Sou iane Kheli i Touhami. (2025). Place o A i icial In elligence in Oncology: Li e a u e e iew and
Pe spec i es. MAR Oncology and Hema ology. (2025) 5:07
Place o A i icial In elligence in Oncology: Li e a u e e iew and
Pe spec i es
Sou iane Kheli i Touhami1*, Abdelaziz Amma i2, Mohamed Dje ouni3, Fayez Al hobi i4, Abdul ahmane
Al ubayee5.
1,3,4,5. Medical Oncology consul an a King Abdulaziz Specialis Hospi al Tai Saudi A abia.
2- P o esso o Medical Oncology a Uni e si y 3 o Cons an ine Alge ia.
*Co espondence o: Sou iane Kheli i Touhami, Medical Oncology consul an a King Abdulaziz
Specialis Hospi al Tai Saudi A abia.
Copy igh .
© 2025 Sou iane Kheli i Touhami This is an open access a icle dis ibu ed unde he C ea i e Commons
A ibu ion License, which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided
he o iginal wo k is p ope ly ci ed.
Recei ed: 13 Aug 2025
Published: 20 Aug 2025
MAR Oncology and Hema ology (2025) 5:07
Re iew A icle
Sou iane Kheli i Touhami, MAR Oncology and Hema ology (2025) 5:07.
Page 2 o 10
Sou iane Kheli i Touhami. (2025). Place o A i icial In elligence in Oncology: Li e a u e e iew and
Pe spec i es. MAR Oncology and Hema ology. (2025) 5:07
A i icial In elligence and Heal hca e
A i icial In elligence (AI) is a sub ield o compu e science conce ned wi h algo i hms ha pe o m asks
ypically associa ed wi h human cogni ion. AI has been applied o medical p oblems o many yea s, bu
accele a ed adop ion in heal hca e is now mo i a ed by a g ow h in da a olume and comp ehensi eness,
imp o emen s in compu a ional powe , and an inc ease in inno a i e so wa e de elope s. Mode n AI
echnology can p o ide signi ican insigh and bene i by p ocessing a di e se se o da a ypes ela ed o a
p oblem domain. Beyond he capabili y o ecognize pa e ns in da a, AI algo i hms can syn hesize la ge
olumes o in o ma ion om mul iple sou ces quickly. In heal hca e, common applica ions include iden i ying
condi ions, isk ac o s, o pa e ns o suppo clinical decision-making, popula ion heal h, and disco e y.
Cance is a b oad class o de as a ing diseases ha g ows exponen ially in complexi y as i p og esses, c ea ing
a di icul analy ical space o easoning and managemen . Ongoing wo k sugges s ha AI may be pa icula ly
well sui ed o mee ing his challenge, bu widesp ead impac in he clinical domain emains elusi e.
A i icial In elligence and Oncology
Cance is a leading cause o dea h wo ldwide and is p ojec ed o become mo e common wi h age. This ou look
highligh s he impo ance o seconda y p e en ion h ough ea ly diagnosis and imp o ed ea men wo a eas
ha ha e unde gone signi ican ans o ma ion aided by a i icial in elligence. AI enhances ea ly diagnosis by
enabling au oma ed, imp o ed accu acy in cance iden i ica ion. I suppo s imp o ed pa ien ou comes
h ough ea men planning, umo de ec ion, segmen a ion, and g ading. Oncology encompasses he s udy
and ea men o malignan g ow hs o med by he con e gence o uncon olled p oli e a ion, diminished
apop osis, and al e ed cell senescence mechanisms. Cance can a ise wi hin almos all issue ypes and is
di ided in o a ious g oups. Indi idual issue ypes may be associa ed wi h a a ie y o di e en umo s ha
span a wide clinical and biological spec um and ha e di e en p ognoses. These include benign and malignan
ca ego ies, egimes o clinical beha io , and li e span. Following app op ia e egene a i e, p oli e a ion and
di e en ia ion mechanisms, a a ie y o changes a e associa ed wi h he o ma ion o he cance ous s a e which
include cellula ea u es such as inc eased adap abili y, issue in asion and sys emic sp ead. T ea men op ions
o oncology pa ien s include adio he apy, su ge y and sys emic he apies which can ei he be in he o m o
chemo he apy, a ge ed agen s o immuno he apy.
A i icial In elligence and Cance Diagnosis
A i icial In elligence in Image Analysis and Radiology : AI d i en image analysis in oncologic adiology
enhances de ec ion, cha ac e iza ion, and moni o ing o malignancies, suppo s p ecision medicine, and
op imizes wo k low. While p omising, success ul clinical in eg a ion equi es add essing challenges ela ed
Sou iane Kheli i Touhami, MAR Oncology and Hema ology (2025) 5:07.
Page 3 o 10
Sou iane Kheli i Touhami. (2025). Place o A i icial In elligence in Oncology: Li e a u e e iew and
Pe spec i es. MAR Oncology and Hema ology. (2025) 5:07
o da a quali y, in e p e abili y, and egula o y compliance. Ongoing esea ch and collabo a ion be ween
clinicians, da a scien is s, and indus y a e essen ial o ealizing he ull po en ial o AI in oncologic imaging
[1].
Pa hology and His opa hology : A i icial in elligence is inc easingly in eg a ed in o pa hology and
his opa hology, ans o ming diagnos ic wo k lows and esea ch. AI, pa icula ly deep lea ning, enables
au oma ed analysis o digi al whole-slide images, suppo ing asks such as umo de ec ion, g ading, and
sub yping wi h high accu acy and ep oducibili y. AI-d i en image analysis can educe in e obse e
a iabili y, imp o e e iciency, and assis in quan i ying bioma ke s and he apeu ic a ge s [2].
Radiomics and compu a ional pa hology app oaches allow ex ac ion o high-dimensional ea u es om issue
images, suppo ing p ognos ica ion and p ecision medicine. AI also acili a es quali y con ol, iage, and
educa ional applica ions. Howe e , challenges emain ega ding da a s anda diza ion, model in e p e abili y,
ex e nal alida ion, and egula o y e hical conside a ions. Ongoing esea ch ocuses on explainable AI and
obus clinical in eg a ion [3].
A i icial In elligence and Cance T ea men Planning
A i icial in elligence is inc easingly u ilized in cance ea men planning, pa icula ly in adia ion oncology
and sys emic he apy selec ion. AI algo i hms especially hose based on machine lea ning and deep lea ning
can au oma e and op imize complex planning asks, such as umo and o gan-a - isk segmen a ion, dose
dis ibu ion p edic ion, and plan quali y assessmen [4].
These sys ems imp o e e iciency, educe in e obse e a iabili y, and can gene a e ea men plans ha
ma ch o exceed he quali y o hose p oduced by human expe s. AI also suppo s indi idualized he apy by
in eg a ing mul imodal da a (imaging, pa hology, genomics) o p edic ea men esponse and pe sonalize
egimens. Despi e hese ad ances, challenges emain ega ding da a quali y, model in e p e abili y, and
clinical in eg a ion[5].
Machine Lea ning Algo i hms in Oncology
Machine lea ning (ML) algo i hms a e inc easingly u ilized in oncology o enhance cance de ec ion,
diagnosis, p ognosis, and ea men pe sonaliza ion. Supe ised lea ning me hods such as suppo ec o
machines (SVM), andom o es s, and deep lea ning (especially con olu ional neu al ne wo ks, CNNs) a e
widely applied o imaging, pa hology, and mul i-omics da a o umo classi ica ion, segmen a ion, and isk
s a i ica ion [6]
Unsupe ised lea ning (e.g., clus e ing) helps iden i y no el cance sub ypes and pa ien g oups. ML models
can p edic ea men esponse, ecu ence isk, and su i al, suppo ing p ecision oncology. In eg a ion o
Sou iane Kheli i Touhami, MAR Oncology and Hema ology (2025) 5:07.
Page 4 o 10
Sou iane Kheli i Touhami. (2025). Place o A i icial In elligence in Oncology: Li e a u e e iew and
Pe spec i es. MAR Oncology and Hema ology. (2025) 5:07
ML in o clinical wo k lows imp o es e iciency and may educe diagnos ic e o s, bu challenges emain
ega ding da a quali y, model in e p e abili y, and gene alizabili y. Ongoing esea ch ocuses on explainable
AI and obus alida ion o sa e clinical adop ion [7]
Na u al Language P ocessing in Cance Resea ch
Na u al language p ocessing (NLP) is inc easingly used in cance esea ch o ex ac , s uc u e, and analyze
in o ma ion om uns uc u ed clinical ex , such as elec onic heal h eco ds (EHRs), pa hology epo s,
adiology epo s, and pa ien na a i es [8]
NLP enables au oma ed iden i ica ion o cance diagnoses, s aging, ea men egimens, ad e se e en s, and
ou comes, signi ican ly educing manual cha e iew wo kload and imp o ing da a quali y o esea ch and
clinical decision suppo . Ad anced NLP models, including ans o me -based a chi ec u es (e.g., BERT,
GPT), ha e imp o ed he accu acy o in o ma ion ex ac ion, coho iden i ica ion, and pheno yping. NLP
also acili a es he analysis o pa ien - epo ed ou comes, social de e minan s o heal h, and pa ien
pe spec i es, suppo ing pa ien -cen e ed esea ch and p ecision oncology. Key challenges include a iabili y
in clinical language, da a p i acy, gene alizabili y ac oss ins i u ions, and in eg a ion in o clinical wo k lows
[9].
P edic i e Analy ics in Oncology
P edic i e analy ics in oncology le e ages s a is ical and machine lea ning models o es ima e indi idual
pa ien isk, guide clinical decision-making, and imp o e ou comes. Risk s a i ica ion models use clinical,
pa hological, imaging, and molecula da a o ca ego ize pa ien s by likelihood o ecu ence, p og ession, o
ea men esponse. Su i al p edic ion models, including Cox p opo ional haza ds, andom o es s, and deep
lea ning app oaches, es ima e o e all o disease- ee su i al p obabili ies. These ools suppo pe sonalized
ea men planning, iden i y high- isk pa ien s o in ensi ied he apy, and in o m p ognosis discussions.
In eg a ion o mul i-omics and eal-wo ld da a is enhancing model accu acy. Key challenges include model
alida ion, in e p e abili y, and in eg a ion in o clinical wo k lows [10].
A i icial In elligence and Clinical T ials
A i icial in elligence is inc easingly ans o ming he design, conduc , and analysis o clinical ials. AI
applica ions include:
Pa ien Rec ui men and Eligibili y: AI algo i hms e icien ly sc een elec onic heal h eco ds and o he
da a sou ces o iden i y eligible pa icipan s, imp o ing ec ui men speed and di e si y [11].
T ial Design Op imiza ion: Machine lea ning models assis in adap i e ial designs, endpoin selec ion, and
Sou iane Kheli i Touhami, MAR Oncology and Hema ology (2025) 5:07.
Page 5 o 10
Sou iane Kheli i Touhami. (2025). Place o A i icial In elligence in Oncology: Li e a u e e iew and
Pe spec i es. MAR Oncology and Hema ology. (2025) 5:07
sample size es ima ion, enhancing ial e iciency and in o ma i eness.
Risk Assessmen and Moni o ing: AI suppo s eal- ime sa e y moni o ing, ad e se e en p edic ion, and
ope a ional isk assessmen , enabling p oac i e isk-based moni o ing.
Da a Collec ion and Analysis: Na u al language p ocessing and AI-d i en ools au oma e ex ac ion and
ha moniza ion o da a om uns uc u ed sou ces, while ad anced analy ics enable deepe insigh s om
complex da ase s.
Pa ien -Repo ed Ou comes: AI-powe ed cha bo s and digi al ools acili a e eal- ime collec ion and
analysis o pa ien - epo ed ou comes, suppo ing pe sonalized and da a-d i en decision-making [12].
D ug De elopmen : AI accele a es d ug disco e y, epu posing, and bioma ke iden i ica ion, s eamlining
he ansi ion om p eclinical o clinical phases[13].
E hical Conside a ions in A i icial In elligence Applica ions
AI applica ions in heal hca e exempli y he echnological e olu ion ha can ad ance cance ca e and esea ch
as one o he mos p omising domains o hem. By adop ing e hical guidelines and open communica ion wi h
he end-use , e hical challenges ini ially encoun e ed in AI applica ions may be add essed and e en
ou weighed by he bene i s o hei implemen a ion [14]
Regula o y Challenges and F amewo ks
A i icial in elligence in oncology p esen s unique egula o y challenges due o i s complexi y, adap i e
lea ning capabili ies, and in eg a ion in o clinical decision-making. Key issues include ensu ing sa e y,
e icacy, anspa ency, da a p i acy, and ongoing pe o mance moni o ing. Regula o y amewo ks mus
add ess algo i hm alida ion, bias mi iga ion, explainabili y, and pos -ma ke su eillance.
FDA Guidelines:
The U.S. Food and D ug Adminis a ion (FDA) egula es AI-based medical de ices and so wa e as a medical
de ice (SaMD) p ima ily h ough isk-based pa hways (510 k), De No o, P ema ke App o al). The FDA
emphasizes a “ o al p oduc li ecycle” app oach, equi ing obus p ema ke alida ion, Good Machine
Lea ning P ac ices (GMLP), and eal-wo ld pe o mance moni o ing. The FDA’s Digi al Heal h Cen e o
Excellence and AI/ML-Based SaMD Ac ion Plan p o ide guidance on anspa ency, algo i hm change
p o ocols, and pos -ma ke o e sigh . Fo oncology, he FDA has issued speci ic guidance on clinical ial
design and accele a ed app o al pa hways o AI-enabled he apeu ics and diagnos ics [15].
In e na ional Regula ions:
The Eu opean Union (EU) egula es AI in medical de ices unde he Medical De ice Regula ion (MDR) and
In Vi o Diagnos ic Regula ion (IVDR), wi h addi ional equi emen s o anspa ency, human o e sigh , and
isk managemen . The EU’s p oposed AI Ac in oduces a isk-based amewo k o all AI sys ems, including
Sou iane Kheli i Touhami, MAR Oncology and Hema ology (2025) 5:07.
Page 6 o 10
Sou iane Kheli i Touhami. (2025). Place o A i icial In elligence in Oncology: Li e a u e e iew and
Pe spec i es. MAR Oncology and Hema ology. (2025) 5:07
hose in heal hca e. O he ju isdic ions (e.g., Canada, Japan, Aus alia) a e de eloping o upda ing egula o y
amewo ks o add ess AI’s unique isks, o en emphasizing ha moniza ion wi h in e na ional s anda ds (e.g.,
IMDRF, ISO/IEC) [16].
Case S udies o A i icial In elligence in Oncology
A i icial in elligence has achie ed se e al success ul implemen a ions in oncology, pa icula ly in he
domains o cance de ec ion, diagnosis, p ognosis, and ea men planning:
A i icial In elligence in Imaging and Pa hology: Deep lea ning models, especially con olu ional neu al
ne wo ks (CNNs), ha e demons a ed high accu acy in de ec ing and classi ying umo s in adiology (e.g.,
mammog aphy, CT, MRI) and digi al pa hology slides. AI-based sys ems a e now FDA-app o ed o b eas
cance sc eening and lung nodule de ec ion, imp o ing sensi i i y and wo k low e iciency [17].
A i icial In elligence in Risk S a i ica ion and P ognosis: Machine lea ning algo i hms a e used o p edic
ecu ence isk and su i al in b eas , lung, and p os a e cance s, suppo ing pe sonalized ea men decisions.
Radio he apy Planning: AI-d i en au o-segmen a ion ools s eamline a ge and o gan-a - isk delinea ion,
educing in e -obse e a iabili y and planning ime.
Clinical Decision Suppo : AI models in eg a e mul i-omics, clinical, and imaging da a o ecommend
indi idualized he apies, including immuno he apy esponse p edic ion and molecula sub yping.
Clinical T ials: AI acili a es pa ien ma ching o oncology ials by apidly sc eening elec onic heal h
eco ds o eligibili y, inc easing en ollmen e iciency and ial di e si y.
These implemen a ions ha e led o imp o ed diagnos ic accu acy, wo k low e iciency, and mo e pe sonalized
ca e, hough ongoing alida ion and o e sigh emain essen ial [18].
Fu u e o A i icial In elligence in Oncology
Fu u e T ends in AI and Oncology: Eme ging Technologies and In eg a ion wi h O he Disciplines –
Sho Summa y
A i icial in elligence in oncology is apidly e ol ing, wi h se e al u u e ends poised o ans o m cance
ca e:
Eme ging Technologies:
Mul imodal AI: In eg a ion o da a om adiology, pa hology, genomics, and clinical eco ds enables
comp ehensi e umo cha ac e iza ion and mo e accu a e p edic ion o ea men esponse and ou comes.
Radiogenomics and Radiomics: AI-d i en analysis o imaging ea u es linked wi h molecula and gene ic
p o iles suppo s non-in asi e umo sub yping and pe sonalized he apy selec ion.
Sel -supe ised and Gene a i e Models: These ad anced machine lea ning app oaches can unco e no el
Sou iane Kheli i Touhami, MAR Oncology and Hema ology (2025) 5:07.
Page 7 o 10
Sou iane Kheli i Touhami. (2025). Place o A i icial In elligence in Oncology: Li e a u e e iew and
Pe spec i es. MAR Oncology and Hema ology. (2025) 5:07
bioma ke s, au oma e ea u e ex ac ion, and gene a e syn he ic da a o a e cance ypes, imp o ing model
obus ness.
Real- ime and Edge AI: Deploymen o AI models a he poin o ca e (e.g., in ope a ing ooms o adiology
sui es) allows o immedia e decision suppo and wo k low op imiza ion.
AI in D ug Disco e y: Machine lea ning accele a es a ge iden i ica ion, compound sc eening, and
bioma ke disco e y, s eamlining he de elopmen o new cance he apies [19].
In eg a ion o A i icial In elligence wi h O he Disciplines:
Bioin o ma ics and Mul iomics: AI is inc easingly used o analyze complex mul iomics da ase s (genomics,
ansc ip omics, p o eomics, me abolomics), ad ancing p ecision oncology and bioma ke disco e y.
Digi al Pa hology and Compu a ional Pa hology: AI enables high- h oughpu , ep oducible analysis o
his opa hology slides, suppo ing diagnos ic consis ency and no el insigh s in o umo biology.
In e disciplina y Collabo a ion: AI os e s collabo a ion be ween oncologis s, adiologis s, pa hologis s,
gene icis s, da a scien is s, and enginee s, d i ing inno a ion in ansla ional esea ch and clinical p ac ice.
Ca dio-oncology and O he Subspecial ies: AI is being applied o p edic and manage cance he apy-
ela ed oxici ies, such as ca dio oxici y, in eg a ing oncology wi h ca diology and o he special ies [20].
Key Challenges and Di ec ions:
Fu u e p og ess depends on add essing da a quali y, model in e p e abili y, egula o y ha moniza ion, and
equi able access o AI echnologies ac oss di e se heal hca e se ings.
Challenges and Limi a ions o A i icial In elligence in Oncology
AI in oncology o e s signi ican p omise bu aces no able challenges and limi a ions:
Technical Ba ie s:
Da a Quali y and He e ogenei y: AI models equi e la ge, high-quali y, and well-anno a ed da ase s.
Va iabili y in imaging p o ocols, elec onic heal h eco ds, and molecula da a can limi model gene alizabili y
and pe o mance.
Algo i hm Bias and Valida ion: Models may inhe i biases om aining da a, leading o educed accu acy
in unde ep esen ed popula ions. Ex e nal alida ion and obus es ing ac oss di e se coho s a e o en
lacking.
Explainabili y and T anspa ency: Many AI models, especially deep lea ning sys ems, unc ion as “black
boxes,” making i di icul o clinicians o in e p e o us hei ou pu s.
In eg a ion and In e ope abili y: Seamless in eg a ion wi h exis ing clinical wo k lows and elec onic
heal h eco d sys ems emains a challenge, o en equi ing signi ican IT in as uc u e and suppo .
Regula o y and Secu i y Conce ns: Ensu ing ongoing sa e y, e icacy, and da a p i acy is complex,
Sou iane Kheli i Touhami, MAR Oncology and Hema ology (2025) 5:07.
Page 8 o 10
Sou iane Kheli i Touhami. (2025). Place o A i icial In elligence in Oncology: Li e a u e e iew and
Pe spec i es. MAR Oncology and Hema ology. (2025) 5:07
especially o adap i e algo i hms ha e ol e o e ime [21].
Accep ance by Heal hca e P o essionals:
T us and Skep icism: Clinicians may be eluc an o ely on AI due o conce ns abou eliabili y, loss o
clinical au onomy, and lack o unde s anding o AI mechanisms.
T aining and Educa ion: Limi ed exposu e o AI concep s and insu icien aining impede adop ion and
e ec i e use.
Wo k low Dis up ion: Pe cei ed o eal inc eases in wo kload, wo k low changes, and lack o use - iendly
in e aces can hinde accep ance.
E hical and Legal Conce ns: Unclea accoun abili y, medicolegal isks, and e hical dilemmas ega ding AI-
d i en decisions con ibu e o hesi ancy [22].
Conclusion
A i icial in elligence is apidly ans o ming oncology by imp o ing cance de ec ion, diagnosis, isk
s a i ica ion, and ea men pe sonaliza ion. AI-d i en ools enhance he accu acy o imaging and pa hology
in e p e a ion, suppo p ecision medicine h ough in eg a ion o complex da a, and s eamline clinical
wo k lows. While A i icial In elligence shows p omise in op imizing sc eening and decision-making,
challenges emain ega ding da a quali y, model anspa ency, and clinical in eg a ion. Con inued esea ch,
alida ion, and mul idisciplina y collabo a ion a e essen ial o ensu e sa e, e ec i e, and equi able adop ion
o A i icial In elligence in cance ca e.
Re e ences
1. Chen MM, Te zic A, Becke AS, Johnson JM. "A i icial in elligence in oncologic imaging." Eu opean
Jou nal o Radiology Open. 2022;9:100421.
2. Moxley-Wyles B, Colling R, Ve ill C. A i icial in elligence in pa hology: an o e iew. Diagnos ic
His opa hology. 2020;26(9):397-403.
3. Abdelsamea MM, Zidan U, Senousy Z, e al. A su ey on a i icial in elligence in his opa hology image
analysis. Wiley In e disciplina y Re iews: Da a Mining and Knowledge Disco e y. 2022;12(2):e1474.
4. Kalsi S, F ench H, Chhaya S, e al. The e ol ing ole o a i icial in elligence in adio he apy ea men
planning—A li e a u e e iew. Clinical Oncology. 2024;36(6):e393-e404.
5. Sheng Y, Zhang J, Ge Y, e al. A i icial in elligence applica ions in in ensi y modula ed adia ion
ea men planning: an o e iew. Quan i a i e Imaging in Medicine and Su ge y. 2021;11(12):4859-4876.
6. Kou ou K, Exa chos KP, Papaloukas C, e al. Applied machine lea ning in cance esea ch: a
sys ema ic e iew o pa ien diagnosis, classi ica ion and p ognosis. Compu S uc Bio echnol J.
Sou iane Kheli i Touhami, MAR Oncology and Hema ology (2025) 5:07.
Page 9 o 10
Sou iane Kheli i Touhami. (2025). Place o A i icial In elligence in Oncology: Li e a u e e iew and
Pe spec i es. MAR Oncology and Hema ology. (2025) 5:07
2021;19:5546-5555.
7. Ja e D, S ide E, Vallis KA, Gooding MJ. Applica ions and limi a ions o machine lea ning in
adia ion oncology. B J Radiol. 2019;92(1100):20190001.
8. Yim W, Ye isgen M, Ha is WP, Kwan SW. Na u al language p ocessing in oncology: a e iew. JAMA
Oncology. 2016;2(6):797-804.
9. Zeng J, Bane jee I, Hen y AS, e al. Na u al language p ocessing o iden i y cance ea men s wi h
elec onic medical eco ds. JCO Clinical Cance In o ma ics. 2021;5:1102-1111.
10. Pa ikh RB, Obe meye Z, Na a he AS. Using big da a and p edic i e analy ics o de e mine pa ien
isk in oncology. ASCO Educa ional Book. 2019;39:53-61.
11. Askin S, Bu khal e D, Calado G, e al. A i icial in elligence applied o clinical ials: oppo uni ies
and challenges. Heal h and Technology. 2023;13:1001–1012.
12. Chow R, Mid oni J, Kau J, e al. Use o a i icial in elligence o cance clinical ial en ollmen : a
sys ema ic e iew and me a-analysis. JNCI: Jou nal o he Na ional Cance Ins i u e. 2023;115(4):365–375.
13. Ri e a SC, Liu X, Chan AW, e al. Guidelines o clinical ial p o ocols o in e en ions in ol ing
a i icial in elligence: he SPIRIT-AI ex ension. The Lance Digi al Heal h. 2020;2(10):e549–e560.
14. Han el, A., P. Walsh, T., M. Ma on, J., L. Kehl, K., Sha p, R., Van Allen, E., and A. Abel, G.
"Pe spec i es o Oncologis s on he E hical Implica ions o Using A i icial In elligence o Cance Ca e."
(2024). ncbi.nlm.nih.go
15. Wa aich HJ, Tazbaz T, Cali RM. FDA pe spec i e on he egula ion o a i icial in elligence in
heal h ca e and biomedicine. JAMA. 2025;334(3):245-252.
16. Wa aich HJ, Tazbaz T, Cali RM. FDA pe spec i e on he egula ion o a i icial in elligence in
heal h ca e and biomedicine. JAMA. 2025;334(3):245-252.
17. Romagnoli A, Fe a a F, Langella R, Zo i A. Heal hca e sys ems and a i icial in elligence: ocus on
challenges and he in e na ional egula o y amewo k. Pha m Res. 2024;41(5):e03685.
18. Bi WL, Hosny A, Schaba h MB, e al. A i icial in elligence in cance imaging: clinical challenges and
applica ions. CA Cance J Clin. 2019;69(2):127-157.
19. Ja e D, S ide E, Vallis KA, Gooding MJ. Applica ions and limi a ions o machine lea ning in
adia ion oncology. B J Radiol. 2019;92(1100):20190001.
20. Xu J, Yang P, Xue S, e al. T ansla ing cance genomics in o p ecision medicine wi h a i icial
in elligence: applica ions, challenges and u u e pe spec i es. Hum Gene . 2019;138(2):109-124.
21. Wang X, Guo J, Gu D, e al. T acking knowledge e olu ion, ho spo s and u u e di ec ions o eme ging
echnologies in cance s esea ch: a bibliome ics e iew. J Cance . 2019;10(12):2643-2653.
22. Chua IS, Gaziel-Yablowi z M, Ko ach ZT, e al. A i icial in elligence in oncology: pa h o
