of the skin in thermoregulation Thermoregulation is the ability of an organism to maintain its core body temperature when the temperature of the surrounding environment is different [14]. When the core body temperature increases significantly, it is called hyperthermia, when the opposite occurs it is called hypothermia. The thermoregulation process maintains the core body temperature unchanged, within a constant range, varying only about 0.6ºC, even when the organism is exposed to major cold or heat variations [21]. Thermoregulation is performed by a physiological control system, consisting of the central nervous system and peripheral thermoreceptors, an afferent conduction system, a central control for the integration of thermal impulses and an efferent response system which leads to compensatory responses [21]. The hypothalamus lies at the central nervous control system (brain) and regulates the body core temperature by integrating thermal impulses coming from peripheral thermoreceptors. The hypothalamus receives inputs from two sets of thermoreceptors, receptors which monitor the temperature of the blood that goes through the brain, and skin receptors, which monitor exterior temperature. Both supply the necessary information for the body to make appropriate adjustments. When the integrated impulse exceeds or falls below the temperature range, autonomic thermoregulatory responses happen in order to maintain appropriate body temperature. Due to skin relationship with the circulatory system, skin vessels dilate whenever the temperature rises above the expected, favouring heat exchange with the environment. This process is called vasodilation. Vasoconstriction is the opposite of vasodilation. In case of the 14 Table 2:Different diagnosis techniques and treatment options to the various types of skin cancer [5, 17]. Types Diagnosis Treatment Melanoma Medical history and physical exam (ABCD and Woods examination) Biopsy (punch or excisional or incisional) to confirm. To determine the extent: Sentinel lymph node biopsy Imaging tests (CT, MRI and PET) Surgery Chemotherapy Immunotherapy Radiation therapy Targeted therapies (vemurafenib, trametinib, dabrafenib, and ipilimumab) Basal cell carcinoma Medical history and physical exam Biopsy (punch or excisional or incisional) to confirm. Surgery (excision, Mosh surgery and Cryosurgery) Curettage and electrodessication Topical treatments (imiquimod and fluorouracil) Medication – vismodegib (adavance stage) Squamous cell carcinoma Medical history and physical exam Biopsy (punch or excisional or incisional) to confirm. Surgery (excision, Mosh surgery and Cryosurgery) Curettage and electrodessication Radiation therapy Photodynamic therapy Laser therapy Chemotherapy creams (5-fluorouracil) Merkel cell carcinoma Medical history and physical exam Biopsy (punch or excisional or incisional) to confirm. To determine the extent: Sentinel lymph node biopsy Imaging tests (CT, MRI and PET) Surgery (excision and Mosh surgery) Lymphadenectomy Radiation therapy Chemotherapy 2.2 Thermology Thermal energy transfer can occur through four different mechanisms individually, or often simultaneously: conduction, convection, radiation and evaporation. The necessary condition for a heat transfer is the existence of a temperature differential or thermal imbalance. Conduction can be described as a mechanism that transfers thermal energy through a contact between two solid masses [51]. Convection is a process of thermal energy transfer between a solid surface and an adjacent moving substance (liquid or gas). In natural convection, the fluid motion is due to density difference, which is caused by temperature difference. In induced convection, the fluid motion is caused by external forces (for example a ventilator) [51]. Energy transfer by radiation results from the radiant emission (electromagnetic or photonic) by a body with a temperature above -273.15 °C (0 Kelvin). Radiation is a volumetric phenomenon because it is related to the system volume. All solids, liquids and gases emit, absorb, or transmit radiation in various degrees. In solids, however, the radiation is normally considered as a superficial phenomenon. This happens because the internally emitted radiation never reaches the surface and the incident absorbed radiation reaches only a few microns of thickness surface [51]. 15 Thermography is a technique that allows mapping a body or region skin surface temperature. The use of thermography has various applications such as electrical installations, testing of structures and entire buildings, meteorology and medicine [54]. 2.2.1.1 Physical principles The amount of radiated electromagnetic energy of an object is directly related to its temperature. An object can be characterized by its capacity of absorbing or emitting electromagnetic radiation [55]. Infrared radiation occupies in the spectrum the region between the visible and microwave radiations. All objects in the universe emit radiation in the infrared region as a temperature function. When an object becomes warmer, it emits a more intense infrared radiation and radiates with a shorter wavelength. Figure 9 illustrates the electromagnetic spectrum in terms of frequency and wavelength. The human eye cannot detect infrared rays [55, 57-59]. Figure 9:Frequency and wavelength of the electromagnetic spectrum [56]. Stephan-Boltzmann law states that the total energy radiated by a black body, both per unit of surface area and unit of time is directly proportional to the fourth potency of its thermodynamic temperature T. Stephan-Boltzmann law can be expressed as: 𝑅𝑡= 𝜎𝑇4 (1) where 𝑅𝑡 represents the total radiance (w/𝑚2), σ the Stephan-Boltzmann constant (5,6697x10−8 𝑊. 𝑚−2. 𝐾−4), and T the absolute temperature emitted by the material (K). 2.2.1.2 Infrared Cameras Although infrared radiation is not perceptible to the human eye, infrared cameras are able to convert it into an image that depicts thermal variations across an object surface. 16 Thus, infrared cameras are instruments used in thermography to monitor and record the surface temperature of an object [55]. Thermal cameras can have lens with different angles, the normal angle is 24º (for closer distances) and the wide angle lenses 45º (for longer distances) [62]. It is also important to consider other features in respect to these cameras, such as temperature range, thermal sensitivity, thermal resolution, spatial resolution, accuracy and responsivity. The temperature range is the sensor ability to detect heat waves along a certain range of temperatures. Thermal sensitivity defines the ability to distinguish changes between energy levels. It is expressed in terms of temperature. A low value implies high sensitivity. The most commonly used methods for thermal sensitivity are the noise equivalent temperature difference (NETD) and the minimum detectable temperature (MDT). The minimum resolution of an instrument is defined as the smallest difference that the instrument is capable to distinguishing. The resolution can be quantified in two ways, minimum resolvable temperature difference (MRTD) and slit response function (SRF). Spatial resolution is the number of the pixels of image, given by columns and lines. Accuracy is the measure of how close the object temperature is to the true value (typically ± 2% of the overall reading) [57]. 2.2.2 Medical Thermography Contact thermometry for many years was the only available method to determine human temperature. Thermometry was slowly developed by Galileo experiments in 1592 [63, 64]. Sanctorio Sanctorius, an Italian physiologist, who invented a clinical mouth thermometer, gave a major step. With this new technology Sanctorius performed quantitative experiments with temperature, respiration, weight, and insensitive perspiration, which laid the foundation for metabolism investigation. Carl Wunderlich, a German physiology and diagnostic methodologist expert became known for characterizing several conditions [65, 66]. In 1800, William Herschel, an amateur astronomer, discovered the existence of infrared radiation. In 1840, after the death of William, John Herschel, his son drew a simple picture of the evaporation process due to a mixture of carbon and alcohol using focus sunlight, which he called "thermogram" [67]. JD Hardy, an American physiologist, in 1935, demonstrated that the surface of human skin has similar characteristics of a blackbody radiator, being highly efficient at radiating heat exchange [68]. 17 Meanwhile, Professor Samuel Langley provided a major contribution by developing the bolometer, an instrument used in the measurement of incident electromagnetic radiation. This tool forms the basis for a new generation of heat sensors technology. In UK, Dr. Max Cade built an infrared thermographer, using an indium antimonide electronic detector (InSb). Then, it was created in 1960 the Pyroscan, the first British medical thermographer [67]. The first study published with the thermography application in medicine, was performed by Ray Lawson et al., in 1956, where patients with breast cancer were examined by asymmetric hot spots and vascularity in medical thermography images [69]. The role of this technology in medicine has increased [33, 71–73]. Such advances include a new generation of infrared technology, intelligent image processing algorithms, and better physiological understanding of the captured images. Thus, thermography becomes a new way to monitor the temperature, non-invasive, non-ionizing, unexpensive, convenient and well accepted by the patient. Medical thermography is a technique based on skin surface temperature analysis (figure 10). Figure 10: Thermography of woman face in two different positions. This technique is passive, and relies in the registration of the thermal radiation emitted by the skin surface through an appropriate instrument that creates an image of the surface’s temperature. The protocol recommended for medical thermography defines that the environment temperature should be between 18 and 24 ºC to facilitate the thermal equilibrium [9, 74, 75]. 2.2.2.1 Thermography in Dermatology The clinical use of thermography in dermatology was first reported by Brasfield et al. [76] in 1964, which indicated the hyper-thermic finding of cutaneous melanoma. In the following years, many researchers [76-98] had been concerned to investigate the application of medical infrared imaging in the area of dermatology. Thermography research has its focus in four main fields: melanoma diagnosis, follow-up of skin cancer, basal cell carcinoma diagnosis and temperature differences in skin surface. 18 2.2.2.1.1 Melanoma Diagnosis In 1977, Tapernoux B. et al. [77], investigated thermography as a tool for the diagnosis of malignant melanoma, and demonstrated a correlation between hyperthermia and malignancy in about two-thirds of the cases examined. Similarly, Diem et al. [78] evaluated the ability of contact thermography in cutaneous melanomas diagnosis. It was suggested that this technique should be used only as complementary method. It was concluded that positive thermograms should lead to a malignancy diagnosis without being pathyonomic of malignancy. In contrast, Cristofolini et al. [79] suggested that thermography is useless for diagnosis and follow-up of cutaneous malignant melanomas. They concluded that false-negative and false-positive thermograms of metastatic lymph nodes, subcutaneous nodules and removal site of cutaneous malignant melanomas were too numerous. Hartmann M. et al. [80] studied telethermography as a diagnosis of malignant melanoma method. In addition, a complete body thermography is complementary to established examination techniques used in evaluation of patients with malignant melanomas. By making visible hypoor hyper-thermic areas, it helps to detect metastases and confirms or denies suspicious findings obtained by other examination methods. In 1985, Michel U. et al [81] used infrared thermography in a group of clinically suspected patients of having primary malignant melanoma, suggesting that the larger malignant melanoma was warmer in comparison with the smallest. The highest measurements of infrared radiation were obtained in the nodular and acrolentiginous malignant melanoma, when compared to other melanoma types. Later, studies of Novak OP. et al. [82] proved that the examination thermographic method was very effective in complex situations, such as patients with complex skin melanomas, early tumour metastasis and recurrence of operative treatment. Di Carlo et al. [83] described that the usage of different types of thermography, for example, contact thermography or tele-thermography and their clinical applications, with emphasis to melanoma. Tele-thermography is based on the capture at a distance of the infrared energy emitted by the cutaneous surface, and the conversion of this into a video signal. He suggested the application of thermo stimulation assisted tele-thermography to solve the problem of false-negative results reported by other authors [79, 84]. By this procedure it was possible to evaluate the thermogenic capacity of skin lesions against that of the skin in which they are seated, based on the different times of visualization on the screen. In the case of malignant cutaneous tumours, when the skin area concerned is cooled, a very quick hot spot of the tumour follows, given its high esothermic metabolism, whereas the hypothermia of the healthy skin protracts. It was concluded that thermo stimulation assisted tele-thermography can be considered a valid and very useful diagnostic tool, not only in the 19 diagnosis of skin tumours (particularly melanomas), but also in more extensive applications concerning the microcirculation physiopathology. Buzug TM. et al. [85] investigated the early diagnosis of malignant skin cancer by functional infrared imaging. It was assumed that malignant melanoma had higher metabolism, increasing blood flow and presented slightly higher temperature when compared to the healthy skin. It was concluded that this method turns out to be promising, especially in the case of basal-cell carcinoma, because it yields a clear diagnostic result. Cila Herman et al. [86] explored the feasibility of infrared imaging as a cost-effective and non-invasive technique. It was used in optical measurements for tumours detection, particularly on the screening and early detection of melanoma. The data presented were obtained in patients who had a pigmented lesion previously diagnosed by a biopsy. They compared the differences in thermal responses between normal and malignant tissue with the results obtained by the biopsy. They used a protocol for image acquisition that involves various steps. In short, they applied a stream of cold air to the area of the patient’s skin containing the lesion as well as a 50 mm diameter-surrounding region, for the duration of one minute. After that, they remove the cooling stress to allow the skin to re-warm at room temperature. During the thermal recovery phase, infrared images of pigmented lesion were captured every 2 seconds. Finally, all IR images were saved and stored using the LabVIEW software. IR images are analyzed using a Matlab code in order to obtain accurate transient temperature distributions on the skin surface. The conclusion was that increased metabolic activity of melanoma lesion could be detected through dynamic infrared imaging. Later, in 2012, Amber L. Shada et al. [87] tried to obtain preliminary estimation of the sensitivity of infrared images for the melanoma metastasis detection, and also to detect metastases, primary on melanomas and nevi. It was also evaluated the reliability of thermography in terms of time logistic constraints and patients tolerability. The conclusion was that melanomas with more than 15 mm of diameter are able to be distinguished from other lesions with excellent sensitivity and specificity (95% and 100% respectively). On the other hand, thermography proved to be tolerant and highly reliable in clinical scenario. Finally, they also observed a great potential of this technique in the evaluation of tumour response to systemic therapy. Recently, in 2014 Mathias Bonmarin et al. [88] evaluated lock-in thermal imaging for the early-stage detection of cutaneous melanoma. They used a computational model to derive the transient temperature of skin surface subjected to periodic convective heat transfer. The technique demonstrates a great ability to reject disrupting thermal signals coming from subcutaneous circulatory and metabolic variations. It was concluded that lock-in thermal imaging exhibit a high signal-to-noise ratio and due to its ability on suppress lateral heat spreading, allows an accurate detection of lesion margins. 20 2.2.2.1.2 Follow-up of skin cancer Bourjat P. et al. [89] investigated various malignant melanoma cases which had not been subject of surgical intervention, and proposed a prognostical classification that enables a malignancy evaluation. They also investigated thermography in follow-up detection of local recurrences and/or lymph-node metastases in an infra-clinical state. Later, Michel U. et al. [90] performed the same study, but with patients with malignant melanomas. They used the preoperative infrared tele-thermograms that were obtained from the draining lymph node, in order to be compared with the clinical and histologic research data. They concluded that the majority of cases (around 95%) with metastic spread of malignant melanoma had a considerable degree of hyperthermic radiation in the respective axillary or inguinal area. In 2009, G. A. Santa Cruz et al. [91] investigated the ability of dynamic thermography in monitoring melanoma patients who are treated with Boron neutron capture therapy (BNCT). It was explored the potential of this technique to evaluate early changes in normal skin, as well as the evolution of the tumour after treatment. These authors concluded that thermography might be a useful and sensitive tool to study skin toxicity and control of melanoma treatments with BNCT. This technique can also help locate abnormal regions with high temperature as nodules of melanoma, which are virtually invisible on CT images because of their small contrast differences when compared to normal tissue. In the same year, Babak Kateb et al. [92], briefly discussed the current status of intraoperative imaging modalities including Intraoperative Thermal Imaging (ITI). It was also presented a case report that evaluated the usefulness of ITI in brain tumour detection and its margins, more specifically in a patient with a metastatic intracortical melanoma. The results of this study suggested that thermal imaging could be used to provide a rapid, non-invasive and real-time intra-operative technique. They concluded that ITI was a cost-efficient imaging method to assist surgeons in identifying tumour location and potentially detect vasculature and other acute abnormalities during the surgery. Mikulska et al. [93] performed a comparative study with dermoscopy, thermography and histopathology of cutaneous melanocytic nevi. They concluded that thermography may supplement dermoscopy during examination of atypical melanocytic nevi and other skin lesions suspected of neoplastic metaplasia. Additionally, average temperatures of benign and atypical melanocytic nevi do not differ significantly from the average temperature of the skin without pigmented lesions. A transient thermal imaging system was developed in 2011 by M. Pirtini Çetingul. [1], that allows a more accurate measurement of temperature differences in skin surface. The aim was to aid the detection and diagnosis of metabolically active or malignant skin lesions. It was observed that the feasibility of this technique to distinguish benign from malignant pigmented lesions, and thus, quantify the malignant potential of lesions. The data obtained 21 from benign and malignant images was compared, and more specifically, correlated the thermal recovery images and data with biopsy results to define thermal associations with malignant potential. Geometrical data obtained from the digital and dermoscopy images was also used and a biopsy report in its computational model to simulate the thermal response of the analyzed lesion. At the end, the results from the modelling effort were compared with clinical data. The thermal response of benign lesions was found to be similar to that of healthy skin tissue. This difference can be used to identify malignant lesions and quantify their malignant potential. On the other hand, the computed results showed that the lesion parameters and properties can be estimated and the influence of malignant lesion on the transient thermal response can be quantified with their computational model. 2.2.2.1.3 Basal Cell Carcinoma Diagnosis In 1981, several studies were performed. Ratz JL et al. [94] described Liquid-crystal thermography as it is applied to basal-cell carcinomas. Results indicated that it is useful for moderate to heavy cellular density neoplasms but little help in outlining neoplasms of low cellular density like morpheaform basal-cell carcinomas. In 2011, J.H.Flores Sahagun et al. [95] performed an evaluation of infrared imaging viability in the basal cell carcinoma diagnosis using a normalized conjugated gradients methodology. It was proposed a methodology that has been tested by temperature measurements using infrared imaging and a healthy control group to establish expected normality ranges of basal cell carcinoma patients. The proposed method allows the comparison of measured dimensionless temperature difference values between two symmetric regions of patient’s body, taking into account the skin, the surrounding ambient and the individual core temperatures. It was concluded that the normalized conjugated gradients method was simple and capable to differentiate normal from injured tissue and also the conjugated gradients method was efficient to identify injured tissue. It also concluded that the use of dimensionless temperature compensates the temperature readings with respect to variations in the individual’s metabolism and ambient temperature at the particular location of the exam. 2.2.2.1.4 Temperature differences in skin surface Amalric et al. [96] investigated the ability of dynamic tele-thermography in the direct observation of infrared rays emission from the skin of the human body. A number of dermatological diseases such as malignant melanomas can modify this emission and show a larger variation in skin temperature. They defended that other skin black tumours may also benefit from this method. M. Pirtini Çetingul jointly with other researchers had published several interesting studies [1, 97, 98] with the aim of helping the detection of early malignant lesions in the skin. In 2008, a study [31] based in the use of transient thermal response of skin layers, allowed the determination of the temperature distribution that reflect the properties of subsurface 22 structures such as lesions. The evaluation of this study was performed by sensitivity metric of the following variables: the surface temperature distribution, variations of thermo-physical properties, blood perfusion rate and thickness of skin layer. First, the model was used to investigate the importance of the aforementioned parameters for surface temperature distribution. Next, the time evolution of the infrared signal was analyzed after a cold stress that was applied to human skin. By considering both, the steady state and transient results, information about the size and depth of masses within the skin layers was registered during recovery. It was observed that variations in these parameters had minimal effects on surface temperature distribution. By selecting the duration of cooling stress, the internal temperature of the skin changes, and this change can be felt at different depths. One year later, the same investigators [98] developed a thermographic system that allows accurate measurement of small temperature differences on the skin surface. This study’s purpose was mainly diagnose malignant and pigmentation skin lesions in an early stage of the disease. The method developed by these authors was based on active thermography and a multimodal image analysis strategy, including correction of involuntary body movements and segmentation of pigmented lesions in order to detect malignancy. The system described in this study was tested in a group of patients who had pigmented lesions and also clinical indication for biopsy. The lesions and surrounding tissue received a source of cool air for a period of time (between 30 and 60 seconds). The thermal recovery was recorded by an infrared camera, generating an image of the lesion after the application of the cold source. It was concluded that the benign lesions had a similar thermal recovery to normal skin, while melanomas presented a higher temperature than healthy skin during the process of thermal recovery. 2.2.2.2 Thermal symmetry A region of interest (ROI) is a selected subset of samples within a dataset identified for a particular purpose. Thermal symmetry is defined as the degree of similarity between two ROIs, mirrored across the human body’s longitudinal main axes which are identical in shape, identical in size, and as near identical in position as possible. Based on the fact that the human body is bilateral and practically fully symmetrical with respect to its extremities, the indication is that the human body can have bilateral temperatures symmetry between two ROIs in co-lateral locations. Significant differences in bilateral temperatures symmetry can be related to pathological states such as structural abnormalities of blood vessels, abnormalities of vascular control, local effects on blood vessels, changes in thermal conductivity of the tissues and increased heat production in the tissues. 23 Table 3:Thermal symmetry of some ROIs using Medical Thermography [99, 100]. Region of interest Mean and SD values Face 0.30 ± 0.20 Neck 0.40 ± 0.30 Arm, upper 0.50 ± 0.40 Forearm, dorsal 0.50 ± 0.30 Forearm, volar 0.30 ± 0.20 Trunk, upper 0.30 ± 0.20 Trunk,medium 0.40 ± 0.30 Trunk, lower 0.30 ± 0.20 Back, upper 0.20 ± 0.20 Back, medium 0.30 ± 0.20 Back, lower 0.30 ± 0.30 Leg, anterior 0.30 ± 0.30 Leg, posterior 0.20 ± 0.20 Thigh anterior 0.20 ± 0.20 Thigh posterior 0.20 ± 0.10 The values taken in consideration are the mean temperature of the ROI and its standard deviation. Table 3 shows these values from the regions of the body that were studied [99, 100]. 2.2.3 Non-invasive competing techniques In this subsection some alternatives to medical thermography are presented as well as their advantages and disadvantages and subsequent comparison with infrared thermography. 2.2.3.1 Contact Thermography According Sherman et al. [101], the majority of contact thermography systems consists in series of detectors and flexible pillows, generally about 18 (within 46 cm) on each side, containing crystals vectors that changes colour corresponding to a specific temperature. The crystals that are sensitive to different temperatures are closely spaced to form similar pixels to those that produce colloidal images on a TV screen. In this technique, the pillow is pressed against the body part to be analyzed, and then a colour image is produced with multiple pixels. Temperature differences are required to produce various colours. The differences between each colour may vary from 0.3 to 1.1 °C. This device is theoretically accurate to 0.2 °C [101]. The main advantages of this technique is cost effectiveness, easy to use and calibrate, is suitable for curved surfaces, have a high spatial resolution and it can be reusable [55]. However, the readings are ordinal, but not continuous, because the data are limited to parametric statistical methods and comparisons between devices are limited to nonparametric methods. Contact thermography cannot be used to long term records because the pressure exerted by the plastic screen in the body, changes the heat surface patterns. Also has difficulty reading temperatures on low lighting conditions, has a limited range of temperature and it is a slow technique, because there is the need for a slow photographic record [55, 101]. 30 a) b) Figure 15: Type of neoplasm prevalence in both groups (aFollow-up group and bDiagnosed group). 31 a) b) Figure 16: Skin neoplasm location prevalence in both groups (aFollow-up group and bDiagnosed group). 3.3 Data collection protocol The data collection protocol represents an important role in any biomedical investigation. The protocol adopted for this research follows the guidelines of "Glamorgan Protocol". This protocol recommends the standard knowledge for recording and evaluation of human body with thermal images [55]. 32 The protocol is based on the objective target of the anatomical region of interest, face for the control group and the neoplasm location for the oncological group. To achieve this goal, the anatomical landmarks are aligned in each view to the image edge. It allows also to reproduce body positions views to increase the thermal image reproducibility [55]. The protocol incorporates aspects such as communication, preparation and consideration of the patient, exam guidelines, results evaluation and presentation [108, 109]. The capture protocol recommended for medical thermography defines that the environment where the test is performed should have an ambient temperature between 18ºC and 24ºC, relative humidity below 50%, air flow less than 2m/s and the absence of incident light on the subject. Before data collection, ethical approval was requested and obtained from the ethics committee of the IPO (appendix I). The process of data collection involved several common steps to the three groups under test. Before capture, it was necessary that the individual remained in a heated room, exposing the area where the lesion was located. This process took about 10-15 minutes, in order to obtain the desired temperature. Meanwhile, the individual became aware and filled out some questionnaires. Among these were include informed consent, biometric data questionnaire, and pathology questionnaire (appendix II, III and IV). The biometric data questionnaire include questions about mobility, autonomy, usual activities, discomfort, pain, anxiety and depression. The final score obtained in this questionnaire should be zero, for inclusion the individual in the study. Only the control group filled this questionnaire. The pathology questionnaire includes information on parameters such as sun sensibility, and family history. The sun sensibility was scaled between 0 and 40 points (inversely proportional to the sun sensibility). For the control group, the phase of menstrual cycle (follicular and luteal) was also recorded. It was also essential that the individual knew his weight and height to allow body mass index (BMI) calculation. It is defined as the individual’s body mass divided by the square of their height, given in units of kg/m2. Before the data collection period, some precautions should be taken such as:  Not having a heavy foods prior to data collection;  No participating in heavy activity or consuming alcohol or hot drinks or smoking 2 hours prior to the pictures being taken;  Wearing minimal clothing in order to obtain clear thermograms;  Removing all jewellery prior to the pictures being taken;  Refraining from applying cosmetics and ointments to the skin prior to the pictures being taken. Different capture procedures were adopted for the three groups. For the control group, nine different images of face per person were taken, in three different moments and 33 positions. The positions of the face for each set of three images were frontal, right, and left profiles. Figure 18 shows some thermograms of the control group in three different moments of frontal and right profile. The first three pictures were taken after the acclimatization period and before applying a provocation test (first moment). The provocation test was a thermal stimulus consisted in a stream of cold air from a fan, facing 90º angle, and applied to the face during 2 minutes. It was applied cold air instead of hot, because in humans the colder sensors react faster than the warm sensors, and thus the examination can be done in less time. Immediately after the application of the thermal stimulus, 3 more images (second moment) were captured. The last 3 images were captured after waiting 5 minutes, in order to re-establish the normal temperature (third moment). Thus, it is possible to maximize the physiological response of the face. Figure 17 illustrates the data collection protocol in three different moments. Figure 17: Cold stimulation and recovery collection protocol for three different moments. For the oncologic group, there was a time constraint to perform the examination, only one image was taken by the individual in the neoplasm location, and the thermal stimulus could not be applied. 34 Figure 18: Thermograms of control group in three different moments of frontal and right profile (A1st moment, frontal view; B-2nd moment, frontal view; C-3rd moment, frontal view; D1st moment, right profile view; E2nd moment, right profile view;F-3rd moment, right profile view). The neoplasm location included parts of body such as scalp, right face profile, left face profile, nose, back, chest, upper limbs (hand, arm, shoulder and axilla) and lower limbs (thigh, leg and foot). 3.4 Inclusion and exclusion criteria Table 5 presents the inclusion and exclusion criteria of this study. Table 5: Description of the inclusion and exclusion criteria of this study. Inclusion criteria Exclusion criteria Control Group - Asymptomatic individuals; - Score null in the biometric survey date; - Individuals belonging to University of Porto. - Accented facial hair; - Individuals who have consumed caffeine; - Individuals who are smokers; - Individuals who have makeup or facial cream; - Individuals who have practiced intense physical exercises on the previous day; - Individuals minors (under 18 years). Oncologic Group - Have or had skin neoplasms; - Be a patient on Portuguese Institute of Oncology (IPO). - Accented facial hair; - Individuals who have consumed caffeine; - Individuals who are smokers; - Individuals who have makeup or facial cream; - Individuals who have practiced intense physical exercises on the previous day; - Individuals without neoplasms; -Individuals with suspicious neoplasm diagnosis. 3.5 Analysis protocol Data analysis was performed using the FLIR tools software which allowed the extraction of maximum, minimum and mean temperature values for each region of interest. However, 35 regions of interest were selected in different ways for the two groups. After obtaining the results, they were treated in SPSS Statistical 22 using different statistical methods. All these aspects are described in the following subsection. 3.5.1 Control group analysis The regions of interest (ROIs) of autonomous nervous system (ANS) were selected according to Netter 3-way division [110] both frontal, right and left profile positions. The frontal images were divided into 16 regions (8 on the right side and 8 on the left side) using squares. The profile images were divided into 8 regions also using squares (see figure 19-A). For the same individual, the squares areas were the same for the right and left sides. Then this were combined in ROI 1, ROI 2 and ROI 3 according to Netter classification (see figure 20A). Figure 19: Frontal and profile images with the selected ROIs (A-ANS, B-MS). 36 From the data obtained, it was used the mean value of temperature, both to the left and right side in different positions. Thus, the thermal symmetry of ROI 1, ROI 2 and ROI 3 was calculated by the equation 3. 𝑇ℎ𝑒𝑟𝑚𝑎𝑙 𝑆𝑦𝑚𝑚𝑒𝑡𝑟𝑦 = |𝑅𝑂𝐼𝑙𝑒𝑓𝑡 − 𝑅𝑂𝐼𝑟𝑖𝑔ℎ𝑡| (3) The regions of interest (ROIs) of microvascular system (MS) were selected according to Gratt et al. [111] both front, right and left profile positions. The front images were divided into 18 regions (9 on the right side and 9 on the left side), while the profile images were divided into 10 regions also using squares. Such as the ANS, the squares areas were the same for the right and left sides (figure 19B). The regions were then combined in ROI 1, ROI 2, ROI 3, ROI 4, ROI 5, ROI 5, ROI 7, ROI 8 and ROI 9 from front images, and in ROI 1, ROI 2, ROI 3, ROI 4, ROI 5, ROI 6, ROI 7, ROI 8, ROI 9 and ROI 10 from profile images. So, it was obtained a classification according to Gratt et al. (figure 20-B). Figure 20: Regions of interest of ANS according to Netter (A) and MS according to Grantt (B) [111]. 3.5.2 Diagnosed and Follow-up group analysis The ROIs of oncologic group analysis were selected according the location of neoplasm. For all cases, it was used two squares in the neoplasm region (one square in the left of neoplasm, other in the right of neoplasm) as can be seen in figure 21 [112]. Figure 21: ROI selection on skin neoplasm group, A-Basal Cell carcinoma; B-Squamous Cell carcinoma. 37 The squares had an area of approximately 8x8 pixels. In these groups it was calculated the absolute difference between the left and right side of skin neoplasms (ΔT). 3.5.3 Case-study It was also compared some of the individuals who possessed skin neoplasms on the face with a group of controls that had a BMI within the same class and the same gender. In the control group, roughly the same square area was selected around the neoplasm (see figure 22). Thus, it was possible to compare the mean temperature of the individual with skin neoplasm with the appropriate control group. Figure 22: ROI selection in control individual (A) and basal cell carcinoma individual (B). A total of 4 cases were tested of diagnosed basal cell carcinoma. The minimum number of comparable controls was 14 and the maximum 18. 3.6 Statistical treatment Unlike the parametric tests, the nonparametric tests do not require strong requirements as normality. They are used when dealing with small samples with a distribution that is not normal. They can also be used if, despite the sample having a size greater than 30, when more conservative conclusions are sought. Non-parametric tests were considered using a significance level of 0.05, table 6. The U-Mann Whitney test allows the detection of significant differences between the central values of two situations, when considered different subjects. It is the nonparametric alternative to the student t-test for independent samples, when the conditions for the application of a parametric test are not meet. The Wilcoxon test enables the detection of significant differences between the central values of two situations, when the same subjects are considered. It represent a nonparametric alternative to student’s t-test for paired samples. 38 The Kruskal-Wallis test allows the detection of significant differences between the central values of three or more situations, when different subjects are considered. It represent an nonparametric alternative to ANOVA for independent samples [113]. Table 6: Description of mean distribution equality tests. Parametric tests Non-parametric tests Two paired samples T-student Test* Wilcoxon or Signal Test Two independent samples T-student Test* ** U Mann-Whitney Test More than two independent samples One Way ANOVA * ** Kruskal-Wallis Test The chi-square test for independence verifies if there is a statistically significant evidence of association between two characteristics. In other words, it verifies the independence between two variables of any type that are grouped together in contingency tables. This test should not be used if more than 20% of expected frequencies under the assumption of independence are less than 5, or one of them is zero. If this assumption does not occur, the Fisher’s test will be used. 46 Figure 26: ΔT values of the different type of skin neoplasms in the differences groups. Through the Kruskal-Wallis test, it can be affirmed that the ΔTs values of different skin neoplasm types were similar (pD=0.141 and pF=0.204). 4.1.2.1 Relation between visible skin neoplasms and oncological thermograms In the figure 27 are presented some examples of visual images and thermograms of skin neoplasms. Figure 27: Thermographic and visual images of some skin neoplasms (A-Melanoma in the back, visual image; BMelanoma in the back, thermogram; CMelanoma in the scalp, visual image; DMelanoma in the scalp, thermogram; EBasal Cell carcinoma in face, visual image; FBasal cell carcinoma in face, thermogram; GFollow-up of Squamous cell carcinoma in face, visual image; H-Follow-up of Squamous cell carcinoma in face, thermogram). It can be observed that melanoma is characterized by higher temperatures. On the other hand, in follow-up images there was a uniform distribution of superficial temperature. 4.2 Individuals characteristics and skin temperature This study was focused in the influence of the individual’s characteristics in the skin surface temperature and thermal symmetry for the control group. Relatively to oncologic group, it was focused in the skin surface temperature and ΔTs. Characteristics, such as gender, body mass index, sun sensibility and in the case of females the different phases of menstrual cycle. Age, body mass index and sun sensibility were only addressed in oncological group because the control did not have a wide variability within this scope. 4.2.1 Control Group In this study it was assessed the gender and phase of the menstrual cycle influence between moments for different systems in left and right sides and also in thermal symmetry. The group was composed by 24 females and 16 males, while concerning the menstrual cycle, 14 individuals were in follicular phase and 10 in luteal phase. 47 4.2.1.1 Autonomous Nervous System Firsty, it was studied if the gender influences the ANS. It was evaluated if the female and male responses were equal at all moments for the right side in frontal and profile positions (table 15). Table 15: Mean temperatures to the right side of each gender for three different moments. ROIs Gender 1st Moment 2nd Moment 3rd Moment Frontal R1 F 34.55 ± 0.53 34.55 ± 0.53 34.23 ± 0.77 M 34.55 ± 2.50 34.04 ± 2.50 34.49 ± 0.37 R2 F 34.55 ± 0.72 33.92 ± 0.72 33.71 ± 0.86 M 34.55 ± 0.52 34.19 ± 0.52 34.29 ± 0.49 R3 F 34.55 ± 0.86 33.54 ±0.86 32.77 ± 2.26 Profile M 34.55 ± 0.78 33.80 ± 0.78 33.74 ± 0.62 R1 F 34.55 ± 0.59 32.37 ± 0.71 33.83 ± 0.73 M 34.55 ± 0.49 32.86 ± 0.43 34.04 ± 0.32 R2 F 34.55 ± 0.82 31.72 ± 0.87 32.87 ± 0.89 M 34.55 ± 0.44 32.91 ± 0.42 34.00 ± 0.42 R3 F 34.55 ± 0.84 31.83 ± 0.82 32.98 ± 0.87 M 34.55 ± 0.60 32.87 ± 0.48 33.94 ± 0.60 Using U Mann-Whitey test, it can be stated that exists significant statistical evidence to affirm that the mean temperatures of frontals ROI R1, ROI R2 and ROI R3 were similar between moments I and II. On the other hand, between moments II and III frontals ROI R1 and ROI R2 were different. Regarding the profile position, only ROI R1 was equal between moment I and II for both genders. Between moments II and III and moments I and III, all ROIs were different, with exception to ROI R1 between moment I and III. In table 16 it can be observed the p-value, U, W and Z of U Mann-Whitney tests results. Table 16: Results of p-value, U, W and Z values of U Mann Whitney test between moments for the right side. Moment I to II Moment II to III Moment I to III U W Z p U W Z P U W Z P Frontal R1 192 328 0.00 1.000 108 408 -2.34 0.020 166 466 -0.72 0.471 R2 155 455 -1.02 0.310 110 410 -2.28 0.020 100 400 -2.54 0.011 R3 158 458 -0.94 0.350 127 427 -1.81 0.070 118 418 -2.06 0.040 Profile R1 154 454 -1.07 0.280 112 412 -2.21 0.030 171 471 -0.60 0.551 R2 65 365 -3.52 0.000 34 334 -4.36 0.000 37 337 -4.29 0.000 R3 117 417 0.00 0.040 52 352 -3.88 0.000 69 369 -3.40 0.001 For left side it was also studied the male and female responses in frontal and profile positions (see table 17). 48 Table 17: Mean temperatures to the left side of each gender for three different moments. ROIs Gender 1st Moment 2nd Moment 3rd Moment Frontal R1 F 34.60 ± 0.55 32.54 ± 0.80 34.23 ± 0.73 M 34.62 ± 0.40 32.91 ± 0.35 34.41 ± 0.33 R2 F 33.94 ± 0.77 32.32 ± 0.84 33.74 ± 0.84 M 34.10 ± 0.54 32.76 ± 0.52 34.15 ± 0.40 R3 F 33.46 ± 0.76 31.78 ± 0.87 33.16 ± 0.91 Profile M 33.67 ± 0.81 32.30 ± 0.56 33.61 ± 0.64 R1 F 34.15 ± 0.53 32.38 ± 0.76 33.75 ± 0.72 M 34.13 ± 0.58 32.63 ± 0.46 33.90 ± 0.47 R2 F 33.47 ± 0.87 31.78 ± 0.93 32.87 ± 0.89 M 34.23 ± 0.55 32.88 ± 0.47 34.00 ± 0.51 R3 F 33.47 ± 0.74 32.02 ± 0.78 33.03 ± 0.80 M 34.23 ± 0.68 32.94 ± 0.55 33.96 ± 0.61 Using U Mann-Whitey test it can be referred that there exists significant statistical evidence to affirm that the mean temperatures of all ROIs were similar for all moments in frontal position with an exception of frontal ROI R1 between moments II and III. For profile position all ROIs were different with exception to ROI R1 regarding all moments. In table 18 it can be can observed the results for p-value, U, W and Z. Table 18: Results of p-value, U, W and Z values of U Mann Whitney test between moments for the left side. Moment I to II Moment II to III Moment I to III U W Z p U W Z P U W Z P Frontal R1 190 326 -0.04 0.97 117 417 -2.06 0.04 174 474 -0.50 0.62 R2 178 478 -0.39 0.7 141 441 -1.39 0.16 131 431 -1.68 0.09 R3 160 460 -0.87 0.38 134 434 -1.59 0.11 130 430 -1.70 0.09 Profile R1 192 328 0.00 1.00 146 446 -1.26 0.21 175 475 -0.46 0.65 R2 96 396 -2.64 0.01 0.21 345 -4.05 0.00 46 346 -4.03 0.00 R3 110 410 -2.25 0.02 63 363 -3.56 0.00 64 364 -3.53 0.00 Regarding thermal symmetry, it was also performed a comparison of genders response in frontal and profile positions between moments (table 19). 49 Table 19: Mean temperatures to the thermal symmetry of each gender for three different moments. ROIs Gender 1st Moment 2nd Moment 3rd Moment Frontal R1 F 0.18 ± 0.15 0.26 ± 0.18 1.63 ± 6.94 M 0.14 ± 0.12 0.23 ± 0.22 0.18 ± 0.16 R2 F 0.25 ± 0.27 0.22 ± 0.15 0.23 ± 0.17 M 0.22 ± 0.17 0.21 ± 0.12 0.21 ± 0.17 R3 F 0.25 ± 0.27 0.35 ± 0.28 0.31 ± 0.21 Profile M 0.22 ± 0.17 0.21 ± 0.17 0.21 ± 0.18 R1 F 0.28 ± 0.26 0.29 ± 0.22 0.29 ± 0.21 M 0.24 ± 0.20 0.37 ± 0.40 0.27 ± 0.27 R2 F 0.26 ± 0.16 0.28 ± 0.23 0.21 ± 0.19 M 0.20 ± 0.16 0.22 ± 0.21 0.26 ± 0.19 R3 F 0.22 ± 0.16 0.26 ± 0.22 0.22 ± 0.14 M 0.11 ± 0.12 0.18 ± 0.17 0.20 ± 0.18 Using U Mann-Whitey test, there is a significant statistical evidence the thermal symmetry is similar for all ROIs between moments for both genders (see table 20), with exception to profile R3 between moment II and III. Table 20: Results of p-value, U, W and Z values of U Mann Whitney test between moments for thermal symmetry. Moment I to II Moment II to III Moment I to III U W Z p U W Z P U W Z P Frontal R1 163 299 -0.84 0.4 168 304 -0.66 0.51 147 283 -1.26 0.21 R2 191 491 -0.01 0.99 186 486 -0.15 0.88 187 323 -0.12 0.90 R3 165 301 -0.75 0.46 149 285 -1.19 0.23 135 271 -1.56 0.12 Profile R1 178 314 -0.38 0.70 182 482 -0.27 0.79 174 310 -0.50 0.62 R2 150 286 -1.16 0.24 160 296 -0.88 0.38 157 457 -0.97 0.33 R3 103 239 -2.46 0.01 145 281 -1.29 0.20 162 298 298.5 0.41 Later, it was studied the influence of menstrual cycle phase in ANS. To verify if the menstrual cycle phases had some impact on the mean temperature on skin surface. It was evaluated if the phase’s response were identical at all moments to the right side in frontal and profile positions (table 21). 50 Table 21: Mean temperatures to right side of each phase for three different moments. ROIs Phase 1st Moment 2nd Moment 3rd Moment Frontal R1 1 34,58 ± 0,51 32.72 ± 0.87 34.19 ± 0.86 2 34.52 ± 0.59 32.27 ± 0.72 34.27 ± 0.65 R2 1 34.00 ± 0.62 32.45 ± 0.74 33.73 ± 0.93 2 33.80 ± 0.86 31.99 ± 0.92 33.68 ± 0.79 R3 1 33.67 ± 0.79 31.78 ± 0.89 32.54 ± 2.88 Profile 2 33.37 ± 0.95 31.73 ± 1.74 33.09 ± 0.96 R1 1 34.11 ± 0.61 32.54 ± 0.76 33.79 ± 0.82 2 34.00 ± 0.60 32.13 ± 0.59 33.88 ± 0.63 R2 1 33.5 ± 0.82 31.88 ± 0.78 32.84 ± 0.93 2 33.33 ± 0.86 31.50 ± 0.98 32.91 ± 0.87 R3 1 33.71 ± 0.91 31.98 ± 0.7 32.90 ± 0.92 2 33.55 ± 0.77 31.61 ± 0.96 33.04 ± 0.84 Using U Mann-Whitey test, it can be affirm that in both frontal and profile positions the mean temperatures of right side were similar for all moments in different phases. In appendix V it can be found the results for p, U, W and Z values. After, it was evaluated if the response of the first and second phases were equal at all moments to the left side in frontal and profile positions (see appendix VI). Using U Mann-Whitney test it can be affirm that the mean temperatures are equal for all moments for both phases. The results can be observed in appendix VI. Table 22: Mean temperatures to thermal symmetry of each phase for three different moments. ROIs Phase 1st Moment 2nd Moment 3rd Moment Frontal R1 1 0.17 ± 0.17 0.28 ± 0.14 0.22 ± 0.21 2 0.18 ± 0.12 0.23 ± 0.23 0.20 ± 0.12 R2 1 0.25 ± 0.32 0.22 ± 0.14 0.25 ± 0.18 2 0.26 ± 0.17 0.21 ± 0.17 0.19 ± 0.15 R3 1 0.29 ± 0.27 0.33 ± 0.27 0.30 ± 0.23 Profile 2 0.26 ± 0.19 0.37 ± 0.31 0.33 ± 0.20 R1 1 0.34 ± 0.31 0.26 ± 0.20 0.24 ± 0.23 2 0,20. ± 0.13 0.32 ± 0.25 0.36 ± 0.18 R2 1 0.19 ± 0.15 0.27 ± 0.18 0.18 ± 0.17 2 0.34 ± 0.12 0.30 ± 0.29 0.26 ± 0.21 R3 1 0.24 ± 0.12 0.20 ± 0.16 0.22 ± 0.13 2 0.19 ± 0.16 0.35 ± 0.27 0.21 ± 0.15 Finally, it was evaluated the influence of menstrual phases in thermal symmetry. The results can be observed in table 22 for all moments. 51 Table 23: Results of p-value, U, W and Z values of U Mann Whitney test between moments for thermal symmetry. Moment I to II Moment II to III Moment I to III U W Z P U W Z P U W Z p Frontal R1 64 169 -0.36 0.71 50 105 -1.16 0.24 59 114 -0.66 0.50 R2 55 160 -0.85 0.39 65 120 -0.26 0.79 56 111 -0.82 0.41 R3 69 124 -0.05 0.95 67 172 -0.14 0.88 65 170 -0.26 0.79 Profile R1 48 103 -1.28 0.19 60 165 -0.56 0.57 44 149 -1.51 0.13 R2 29 134 -2.38 0.01 68 123 -0.11 0.90 57 162 -0.73 0.46 R3 59 114 -0.64 0.51 49 154 -1.20 0.22 62 117 -0.44 0.65 Using U Mann-Whitey test, it can be stated that there exists significant statistical evidence to affirm that thermal symmetries were equal for all ROIs in all moments with exception to profile ROI R2 between moment I and II (see table 23). 4.2.1.2 Microvascular System In this system it was also studied the influence of gender and phase of menstrual cycle. Firstly, it was evaluated if the response of the female and male genders were equal at all moments for the right side in frontal and profile positions (see appendix VII). By U MannWhitey test it can be stated that there exist significant statistical evidence to affirm that between moment I and II, all temperatures of all ROIs were similar, with exception to frontal ROI R9 and profile ROIs R6, R7, R8 and R10. Between moments II and III, all temperatures of all ROIs were equal, with exception to frontal R1, R4, R6, R7 and R9 ROIs and profile R1, R3, R6, R7, R8, R9 and R10 ROIs (see appendix VII). Finally, between moments I and III, all temperatures were similar, with exception of frontal R1, R6 and R9 ROIs and profile R1, R2, R6, R7, R8, R9 and R10 ROIs. Regarding the left side, it was performed the same study (appendix VIII). Using U MannWhitey test it can be stated that there exist significant statistical evidence to affirm that all temperatures were similar between moments I and II, with exception to ROIs frontal R5 and R9, and profile R6, R7, R8 and R10. Between moments II and III, all temperatures were identical with exception to frontal R1, R6, R7 and R9 and profile R1, R6, R7, R8, R9 and R10. In sum, between moments I and III all temperatures were equal with exception of profile R1, R3, R6, R7, R8, R9 and R10 and frontal R6, R7 and R8 (appendix VIII). Relatively to thermal symmetry, the same study was performed, also in frontal and profile positions (appendix IX). Using the U Mann-Whitey test it can be referred that there is significant statistical evidence to affirm that the thermal symmetries were similar for all ROIs in all moments, with exception to profile ROI R3 between moments I and II, frontal R3 and R7 and profile R1, R7 and R8 ROIs between moments II and III. Between moments I and III, only frontal R3 and profile R1 and 7 ROIs were different (see appendix IX). Secondly, it was studied the influence of menstrual cycle phases in MS in profile and frontal positions (appendix X). Using U Mann-Whitey test it can be stated that there is significant statistical evidence to affirm that the temperatures of right side were equal for all 52 ROIs in all moments, both profile and frontal positions with exception of frontal ROI R8 between moments II and III. The results can be observed in appendix X. Using U Mann-Whitey test, it can be referred that there is significant statistical evidence to affirm that the temperatures of left side were identical for all ROIs in all moments in both profile and frontal positions (appendix XI). The thermal symmetry response of different phases in each moment for frontal and profile positions was assess using U Mann-Whitney test (appendix XII). Through this test it can be stated that there is significant statistical evidence to affirm that thermal symmetry are similar for all ROIs in all moments, with exception to profile ROI R3 between moments I and II and frontal ROI R9 between moments II and III and between moments I and III. The results can be observed in appendix XII. 4.2.2 Oncologic Group For oncologic group, it was studied the influence of individuals characteristics such as gender, age, BMI, sun sensibility and malignancy in ΔT and mean temperature. In relation to diagnosed group, all mentioned characteristics were studied. Concerning the gender, it were recorded 26 females and 32 males. For malignancy, it were recorded 14 benign skin neoplasms and 44 malignant neoplasms of the diagnosed group. The distribution of individuals in each BMI class for diagnosed group was 1 individual bellow age of 18.5, 20 individuals in class ]18.5-25], 28 individuals in class ]25-30] and 9 individuals in class ]30-35]. The distribution in each subject age was 3 individuals in class [18-30], 2 individuals in class ]3040], 2 individuals in class ]40-50], 4 individuals in class ]50-60], 18 individuals in class ]60-70] and 29 individuals with age higher than 70. For sun sensibility, were recorded 6 individuals in class ]10-20], 22 individuals in class ]20-30] and 30 individuals in class ]30-40]. The obtained results for each characteristic were in table 24. Regarding to follow-up group, it were recorded 24 females and 21 males. For malignancy, it were recorded 3 benign skin neoplasms and 42 malignant neoplasms of the diagnosed group. The distribution of individuals in each BMI class for the diagnosed group was 16 individuals in class ]18.5-25], 19 individuals in class ]25-30], 8 individuals in class ]30-35] and 2 individuals with BMI higher than 35. The distribution in each subject age was 3 individuals in class [18-30], 1 individuals in class ]30-40], 3 individuals in class ]40-50], 7 individuals in class ]50-60], 10 individuals in class ]60-70] and 21 individuals with age higher than 70. 53 Table 24: Obtained results for each characteristic to median temperatures of aright and lefts sides and ΔTs for diagnosed group. Classes Right Left ΔT Subject Age [18,30] 33.00 ± 1.48 33.00 ± 1.34 0.00 ]30,40] 33.00 ± 0.00 32.90 ± 0.00 0.10 ± 0.00 ]40,50] 33.70 ± 2.47 33.60 ± 2.28 0.10 ± 0.01 ]50,60] 34.40 ± 5.00 34.30 ± 4.79 0.10 ± 0.02 ]60,70] 34.15 ± 4.48 34.25 ± 2.10 0.10 ± 0.79 >70 33.80 ± 2.64 34.00 ± 2.73 0.20 ± 0.05 Subject Gender Female 33.55 ± 4.41 33.90 ± 3.69 0.20 ±0.33 Male 34.10 ± 1.28 34.10 ± 1.35 0.10 ±0.04 BMI ]18,5;25] 34.45 ± 3.74 34.25 ± 2.24 0.10 ± 0.48 ]25,30] 33.80 ± 1.62 34.00 ± 1.54 0.10 ± 0.07 ]30,35] 32.90 ± 4.56 32.75 ± 4.59 0.15 ± 0.01 <35 34.05 ± 0.61 34.05 ± 0.13 0.30 Sun Sensibility ]10,20] 33.00 ± 0.00 32.90 ± 0.00 0.10 ± 0.00 ]20,30] 34.40 ± 5.72 34.20 ± 4.76 0.10 ± 0.43 ]30,40] 33.75 ± 1.50 33.95 ± 1.43 0.10 ± 0.05 Malignancy Benign 33.00 ± 0.01 33.00 ± 0.02 0.00 Malign 34.15 ± 3.39 34.05 ± 2.90 0.10 ± 0.21 For sun sensibility, it were recorded 1 individual in class ]10-20], 18 individuals in class ]20-30] and 26 individuals in class ]30-40]. The obtained results for each characteristic are presented in table 25. Table 25: Results follow-up group characteristic to right and lefts sides and ΔT. Characteristics Classes Right Left ΔTs Subject Age [18,30] 33.00 ± 1.48 33.00 ± 1.34 0.00 ]40,50] 33.70 ± 2.47 33.60 ± 2.28 0.10 ± 0.01 ]50,60] 34.40 ± 5.00 34.30 ± 4.79 0.10 ± 0.02 ]60,70] 34.15 ± 4.48 34.25 ± 2.10 0.10 ± 0.79 >70 33.80 ± 2.64 34.00 ± 2.73 0.20 ± 0.05 Subject Gender Female 33.55 ± 4.41 33.90 ± 3.69 0.20 ±0.33 Male 34.10 ± 1.28 34.10 ± 1.35 0.10 ±0.04 BMI ]18,5;25] 34.45 ± 3.74 34.25 ± 2.24 0.10 ± 0.48 ]25,30] 33.80 ± 1.62 34.00 ± 1.54 0.10 ± 0.07 ]30,35] 32.90 ± 4.56 32.75 ± 4.59 0.15 ± 0.01 >35 34.05 ± 0.61 34.05 ± 0.13 0.30 Sun Sensibility ]10,20] 33.00 ± 0.00 32.90 ± 0.00 0.10 ± 0.00 ]20,30] 34.40 ± 5.72 34.20 ± 4.76 0.10 ± 0.43 ]30,40] 33.75 ± 1.50 33.95 ± 1.43 0.10 ± 0.05 Malignancy Benign 33.00 ± 0.01 33.00 ± 0.02 0.00 Malign 34.15 ±3.39 34.05 ± 2.90 0.10 ± 0.21 It was studied if ΔT was related to these characteristics for both groups. For diagnosed group, with the aid of U Mann-Whitney test it can be stated that there is statistical significant evidence to affirm that the ΔTs referring to malignancy were different (table 26). 54 Table 26: U Mann Whitney test results in terms of p-value and ranking values for diagnosed group. Characteristics U W Z P Gender 171.00 402.00 1.94 0.05 Malignancy 16.00 22.00 2.26 0.02 Within diagnosed group, using Kruskal-Wallis test there is statistical significance evidence to affirm that the ΔTs were different for the BMI classes (table 27). Table 27: Kruskal-wallis test results in terms of p-value for diagnosed group. Characteristics p Age 0.33 BMI 0.00 Sun Sensibility 0.53 For the follow-up group, using U Mann-Whitney test there is statistical significance evidence to affirm that the ΔTs referring to gender and malignancy were identical (table 28). Table 28: U Mann Whitney test results. Characteristics U W Z P Gender 343 694 1.15 0.25 Malignancy 209 314 1.82 0.07 Using Kruskal-Wallis test there is statistical significance evidence to affirm that the ΔTs were different in BMI classes (table 29). Table 29: Kruskal-wallis test results in terms of p-value. Characteristics P Age 0.51 BMI 0.04 Sun Sensibility 0.64 4.3 Individuals characteristics and malignancy The fourth part of this study is focused on establishing relations between individual’s characteristics and malignancy. Characteristics such age, gender, body mass index, sun sensibility and family history. It was verified that 83.3% of individuals are within 18 to 30 years old and were benign neoplasms. The prevalence of malign neoplasms was the age group of 40 to 70 years old is also noteworthy (see table 30). Using Pearson chi-square test can be stated that there is statistical significant evidence that the age and malignancy are associated. 55 Table 30: Prevalence of malignancy by age classes. [18,30] ]30,40] ]40,50] ]50,60] ]60,70] >70 p N % N % N % N % N % N % Benign 5 83.3% 2 66.7% 0 0.0% 1 9.1% 4 14.3% 5 100% 0.00 Malign 1 16.7% 1 33.3% 5 100.0% 10 90.9% 24 85.7% 45 90.0% A percentage of 14% of female individuals had benign neoplasms (table 31). With Pearson chi-square test, it can be stated that there is no statistical significant evidence that the gender is related with malignancy. Table 31: Prevalence of malignancy in gender. Female Male P N % N % Benign 7 14.0% 10 18.9% 0.506 Malign 43 86.0% 43 81.1% A percentage of 5.9% of individuals who had BMI less than 18.5 kg/m2 had benign neoplasms. Whereas 44.2% of individuals with BMI between 25 and 30 kg/m2 had malignant neoplasms (table 32). Using Pearson chi-square test there is statistical significant evidence to affirm that the BMI classes and malignancy are not associated. Table 32: Prevalence of malignancy by BMI classes. <18,5 ]18,5;25] ]25,30] ]30,35] >35 P N % N % N % N % N % Benign 1 5.9% 6 35.3% 9 52.9% 1 5.9% 0 0.0% 0.131 Malign 0 0.0% 30 34.9% 38 44.2% 16 18.6% 2 2.3% A percentage of 26.8% of individuals who had a sun sensibility between 30 and 40 points had benign neoplasms. However, 97.5% of individuals, who had a sun sensibility between 20 and 30 points had malignant neoplasms (table 33). Using Pearson chi-square test, it can be stated that there is statistical significant evidence that the sun sensibility is related with malignancy. Table 33: Prevalence of malignancy by sun sensibility classes. ]10,20] ]20,30] ]30,40] P N % N % N % Benign 1 14.3% 1 2.5% 15 26.8% 0.007 Malign 6 85.7% 39 97.5% 41 73.2% A percentage of 50% of individuals who had family history with paternal relatives had benign neoplasms. Similarly, 50% of individuals who had family history with paternal relatives had malignant neoplasms. It should also be noted that 20% of individuals who had family history with a sibling had benign neoplasms, while 80% of individuals who had family history with a sibling had malign neoplasms (figure 28). Thus, using Pearson chi-square test, it can be 62 5.4 Case studies The results obtained for the case studies allowed to conclude that ΔT of the skin neoplasm individual and associated control group were similar for diagnosed group. However, this result is contrary to the literature [95]. Many studies shows that in the case of basal cell carcinoma had higher metabolism, increasing blood flow and presented slightly higher temperature when compared to surrounding healthy skin. Therefore, it should have been obtained higher temperatures in the basal cell carcinoma participants than in the control group. However, to the results being more accurate and precise, a larger dimension sample of basal cell carcinoma participants is needed. Chapter 6. Conclusion The aim of this research study was the characterization of the temperature at the skin surface in healthy control and skin oncologic groups, which was fully accomplished. It was also assessed the effect of age, BMI, gender, sun sensibility, malignancy, menstrual cycle phase in females, thermal symmetry and ΔTs in the studied population. Reference data was provided for facial temperature distribution in autonomous nervous system and microvascular system. The characteristic factors that had a significant relationship were the BMI for the diagnosed and follow-up groups, and the malignancy factor in the diagnosed group. It can be concluded that medical thermal imaging proved its importance in the support for the identification and assessment of the skin carcinoma treatments. However, it requires significant research improvements and more tests with large samples in order to be actually used as a complementary method of diagnosis in the field of dermatology. Nevertheless, there were many limitations in this study that could affect the global results:  It was not impossible to perform the thermo-stimulus in the oncologic group, due to the reduced appointment time for each individual (less than 10 minutes) and the inexistence of a free examination room;  The impossibility of measure the height and weight of the individuals involved in this study, and thus not accurate. 6.1 Proposed future work In order to take this work into a further step, the following future work is proposed:  Build a larger dataset to yield more diverse in order to have more statistical evidence results;  Develop an efficient cold stimulus for better understanding of the thermophysiology of the skin surface and associated characteristics; 65  Development of a dedicated software for standardizing the image analysis in order to improve the reproducibility and accuracy of the method. 66 References [1] M. Pirtini Çetingül e C. Herman, “Quantification of the thermal signature of a melanoma lesion”, Int. J. Therm. Sci., vol. 50, n. 4, pp. 421–431, Abr. 2011. [2] J. Ferlay, H.-R. Shin, F. Bray, D. Forman, C. Mathers, e D. M. Parkin, “Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008”, Int. J. Cancer, vol. 127, n. 12, pp. 2893–2917, Dez. 2010. [3] P. S. Pinheiro, J. E. Tyczyński, F. Bray, J. Amado, E. Matos, e D. M. Parkin, “Cancer incidence and mortality in Portugal”, Eur. J. Cancer, vol. 39, n. 17, pp. 2507–2520, Nov. 2003. [4] E. de Vries e J. Willem Coebergh, “Cutaneous malignant melanoma in Europe”, Eur. J. Cancer, vol. 40, n. 16, pp. 2355–2366, Nov. 2004. [5] “American Cancer Society”. [On-line]. Disponível em: http://www.cancer.org/. [Acedido: 18-Dez-2013]. [6] E. Losina, R. P. Walensky, A. Geller, F. C. Beddingfield, L. L. Wolf, B. A. Gilchrest, e K. A. Freedberg, “Visual screening for malignant melanoma: a cost-effectiveness analysis”, Arch. Dermatol., vol. 143, n. 1, pp. 21–28, Jan. 2007. [7] H. Tsao, G. S. Rogers, e A. J. Sober, “An estimate of the annual direct cost of treating cutaneous melanoma”, J. Am. Acad. Dermatol., vol. 38, n. 5 Pt 1, pp. 669–680, Mai. 1998. [8] B. Bonevski, R. Sanson-Fisher, P. Hersey, C. Paul, e G. Foot, “Assessing the Perceived Needs of Patients Attending an Outpatient Melanoma Clinic”, J. Psychosoc. Oncol., vol. 17, n. 3–4, pp. 101–118, 2000. [9] R. Vardasca e R. Simoes, “Current Issues in Medical Thermography”, em Topics in Medical Image Processing and Computational Vision, J. M. R. S. Tavares e R. M. N. Jorge, Eds. Springer Netherlands, 2013, pp. 223–237. [10] M. Norris e D. R. Siegfried, Anatomy and Physiology For Dummies. John Wiley & Sons, 2011. [11] S. W. Turner, Atlas of Human Anatomy and Physiology: Handbook. 1857. [12] Fukuyama, “BIochemistry and physiology of the skin”, Arch. Dermatol., vol. 120, n. 5, pp. 685–685, Mai. 1984. [13] T. Igarashi, K. Nishino, e S. K. Nayar, The Appearance of Human Skin: A Survey. Now Publishers Inc, 2007. [14] “Different layers of the skin”. [On-line]. Disponível em: http://commons.wikimedia.org/wiki/File:Anatomy_The_Skin__NCI_Visuals_Online.jpg. [Acedido: 15-Mai-2014]. [15] A. A. Romanovsky, “Thermoregulation: some concepts have changed. Functional architecture of the thermoregulatory system”, Am. J. Physiol. Regul. Integr. Comp. Physiol., vol. 292, n. 1, pp. R37–46, Jan. 2007. [16] J. Bell e S. C. Bell, The Anatomy and Physiology of the Human Body: Containing the Anatomy of the Bones, Muscles, and Joints; and the Heart and Arteries. Collins & Company, 1822. [17] “National Cancer Institute”. [On-line]. Disponível em: http://www.cancer.gov/. [Acedido: 15-Dez-2013]. 67 [18] B. Wingerd, The Human Body: Concepts of Anatomy and Physiology. Lippincott Williams & Wilkins, 2013. [19] S. Connor, The Book of Skin. Cornell University Press, 2004. [20] P. J. Cannon, “Medical physiology”, Ir. J. Med. Sci., vol. 36, n. 10, pp. 473–474, Out. 1961. [21] E. Satinoff, Thermoregulation. Dowden, Hutchinson & Ross, 1980. [22] A. B. Butler, Comparative Vertebrate Neuroanatomy: Evolution and Adaptation, Auflage: 2. Auflage. Hoboken, N.J: John Wiley & Sons, 2005. [23] K. L. Moore, A. F. Dalley, e A. M. R. Agur, Clinically Oriented Anatomy. Wolters Kluwer Health/Lippincott Williams & Wilkins, 2010. [24] F. O. Stephens e K. Aigner, Basics of Oncology. Springer, 2009. [25] L. J. Kleinsmith, Principles of cancer biology. Pearson Benjamin Cummings, 2006. [26] R. W. Ruddon, Cancer Biology. Oxford University Press, 2007. [27] P. C. Nasca e H. Pastides, Fundamentals of Cancer Epidemiology. Jones & Bartlett Learning, 2001. [28] C. S. M. Wong, R. C. Strange, e J. T. Lear, “Basal cell carcinoma”, BMJ, vol. 327, n. 7418, pp. 794–798, Abr. 2003. [29] V. Madan, J. T. Lear, e R.-M. Szeimies, “Non-melanoma skin cancer”, The Lancet, vol. 375, n. 9715, pp. 673–685, 20. [30] R. E. Kwa, K. Campana, e R. L. Moy, “Biology of cutaneous squamous cell carcinoma”, J. Am. Acad. Dermatol., vol. 26, n. 1, pp. 1–26, Jan. 1992. [31] L. M. Ferrucci, B. Cartmel, A. M. Molinaro, D. J. Leffell, A. E. Bale, e S. T. Mayne, “Indoor tanning and risk of early-onset basal cell carcinoma”, J. Am. Acad. Dermatol., vol. 67, n. 4, pp. 552–562, Out. 2012. [32] F. Aubry e B. Macgibbon, “Risk factors of squamous cell carcinoma of the skin. A casecontrol study in the montreal region”, Cancer, vol. 55, n. 4, pp. 907–911, 1985. [33] M. Anbar, “Quantitative and dynamic telethermometry-a fresh look at clinical thermology”, IEEE Eng. Med. Biol. Mag., vol. 14, n. 1, pp. 15–16, 1995. [34] N. M. Wikonkal e D. E. Brash, “Ultraviolet radiation induced signature mutations in photocarcinogenesis”, J. Investig. Dermatol. Symp. Proc. Soc. Investig. Dermatol. Inc Eur. Soc. Dermatol. Res., vol. 4, n. 1, pp. 6–10, Set. 1999. [35] B. K. Armstrong e A. Kricker, “The epidemiology of UV induced skin cancer”, J. Photochem. Photobiol. B, vol. 63, n. 1–3, pp. 8–18, Out. 2001. [36] “WHO | Sunbeds, tanning and UV exposure”. [On-line]. Disponível em: http://www.who.int/mediacentre/factsheets/fs287/en/. [Acedido: 18-Dez-2013]. [37] R. P. Gallagher, J. J. Spinelli, e T. K. Lee, “Tanning Beds, Sunlamps, and Risk of Cutaneous Malignant Melanoma”, Cancer Epidemiol. Biomarkers Prev., vol. 14, n. 3, pp. 562–566, Jan. 2005. [38] C. D’Arcy, J. Holman, e B. K. Armstrong, “Pigmentary Traits, Ethnic Origin, Benign Nevi, and Family History as Risk Factors for Cutaneous Malignant Melanoma”, J. Natl. Cancer Inst., vol. 72, n. 2, pp. 257–266, Jan. 1984. [39] A. Green, V. Siskind, C. Bain, e J. Alexander, “Sunburn and malignant melanoma.”, Br. J. Cancer, vol. 51, n. 3, pp. 393–397, Mar. 1985. [40] J. A. Centeno, F. G. Mullick, L. Martinez, N. P. Page, H. Gibb, D. Longfellow, C. Thompson, e E. R. Ladich, “Pathology related to chronic arsenic exposure.”, Environ. Health Perspect., vol. 110, n. Suppl 5, pp. 883–886, Out. 2002. [41] R. L. Shannon e D. S. Strayer, “Arsenic-induced Skin Toxicity”, Hum. Exp. Toxicol., vol. 8, n. 2, pp. 99–104, Jan. 1989. [42] H.-S. Yu, W.-T. Liao, e C.-Y. Chai, “Arsenic Carcinogenesis in the Skin”, J. Biomed. Sci., vol. 13, n. 5, pp. 657–666, Set. 2006. [43] J. P. Callen, D. R. Bickers, e R. L. Moy, “Actinic keratoses”, J. Am. Acad. Dermatol., vol. 36, n. 4, pp. 650–653, Abr. 1997. [44] M. Lebwohl, “Actinic keratosis: epidemiology and progression to squamous cell carcinoma”, Br. J. Dermatol., vol. 149, pp. 31–33, 2003. [45] H. Pfister, “Chapter 8: Human Papillomavirus and Skin Cancer”, JNCI Monogr., vol. 2003, n. 31, pp. 52–56, Jan. 2003. 68 [46] M. H. Greene, W. H. Clark Jr, M. A. Tucker, K. H. Kraemer, D. E. Elder, e M. C. Fraser, “High risk of malignant melanoma in melanoma-prone families with dysplastic nevi”, Ann. Intern. Med., vol. 102, n. 4, pp. 458–465, Abr. 1985. [47] W. P. Carey, C. J. Thompson, M. Synnestvedt, D. Guerry, A. Halpern, D. Schultz, e D. E. Elder, “Dysplastic nevi as a melanoma risk factor in patients with familial melanoma”, Cancer, vol. 74, n. 12, pp. 3118–3125, 1994. [48] Harrison’s principles of internal medicine. New York: McGraw-Hill medical, 2012. [49] R. J. Friedman, D. S. Rigel, e A. W. Kopf, “Early detection of malignant melanoma: The role of physician examination and self-examination of the skin”, CA. Cancer J. Clin., vol. 35, n. 3, pp. 130–151, 1985. [50] “ABCD analysis”. [On-line]. Disponível em: http://egeszsegedrearpikam.blogspot.pt/2012_06_01_archive.html. [Acedido: 15-Mai2014]. [51] F. P. Incropera, Fundamentals of heat and mass transfer. John Wiley, 2007. [52] G. J. Tortora e B. H. Derrickson, Principles of Anatomy and Physiology. John Wiley & Sons, 2008. [53] “Heat Loss”. [On-line]. Disponível em: http://general.utpb.edu/FAC/eldridge_j/KINE6362/unit4_l1.html. [Acedido: 31-Mar2014]. [54] N. P. Avdelidis e A. Moropoulou, “Applications of infrared thermography for the investigation of historic structures”, J. Cult. Herit., vol. 5, n. 1, pp. 119–127, Jan. 2004. [55] R. Â. R. Vardasca, “The effect of work related mechanical stress on the peripheral temperature of the hand”, Thesis, Faculty of Advanced Technology, 2010. [56] C. Herman, “Emerging technologies for the detection of melanoma: achieving better outcomes”, Clin. Cosmet. Investig. Dermatol., vol. 5, pp. 195–212, Nov. 2012. [57] R. A. Thomas, The Thermography Monitoring Handbook. Coxmoor Publishing Company, 1999. [58] M. Massoud, Engineering Thermofluids: Thermodynamics, Fluid Mechanics, and Heat Transfer. Springer, 2005. [59] H. Kangro, Early History of Planck’s Radiation Law. Taylor & Francis, 1976. [60] “Plank’s Radiantion Law”. [On-line]. Disponível em: http://forums.tibiabr.com/showthread.php?396464-Curiosidades-F%EDsicas-IRedescobrindo-o-Fogo#.Uyr8lIVOBco. [Acedido: 20-Mar-2014]. [61] “Wien’s Displacement Law”. [On-line]. Disponível em: http://www.skinnerscience.com/Year%2012,13/astrophysics_definitions.htm. [Acedido: 20-Mar-2014]. [62] “FLIR® Thermal Imaging, Night Vision and Infrared Camera Systems”. [On-line]. Disponível em: http://www.flir.com/PT/. [Acedido: 15-Nov-2013]. [63] E. F. J. Ring, “The historical development of temperature measurement in medicine”, Infrared Phys. Technol., vol. 49, n. 3, pp. 297–301, Jan. 2007. [64] D. P. L. Pires, J. C. Afonso, e F. A. B. Chaves, “The thermometry in the XIXth and the XXth centuries”, Rev. Bras. Ensino Física, vol. 28, n. 1, pp. 101–114, Jan. 2006. [65] Mackowiak PA, Wasserman SS, e Levine MM, “A critical appraisal of 98.6°f, the upper limit of the normal body temperature, and other legacies of carl reinhold august wunderlich”, JAMA, vol. 268, n. 12, pp. 1578–1580, Set. 1992. [66] J. M. S. Pearce, “A brief history of the clinical thermometer”, QJM, vol. 95, n. 4, pp. 251–252, Jan. 2002. [67] J.Ring, “History of Thermology and Thermography: Pioneers and Progress”, Thermol. Int., 2012. [68] J. D. Hardy, “The radiation of heat from the human body”, J. Clin. Invest., vol. 13, n. 4, pp. 593–604, Jul. 1934. [69] R. Lawson, “Implications of Surface Temperatures in the Diagnosis of Breast Cancer”, Can. Med. Assoc. J., vol. 75, n. 4, pp. 309–310, Ago. 1956. [70] “Thermography Research”. [On-line]. Disponível em: http://www.thermology.com/research.html. [Acedido: 26-Dez-2013]. [71] J. F. Head e R. L. Elliott, “Infrared imaging: making progress in fulfilling its medical promise”, IEEE Eng. Med. Biol. Mag., vol. 21, n. 6, pp. 80–85, 2002. 69 [72] J. F. Head, F. Wang, C. A. Lipari, e R. L. Elliott, “The important role of infrared imaging in breast cancer”, IEEE Eng. Med. Biol. Mag., vol. 19, n. 3, pp. 52–57, 2000. [73] B. F. Jones e P. Plassmann, “Digital infrared thermal imaging of human skin”, IEEE Eng. Med. Biol. Mag., vol. 21, n. 6, pp. 41–48, 2002. [74] U. I. Bajwa, R. Vardasca, F. Ring, e P. Plassmann, “Comparison of boundary detection techniques to improve image analysis in medical thermography”, Imaging Sci. J., vol. 58, n. 1, pp. 12–19, Fev. 2010. [75] L. J. Jiang, E. Y. K. Ng, A. C. B. Yeo, S. Wu, F. Pan, W. Y. Yau, J. H. Chen, e Y. Yang, “A perspective on medical infrared imaging”, J. Med. Eng. Technol., vol. 29, n. 6, pp. 257–267, Dez. 2005. [76] R. D. Brasfield, J. S. Laughlin, e R. S. Sherman, “Thermography in the Management of Cancer: A Preliminary Report”, Ann. N. Y. Acad. Sci., vol. 121, n. 1, pp. 235–247, 1964. [77] B. Tapernoux e C. Hessler, “Thermography of malignant melanomas”, J. Dermatol. Surg. Oncol., vol. 3, n. 3, pp. 299–302, Jun. 1977. [78] E. Diem e G. Wolf, “[Contact thermographic studies on primary cutaneous melanomas]”, Hautarzt Z. Für Dermatol. Venerol. Verwandte Geb., vol. 28, n. 9, pp. 475–477, Set. 1977. [79] M. Cristofolini, B. Perani, F. Piscioli, G. Recchia, e G. Zumiani, “Uselessness of thermography for diagnosis and follow-up of cutaneous malignant melanoma”, Tumori, vol. 67, n. 2, pp. 141–143, Abr. 1981. [80] M. Hartmann, J. Kunze, e S. Friedel, “Telethermography in the diagnostics and management of malignant melanomas”, J. Dermatol. Surg. Oncol., vol. 7, n. 3, pp. 213–218, Mar. 1981. [81] U. Michel, O. P. Hornstein, e A. Schönberger, “[Infrared thermography in malignant melanoma. Diagnostic potential and limits]”, Hautarzt Z. Für Dermatol. Venerol. Verwandte Geb., vol. 36, n. 2, pp. 83–89, Fev. 1985. [82] O. P. Novak e B. T. Bilyns’kyĭ, “[Thermography in the complex examination of patients with skin melanoma]”, Likarska Sprava Minist. Okhorony Zdorovia Ukraïny, n. 11–12, pp. 66–69, Dez. 1992. [83] A. Di Carlo, “Thermography and the possibilities for its applications in clinical and experimental dermatology”, Clin. Dermatol., vol. 13, n. 4, pp. 329–336, Ago. 1995. [84] R. Amalric, C. Altschuler, D. Giraud, L. Thomassin, e J. M. Spitalier, “Value of Infrared Thermography in the Assessment of Malignant Melanomas of the Skin”, em Recent Advances in Medical Thermology, E. F. J. Ring e B. Phillips, Eds. Springer New York, 1984, pp. 623–629. [85] T. M. Buzug, S. Schumann, L. Pfaffmann, U. Reinhold, e J. Ruhlmann, “Functional infrared imaging for skin-cancer screening”, Conf. Proc. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. IEEE Eng. Med. Biol. Soc. Conf., vol. 1, pp. 2766–2769, 2006. [86] C. Herman e M. P. Cetingul, “Quantitative visualization and detection of skin cancer using dynamic thermal imaging”, J. Vis. Exp. JoVE, n. 51, 2011. [87] A. L. Shada, L. T. Dengel, G. R. Petroni, M. E. Smolkin, S. Acton, e C. L. Slingluff Jr., “Infrared thermography of cutaneous melanoma metastases”, J. Surg. Res., vol. 182, n. 1, pp. e9–e14, Jun. 2013. [88] M. Bonmarin e F.-A. Le Gal, “Lock-in thermal imaging for the early-stage detection of cutaneous melanoma: A feasibility study”, Comput. Biol. Med., vol. 47, pp. 36–43, Abr. 2014. [89] P. Bourjat, M. Gautherie, e E. Grosshans, “Diagnosis, follow-up and prognosis of malignant melanomas by thermography”, Bibl. Radiol., n. 6, pp. 115–127, 1975. [90] U. Michel, O. P. Hornstein, e A. Schönberger, “[Thermographico-histologic study of the lymph drainage areas in malignant melanoma]”, Hautarzt Z. Für Dermatol. Venerol. Verwandte Geb., vol. 37, n. 1, pp. 12–16, Jan. 1986. [91] G. A. Santa Cruz, J. Bertotti, J. Marín, S. J. González, S. Gossio, D. Alvarez, B. M. C. Roth, P. Menéndez, M. D. Pereira, M. Albero, L. Cubau, P. Orellano, e S. J. Liberman, “Dynamic infrared imaging of cutaneous melanoma and normal skin in patients treated with BNCT”, Appl. Radiat. Isot. Data Instrum. Methods Use Agric. Ind. Med., vol. 67, n. 7–8 Suppl, pp. S54–58, Jul. 2009. 70 [92] B. Kateb, V. Yamamoto, C. Yu, W. Grundfest, e J. P. Gruen, “Infrared thermal imaging: a review of the literature and case report”, NeuroImage, vol. 47 Suppl 2, pp. T154–162, Ago. 2009. [93] D. Mikulska, “[Thermographic examination of cutaneous melanocytic nevi]”, Ann. Acad. Med. Stetin., vol. 55, n. 1, pp. 31–38; discussion 38, 2009. [94] J. L. Ratz e P. L. Bailin, “Liquid-crystal thermography in determining the lateral extents of basal-cell carcinomas”, J. Dermatol. Surg. Oncol., vol. 7, n. 1, pp. 27–31, Jan. 1981. [95] J. H. Flores-Sahagun, J. V. C. Vargas, e F. A. Mulinari-Brenner, “Analysis and diagnosis of basal cell carcinoma (BCC) via infrared imaging”, Infrared Phys. Technol., vol. 54, n. 5, pp. 367–378, Set. 2011. [96] R. Amalric, E. Calas, P. Y. Castelain, C. Altschuler, e J. M. Spitalier, “[Dynamic telethermography in dermatology]”, Ann. Dermatol. Syphiligr. (Paris), vol. 102, n. 2, pp. 157–164, 1975. [97] M. P. Cetingul e C. Herman, “Identification of skin lesions from the transient thermal response using infrared imaging technique”, em 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, 2008. ISBI 2008, 2008, pp. 1219–1222. [98] M. Pirtini Çetingül e C. Herman, “The Assessment of Melanoma Risk Using the Dynamic Infrared Imaging Technique”, J. Therm. Sci. Eng. Appl., vol. 3, n. 3, pp. 031006– 031006, Ago. 2011. [99] R. Vardasca, F. Ring, P. Plassmann, e C. Jones, “Thermal symmetry of the upper and lower extremities in healthy subjects”, Thermol. Int., vol. 22, pp. 53–60, Fev. 2012. [100] H. H. Niu, P. W. Lui, J. S. Hu, C. K. Ting, Y. C. Yin, Y. L. Lo, L. Liu, e T. Y. Lee, “Thermal symmetry of skin temperature: normative data of normal subjects in Taiwan”, Zhonghua Yi Xue Za Zhi Chin. Med. J. Free China Ed, vol. 64, n. 8, pp. 459– 468, Ago. 2001. [101] R. A. Sherman, A. L. Woerman, e K. W. Karstetter, “Comparative effectiveness of videothermography, contact thermography, and infrared beam thermography for scanning relative skin temperature”, J. Rehabil. Res. Dev., vol. 33, n. 4, pp. 377–386, Out. 1996. [102] K. U. Frerichs e G. Z. Feuerstein, “Laser-doppler flowmetry”, Mol. Chem. Neuropathol., vol. 12, n. 1, pp. 55–70, Jan. 1990. [103] O. Schlager, M. E. Gschwandtner, K. Herberg, T. Frohner, M. Schillinger, R. Koppensteiner, e W. Mlekusch, “Correlation of infrared thermography and skin perfusion in Raynaud patients and in healthy controls”, Microvasc. Res., vol. 80, n. 1, pp. 54–57, Jul. 2010. [104] A. Serov, B. Steinacher, e T. Lasser, “Full-field laser Doppler perfusion imaging and monitoring with an intelligent CMOS camera”, Opt. Express, vol. 13, n. 10, pp. 3681– 3689, Mai. 2005. [105] H. Fujii, K. Nohira, Y. Yamamoto, H. Ikawa, e T. Ohura, “Evaluation of blood flow by laser speckle image sensing. Part 1”, Appl. Opt., vol. 26, n. 24, pp. 5321–5325, Dez. 1987. [106] J. D. Briers, G. Richards, e X. W. He, “Capillary Blood Flow Monitoring Using Laser Speckle Contrast Analysis (LASCA)”, J. Biomed. Opt., vol. 4, n. 1, pp. 164–175, Jan. 1999. [107] “IBM SPSS software”, 12-Jun-2014. [On-line]. Disponível em: http://www01.ibm.com/software/analytics/spss/. [Acedido: 25-Ago-2014]. [108] C. Hildebrandt, C. Raschner, e K. Ammer, “An Overview of Recent Application of Medical Infrared Thermography in Sports Medicine in Austria”, Sensors, vol. 10, n. 5, pp. 4700–4715, Mai. 2010. [109] Schwartz,Robert G. , MD, Chair, “Guidelines For Neuromusculoskeletal Thermography”, Thermology international 2006; 16: 5-9. [110] F. H. Netter, Netter Atlas De Anatomia Humana. Elsevier Brasil, 2011. [111] B. M. Gratt, A. Pullinger, E. A. Sickles, e J. J. Lee, “Electronic thermography of normal facial structures: A pilot study”, Oral Surg. Oral Med. Oral Pathol., vol. 68, n. 3, pp. 346–351, Set. 1989. 71 [112] R. E. Mangine, K. A. Siqueland, e F. R. Noyes, “The use of thermography for the diagnosis and management of patellar tendinitis”, J. Orthop. Sports Phys. Ther., vol. 9, n. 4, pp. 132–140, 1987. [113] A. F. Luis Matinez, Análise de dados com SPSS, Primeiros passos., 3a ed. Escolar editora, 2010. [114] S. Pothiawala, A. A. Qureshi, Y. Li, e J. Han, “Obesity and the incidence of skin cancer in US Caucasians”, Cancer Causes Control CCC, vol. 23, n. 5, pp. 717–726, Mai. 2012. [115] “Medical Imaging @ UoG”. [On-line]. Disponível em: http://www.medimaging.org/. [Acedido: 16-Nov-2013]. [116] J. Melia, C. Harland, S. Moss, J. R. Eiser, e L. Pendry, “Feasibility of targeted early detection for melanoma: a population-based screening study”, Br. J. Cancer, vol. 82, n. 9, pp. 1605–1609, Mai. 2000. [117] J. Tacke, J. Dietrich, B. Steinebrunner, e A. Reifferscheid, “Assessment of a new questionnaire for self-reported sun sensitivity in an occupational skin cancer screening program”, BMC Dermatol., vol. 8, n. 1, p. 4, Out. 2008. 78 1.De acordo com os tópicos abaixo descritos, selecione para cada pergunta, a opção com qual se identifica. Questões Respostas Pontuação Qual é a cor da sua pele quando não está bronzeada? Avermelhada Clara/Pálida Castanho claro Castanho escuro 1 2 3 4 Tem sardas? Sim, muitas Sim, algumas Sim, poucos Não 1 2 3 4 Como é que a sua pele reage à exposição solar? Muito sensível, susceptível a irritação Sensível, por vezes com irritação Geralmente sensível, raramente irritação Insensível, nenhuma irritação 1 2 3 4 Quanto tempo é que se pode expor ao sol, num céu sem nuvens, ao meio dia, sem ficar com queimaduras? Menos de 15 minutos Entre 15 a 20 minutos Entre 25 e 40 minutos Mais de 40 minutos 1 2 3 4 Como é que a sua pele reage à longa exposição solar? Sempre com queimaduras solares Principalmente com queimaduras solares Muitas vezes com queimaduras solares Raramente ou nunca com queimaduras solares 1 2 3 4 Que tipo de queimaduras costuma desenvolver sempre que se expõe ao sol durante muito tempo? Vermelhão forte, bolhas dolorosas, descamação da pele Vermelho claro, seguido por descamação Avermelhamento, seguido por vezes por descamação Quase nunca vermelhão e descamação 1 2 3 4 Alguma vez experimentou efeito de bronzeamento após banhos de sol durante um período de tempo prolongado? Nunca Quase nunca Frequentemente Quase sempre 1 2 3 4 Como funciona o seu bronzeado após banhos de sol repetidos? Nunca ou quase nunca Bronze mínimo Progressivo, bronze claro Rápido, bronze profundo 1 2 3 4 Como descreve a sua cor natural de cabelo? Vermelho até castanho avermelhado Castanho claro até loiro Loiro escuro até castanho Castanho escuro até preto 1 2 3 4 Como descreve a sua cor de olhos? Azul claro, cinza, ou verde claro Azul, cinza ou verde Castanho claro ou cinza escuro Castanho escuro 1 2 3 4 Total 2. Tem parentes com melanomas? a. Sim b. Não c. Não sei 2.1.Se respondeu sim à questão anterior, indique qual o grau de parentesco. a. Pai/mãe b. Tio/Tia c. Irmão/Irmã d. Filho/Filha e. Outro 79 *Com base em [116, 117]. Nota importante: A pontuação obtida neste questionário diz respeito à sensibilidade solar da pele. Tendo um mínimo de 10, representando alta sensibilidade solar, enquanto que um máximo de 40, representa baixa sensibilidade solar 80 APPENDIXE V: ANS results of p-value and ranking values to right side of each phase Moment I to II Moment II to III Moment I to III U W Z p U W Z p U W Z p Frontal R1 66.0 121.0 -0.2 0.80 44.0 99.0 -1.5 0.10 69.5 124.5 0.0 1.00 R2 63.5 118.5 -0.4 0.70 47.5 102.5 -1.3 0.20 63.0 118.0 -0.4 0.70 R3 56.5 111.5 -0.8 0.40 59.5 114.5 -0.6 0.50 65.5 120.5 -0.3 0.80 Profile R1 64.0 119.0 -0.4 0.70 42.0 97.0 -1.6 0.10 70.0 125.0 0.0 1.00 R2 62.0 117.0 -0.5 0.60 52.0 107.0 -1.1 0.30 70.0 125.0 0.0 1.00 R3 61.0 116.0 -0.5 0.60 56.5 111.5 -0.8 0.40 66.5 171.5 -0.2 0.80 81 APPENDICE VI: ANS mean temperatures to left side of each phase and results of p-value and ranking values ROIs Phase 1st Moment 2nd Moment 3rd Moment Frontal R1 1 34.64 ± 0.43 32.76 ± 0.84 34.21 ± 0.73 2 34.54 ± 0.71 32.24 ± 0.67 34.25 ± 0.77 R2 1 33.97 ± 0.69 32.46 ± 0.66 33.72 ± 0.82 2 33.90 ± 0.90 32.11 ± 1.04 33.78 ± 0.91 R3 1 33.72 ± 0.69 31.95 ± 0.70 33.18 ± 0.99 Profile 2 33.10 ± 0.74 31.56 ± 1.07 33.12 ± 0.84 R1 1 34.28 ± 0.44 32.54 ± 0.79 33.73 ± 0.75 2 33.98 ± 0.62 32.15 ± 0.69 33.78 ± 0.72 R2 1 33.61 ± 0.79 31.90 ± 0.85 32.87 ± 0.84 2 33.26 ± 0.98 31.61 ± 1.06 32.87 ± 1.02 R3 1 33.79 ± 0.74 32.12 ± 0.64 32.99 ± 0.86 2 33.51 ± 0.74 31.87 ± 0.97 33.09 ± 0.74 Moment I to II Moment II to III Moment I to III U W Z p U W Z p U W Z p Frontal R1 66 121 -0.24 0.81 39 94 -1.79 0.07 67 172 -0.18 0.86 R2 69 124 -0.06 0.95 56 111 -0.82 0.41 66 171 -0.20 0.84 R3 36 91 -1.96 0.05 55 110 -0.85 0.4 60 115 -0.56 0.58 Profile R1 50 105 -1.18 0.24 50 105 -1.14 0.25 70 125 0.00 1.00 R2 53 108 -1.00 0.32 62 117 -0.47 0.64 65 170 -0.29 0.77 R3 55 110 -0.88 0.38 60 115 -0.56 0.58 68 173 -0.12 0.91 82 Appendix VII: MS mean temperatures to right side of each gender and results of p-value and U,W an Z values ROIs Gender 1st Moment 2nd Moment 3rd Moment Frontal R1 F 32.06 ± 1.56 30.33 ± 1.44 31.98 ± 1.86 M 32.41 ± 1.16 31.51 ± 1.28 33.43 ± 1.41 R2 F 33.88 ± 0.91 32.38 ± 0.93 33.75 ± 1.02 M 33.98 ± 0.73 32.75 ± 0.76 34.26 ± 0.80 R3 F 33.99 ± 0.81 32.05 ± 1.00 33.66 ± 0.97 M 33.96 ± 0.78 32.47 ± 0.71 33.98 ± 0.71 R4 F 34.56 ± 0.53 32.51 ± 0.81 34.11 ± 0.78 M 34.68 ± 0.43 33.03 ± 0.41 34.49 ± 0.35 R5 F 34.28 ± 0.46 32.94 ± 0.59 33.95 ± 0.58 M 34.09 ± 0.44 33.06 ± 0.39 33.97 ± 0.24 R6 F 33.18± 1.06 31.08 ± 1.07 32.46 ± 1.18 M 33.81 ±0.62 32.17 ± 0.53 33.78 ± 0.58 R7 F 33.59 ± 0.95 31.78 ± 1.13 33.07 ± 1.16 M 34.01 ± 0.88 32.63 ± 0.69 33.94 ± 0.66 R8 F 34.60 ± 0.50 33.19 ± 0.79 34.36 ± 0.64 M 34.78 ± 0.31 33.53 ± 0.45 34.74 ± 0.41 R9 F 32.93 ± 0.68 31.26 ± 0.72 32.32 ± 0.73 M 33.67 ± 0.45 32.31 ± 0.30 33.39 ± 0.49 Profile R1 F 30.90 ± 2.25 29.72 ± 2.07 31.04 ± 2.59 M 31.73 ± 1.77 31.29 ± 1.90 32.96 ± 2.13 R2 F 33.47 ± 1.15 32.32 ± 0.99 33.59 ± 0.97 M 33.13 ± 0.96 32.44 ± 1.11 33.58 ± 1.10 R3 F 33.71 ± 0.79 31.85 ± 0.94 33.45 ± 0.90 M 33.96 ± 0.47 32.50± 0.64 33.96 ± 0.57 R4 F 34.26 ± 0.66 32.60 ± 0.81 33.86 ± 0.77 M 34.38 ± 0.53 33.03 ± 0.45 34.23 ± 0.34 R5 F 33.98 ± 0.50 32.75 ± 0.65 33.71 ± 0.64 M 33.91 ± 0.57 33.05 ± 0.40 33.91 ± 0.29 R6 F 32.60 ± 1.12 30.77 ± 1.14 32.05 ± 1.19 M 33.76 ± 0.61 32.30 ± 0.46 33.62 ± 0.52 R7 F 33.39 ± 0.89 31.6 ± 0.99 32.84 ± 1.07 M 34.28 ± 0.62 32.99 ± 0.58 34.07 ± 0.60 R8 F 33.37± 0.88 31.70 ± 0.84 32.73 ± 0.84 M 34.19 ± 0.55 32.91 ± 0.41 33.86 ± 0.50 R9 F 34.45 ± 0.82 32.7 ± 0.94 33.68 ± 0.77 M 34.45 ± 0.57 33.44 ± 0.66 34.31 ± 0.48 83 R10 F 33.16 ± 0.99 31.40 ± 0.92 32.46 ± 0.94 M 34.16 ± 0.59 32.70 ± 0.50 33.85 ± 0.63 ROI's Moment I to II Moment II to III Moment I to III U W Z p U W Z p U W Z p Frontal R1 166 466 -0.70 0.48 100 400 -2.54 0.01 94 394 -2.69 0.01 R2 187 487 -0.14 0.89 158 458 -0.94 0.35 132 432 -1.65 0.10 R3 186 322 -0.17 0.87 151 451 -1.12 0.26 153 453 -1.08 0.28 R4 173 473 -0.51 0.61 107 407 -2.35 0.02 143 443 -1.36 0.18 R5 147 283 -1.23 0.22 172 472 -0.55 0.58 176 312 -0.43 0.67 R6 130 430 -1.70 0.09 70 370 -3.37 0.00 55 355 -3.79 0.00 R7 140 440 -1.44 0.15 115 415 -2.12 0.03 98 98 -2.60 0.01 R8 155 455 -1.03 0.30 137 437 -1.51 0.13 126 426 -1.83 0.07 R9 72 372 -3.32 0.00 39 339 -4.23 0.00 35 335 -4.34 0.00 Profile R1 150 450 -1.15 0.25 105 405 -2.39 0.02 101 401 -2.51 0.01 R2 143 279 -1.34 0.18 178 478 -0.37 0.71 190 326 -0.06 0.96 R3 162 462 -0.82 0.41 109 409 -2.29 0.02 135 435 -1.58 0.12 R4 180 480 -0.32 0.75 139 439 -1.47 0.14 152 452 -1.11 0.27 R5 186 322 -0.15 0.88 150 450 -1.16 0.25 173 473 -0.51 0.61 R6 78 378 -3.14 0.00 31 331 -4.44 0.00 39 339 -4.21 0.00 R7 82 382 -3.04 0.00 33 333 -4.4 0.00 61 361 -3.62 0.00 R8 86 386 -2.92 0.00 37 337 -4.29 0.00 48 348 -3.98 0.00 R9 180 316 -0.33 0.74 100 400 -2.53 0.01 92 392 -2.75 0.01 R10 79 379 -3.11 0.00 34 334 -4.37 0.00 35 335 -4.33 0.00 84 Appendix VIII: MS mean temperatures to left side of each gender and results of p-value and U,W an Z values ROIs Gender 1st Moment 2nd Moment 3rd Moment Frontal R1 F 32.15 ± 1.57 30.59 ± 1.51 32.16 ± 1.90 M 32.54 ± 1.25 31.66 ± 1.35 33.36 ± 1.48 R2 F 33.98 ± 0.64 32.54 ± 0.74 33.87 ± 0.82 M 33.84 ± 0.75 32.66 ± 0.79 34.01 ± 0.64 R3 F 33.95 ± 0.71 32.08 ± 0.84 33.50 ± 0.96 M 33.87 ± 0.86 32.44 ± 0.70 33.86 ± 0.69 R4 F 34.60 ± 0.54 32.60 ± 0.81 34.10 ± 0.83 M 34.61 ± 0.41 32.93 ± 0.34 34.39 ± 0.31 R5 F 34.30 ± 0.51 32.95 ± 0.63 33.95 ± 0.59 M 34.06 ± 0.38 33.00 ± 0.37 33.93 ± 0.23 R6 F 33.18 ± 1.08 31.18 ± 1.05 32.74 ± 1.18 M 33.79 ± 0.67 32.17 ± 0.63 33.75 ± 0.70 R7 F 33.64 ± 0.86 31.95 ± 0.83 33.17 ± 1.00 M 33.88 ± 0.78 32.61 ± 0.57 33.78 ± 0.65 R8 F 34.47 ± 0.66 33.08 ± 0.76 34.13 ± 0.61 M 34.65 ± 0.46 33.36 ± 0.40 34.54 ± 0.38 R9 F 32.93 ± 0.81 31.30 ± 0.91 32.29 ± 0.86 M 33.56 ± 0.47 32.31 ± 0.77 33.35 ± 0.42 Profile R1 F 30.93 ± 2.32 29.80 ± 2.17 31.03 ± 2.96 M 31.69 ± 1.78 31.29 ± 1.99 32.94 ± 2.22 R2 F 33.42 ± 0.91 32.32 ± 0.91 33.57 ± 1.04 M 32.92 ± 1.06 32.36 ± 1.18 33.59 ± 0.89 R3 F 33.61 ± 0.70 31.99 ± 0.90 33.32 ± 0.93 M 33.92 ± 0.62 32.58 ± 0.77 33.88 ± 0.60 R4 F 34.39 ± 0.55 32.71 ± 0.72 33.95 ± 0.65 M 34.36 ± 0.59 32.91 ± 0.45 34.15 ± 0.45 R5 F 34.00 ± 0.57 32.76 ± 0.74 33.68 ± 0.85 M 33.93 ± 0.45 32.94 ± 0.45 33.21 ± 2.69 R6 F 32.75 ± 1.10 31.07 ± 1.13 32.22 ± 1.22 M 33.76 ± 0.68 32.27 ± 0.65 33.62 ± 0.70 R7 F 33.60 ± 0.78 32.00 ± 0.79 33.14 ± 0.93 M 34.27 ± 0.62 33.03 ± 0.48 34.09 ± 0.52 R8 F 33.38 ± 0.89 31.85 ± 0.81 32.72 ± 0.90 M 34.24 ± 0.58 32.92 ± 0.47 33.93 ± 0.50 R9 F 34.28 ± 0.77 32.73 ± 0.85 33.62 ± 0.61 M 34.64 ± 0.44 33.38 ± 0.78 34.35 ± 0.57 R10 F 33.02 ± 0.99 31.42 ± 0.96 32.35 ± 0.92 M 34.18 ± 0.65 32.84 ± 0.61 33.96 ± 0.74 85 ROI's Moment I to II Moment II to III Moment I to III U W Z p U W Z p U W Z p Frontal R1 159 459 -0.89 0.36 112 412 -2.21 0.02 121 421 -1.96 0.05 R2 172 308 -0.55 0.58 189 489 -0.06 0.94 178 478 -0.37 0.70 R3 182 318 -0.26 0.79 148 448 -1.20 0.22 144 444 -1.31 0.19 R4 190 326 -0.05 0.95 134 434 -1.60 0.10 164 464 -0.77 0.43 R5 121 257 -1.96 0.04 190 490 -0.05 0.95 189 325 -0.08 0.93 R6 139 439 -1.45 0.14 89 389 -2.84 0.04 86 386 -2.93 0.03 R7 166 466 -0.71 0.47 99 399 -2.55 0.01 111 411 -2.22 0.02 R8 157 457 -0.96 0.33 146 446 -1.25 0.20 117 417 -2.06 0.03 R9 111 411 -2.24 0.02 75 375 -3.21 0.01 52 352 -3.87 0.00 Profile R1 147 447 -1.24 0.21 117 417 -2.07 0.03 118 418 -2.04 0.04 R2 135 271 -1.56 0.11 177 313 -0.41 0.67 183 319 -0.23 0.81 R3 144 444 -1.31 0.18 121 421 -1.94 0.05 118 418 -2.04 0.04 R4 186 322 -0.16 0.86 151 451 -1.12 0.26 160 460 -0.88 0.37 R5 175 311 -0.45 0.64 177 477 -0.40 0.68 179 315 -0.36 0.71 R6 98 398 -2.59 0.00 68 368 -3.41 0.01 51 351 -3.88 0.00 R7 102 402 -2.47 0.01 48 348 -3.98 0.00 65 365 -3.50 0.00 R8 83 383 -2.99 0.00 52 352 -3.87 0.00 47 347 -4.00 0.00 R9 142 442 -1.38 0.16 108 408 -2.30 0.02 70 370 -3.36 0.01 R10 64 364 -3.52 0.00 37 337 -4.27 0.00 29 329 -4.50 0.00 86 Appendix IX: MS thermal symmetry of each gender and results of pvalue, U,W and Z values ROIs Gender 1st Moment 2nd Moment 3rd Moment Frontal R1 F 0.26 ± 0.02 0.32 ± 0.31 0.22 ± 0.16 M 0.34 ± 0.52 0.39 ± 0.48 0.22 ± 0.18 R2 F 0.41 ± 0.37 0.52 ± 0.40 0.45 ± 0.44 M 0.34 ± 0.25 0.48 ± 0.48 0.40 ± 0.30 R3 F 0.30 ± 0.20 0.38 ± 0.20 0.39 ± 0.33 M 0.19 ± 0.15 0.19 ± 0.16 0.21 ± 0.17 R4 F 0.15 ± 0.14 0.26 ± 0.26 0.20 ± 0.15 M 0.14 ± 0.14 0.24 ± 0.21 0.19 ± 0.16 R5 F 0.15 ± 0.08 0.16 ± 0.12 0.14 ± 0.11 M 0.16 ± 0.15 0.14 ± 0.14 0.09 ± 0.10 R6 F 0.47 ± 0.63 0.30 ± 0.24 0.44 ± 0.45 M 0.25 ± 0.18 0.29 ± 0.22 0.28 ± 0.19 R7 F 0.37 ± 0.41 0.54 ± 0.31 0.43 ± 0.52 M 0.33 ± 0.21 0.33 ± 0.16 0.22 ± 0.18 R8 F 0.30 ± 0.41 0.36 ± 0.24 0.35 ± 0.39 M 0.25 ± 0.23 0.28 ± 0.17 0.27 ± 0.23 R9 F 0.35 ± 0.29 0.41 ± 0.32 0.35 ± 0.23 M 0.34 ± 0.26 0.38 ± 0.52 0.40 ± 0.31 Profile R1 F 0.83 ± 0.68 1.03 ± 0.76 0.90 ± 0.59 M 0.87 ± 0.56 0.51 ± 0.44 0.44 ± 0.40 R2 F 0.55 ± 0.40 0.56 ± 0.44 0.44 ± 0.38 M 0.39 ± 0.48 0.50 ± 0.39 0.35 ± 0.27 R3 F 0.40 ± 0.29 0.46 ± 0.41 0.34 ± 0.30 M 0.22 ± 0.21 0.39 ± 0.30 0.20 ± 0.24 R4 F 0.22 ± 0.27 0.40 ± 0.26 0.23 ± 0.28 M 0.26 ± 0.15 0.38 ± 0.34 0.22 ± 0.21 R5 F 0.20 ± 0.15 0.33 ± 0.43 0.35 ± 0.65 M 0.32 ± 0.21 0.21 ± 0.20 0.86 ± 2.58 R6 F 0.31 ± 0.23 0.43 ± 0.39 0.34 ± 0.34 M 0.28 ± 0.21 0.37 ± 0.28 0.35 ± 0.27 R7 F 0.31 ± 0.26 0.53 ± 0.40 0.50 ± 0.38 M 0.18 ± 0.14 0.26 ± 0.25 0.24 ± 0.16 R8 F 0.28 ± 0.24 0.30 ± 0.26 0.22 ± 0.18 M 0.31 ± 0.21 0.15 ± 0.12 0.20 ± 0.20 87 R9 F 0.44 ± 0.45 0.41 ± 0.35 0.35 ± 0.25 M 0.38 ± 0.32 0.31 ± 0.19 0.28 ± 0.29 R10 F 0.39 ± 0.23 0.27 ± 0.23 0.31 ± 0.24 M 0.34 ± 0.17 0.33 ± 0.23 0.31 ± 0.28 ROI's Moment I to II Moment II to III Moment I to III U W Z p U W Z p U W Z p Frontal R1 175 311 -0.48 0.63 181 481 -0.29 0.76 186 322 -0.15 0.87 R2 181 317 -0.30 0.75 174 310 -0.50 0.61 185 485 -0.19 0.84 R3 128 264 -1.79 0.07 93 229 -2.76 0.00 113 249 -2.20 0.02 R4 189 325 -0.07 0.94 190 326 -0.05 0.95 176 312 -0.43 0.66 R5 175 311 -0.47 0.63 168 304 -0.66 0.50 132 268 -1.74 0.08 R6 141 277 -1.42 0.13 190 326 -0.04 0.96 156 292 -1.00 0.31 R7 178 478 -0.37 0.76 104 240 -2.45 0.01 158 294 -0.95 0.33 R8 192 328 0.00 1 158 294 -0.93 0.34 189 325 -0.07 0.94 R9 190 490 -0.04 0.96 146 282 -1.26 0.20 164 464 -0.76 0.44 Profile R1 171 471 -0.58 0.56 110 246 -2.25 0.02 105 241 -2.40 0.01 R2 141 277 -1.41 0.15 173 309 -0.51 0.60 171 307 -0.56 0.56 R3 119 255 -2.04 0.04 178 314 -0.39 0.69 136 272 -1.57 0.11 R4 132 432 -1.69 0.09 172 308 -0.54 0.58 178 478 -0.38 0.70 R5 124 424 -1.91 0.05 156 292 -1.01 0.31 191 327 -0.02 0.97 R6 184 320 -0.22 0.82 182 318 -0.27 0.78 176 476 -0.44 0.65 R7 138 274 -1.51 0.13 111 247 -2.24 0.02 118 254 -2.04 0.04 R8 166 466 -0.72 0.47 118 254 -2.09 0.03 168 304 -0.68 0.49 R9 188 324 -0.09 0.92 167 303 -0.69 0.48 146 282 -1.27 0.20 R10 183 319 -0.23 0.81 155 455 -1.02 0.30 185 321 -0.18 0.85