Towards Flexible Mobile Data Collection in Healthcare
Johannes Schobel, R¨
udiger Pryss, Marc Schickler, Manfred Reichert
Institute of Databases and Information Systems
Ulm University, Germany
{johannes.schobel, ruediger.pryss, marc.schickler, manfred.reichert}@uni-ulm.de
Abstract—The widespread dissemination of smart mobile
devices offers promising perspectives for a variety of healthcare
data collection scenarios. Usually, the implementation of mobile
healthcare applications for collecting patient data is cumber-
some and time-consuming due to scenario-specific requirements
as well as continuous adaptations to already existing mobile
applications. Emerging approaches, therefore, aim to empower
domain experts to create mobile data collection applications
themselves. This paper discusses flexibility issues considered
by a generic and sophisticated framework for realizing mobile
data collection applications. Thereby, flexibility is discussed
along different phases of data collection scenarios. Altogether,
the realized flexibility significantly increases the practical
benefit of smart mobile devices in healthcare data collection
scenarios.
Keywords-Mobile Data Collection, Process-driven Data Col-
lection, Data Collection Flexibility.
I. INTRODUCTION
Using smart mobile devices (e.g., tablets and smart-
phones) in healthcare scenarios has become increasingly
important during the last years. To overcome the draw-
backs of traditional paper-based instruments (e.g., question-
naires), mobile applications specifically tailored to a partic-
ular healthcare scenario shall enable experts from various
domains (e.g., healthcare or psychology) to collect patient
data more effectively. Based on the experiences gathered in
various projects realizing mobile data collection applications
in the large scale, a number of crucial requirements has
been identified. For example, sophisticated operations1for
navigating within user forms were often demanded. In this
context, a flexible customization of the user interface was
crucial, ranging from issues related to multilingualism up to
individual control elements not introduced before2.
The mentioned projects further revealed that mobile data
collection applications frequently need to be adapted to
changing requirements. For example, changes became neces-
sary due to the rapid development cycles of mobile operating
systems (i.e., Android) and emerging demands for new
features (e.g., integrating questionnaires with sensors to
measure and record vital parameters of the patients). In this
context, IT experts needed to be involved to properly address
these requirements, resulting in a cost explosion in most
1Skip questions depending on already given answers.
2Psychological questionnaires may need a neutral position for certain
controls (e.g., slider) not influencing the participant in any way.
projects. What domain experts actually demanded, how-
ever, are tools enabling them to create sophisticated mobile
data collection applications themselves. For this purpose, a
comprehensive technology framework has been developed,
which supports the design,deployment and execution of mo-
bile data collection applications as well as the analysis and
archiving of the data collected. In particular, the framework
enables customizable mobile applications by allowing for
flexible adaptations of the mobile data collection application
during all these phases. Note that this flexibility reduces
development costs significantly, while at the same time it
improves the time to deliver a respective mobile application.
This paper envisions how the developed framework supports
flexible adaptations in healthcare data collection scenarios.
Section II introduces technologies enabling flexibility in
mobile data collection projects. Section III discusses related
work and Section IV concludes the paper.
II. FLEXIBILITY GOALS
When realizing mobile data collection applications, flex-
ibility issues are considered for three phases (cf. Fig. 1):
(1) Design Time. Flexibility is provided through an ad-
vanced configurator component and end-user programming.
The configurator allows creating all elements needed to
design a data collection instrument (e.g., texts or questions).
The latter may be provided in different languages to enable
multilingualism. Through a model-based approach domain
experts are enabled to easily define the logic and structure of
the data collection instrument themselves. Advanced wizards
guide domain experts through the process of defining navi-
gation paths for instruments. Finally, sensors for collecting
data during run time are modeled on an abstract level.
(2) Deployment Time. The designed data collection
instrument is mapped to a process model [1] and is installed
on respective smart mobile devices. Process management
technology, in turn, is utilized to ensure a flexible and robust
execution of the process instance during data collection.
(3) Run Time flexibility requires the consideration of
several issues. First, a (lightweight) process engine for the
robust execution of the process model defining the logic
of the instrument is run on the smart mobile device. This
engine is capable of enacting and interpreting the underlying
process model. The logic to be executed during one specific
process node (e.g., displaying a form or retrieving data from
$OFRKRO
&RQVXPSWLRQ
&LJDUHWWH
&RQVXPSWLRQ
6WDUW)ORZ
$FWLYLW\ ;25MRLQ
'DWD(OHPHQW
:ULWH$FFHVV
5HDG$FFHVV
(QG)ORZ
(7B&RQWURO)ORZB'HIDXOW
(7B'DWD)ORZ
$OFRKRO&LJDUHWWHV
;ŝŐĂƌĞƚƚĞƐсLJĞƐͿ
E;ůĐŽŚŽůсLJĞƐͿ
;25VSOLW
ĞůƐĞ
;ŝŐĂƌĞƚƚĞƐсLJĞƐͿ
E;ůĐŽŚŽůсŶŽͿ
(7B&RQWURO)ORZ
&LJDUHWWHV
$OFRKRO
3DJH
,QWUR
3DJH
*HQHUDO (QG&LJDUHWWHV
DŽĚĞůĂƚĂŽůůĞĐƚŝŽŶ
/ŶƐƚƌƵŵĞŶƚhƐŝŶŐŶĚͲhƐĞƌ
WƌŽŐƌĂŵŵŝŶŐƉƉƌŽĂĐŚĞƐ
WƌŽĐĞƐƐdĞĐŚŶŽůŽŐLJ
;ĞŐWƌŽĐĞƐƐDŽĚĞůͿ
ϭͿĞƐŝŐŶdŝŵĞ ϯͿZƵŶdŝŵĞϮͿĞƉůŽLJŵĞŶƚdŝŵĞ
DŽďŝůĞĂƚĂŽůůĞĐƚŝŽŶƉƉůŝĐĂƚŝŽŶ
^ĞŶƐŽƌƐ;džƚĞƌŶĂů/ŶƚĞƌŶĂů,ĂƌĚǁĂƌĞͿ
hƐĞƌ/ŶƚĞƌĨĂĐĞ
^ĞŶƐŽƌ&ƌĂŵĞǁŽƌŬ
WƌŽĐĞƐƐŶŐŝŶĞ
DŽĚƵůĞϭ
h/
>ŽŐŝĐ
DŽĚƵůĞϮ
h/
>ŽŐŝĐ
DŽĚƵůĞŶ
h/
>ŽŐŝĐ
Z^d
ŽƌĞ
ŶĂůLJƐŝƐ
>ŝŐŚƚǁĞŝŐŚƚWƌŽĐĞƐƐŶŐŝŶĞĨŽƌ
WƌŽĐĞƐƐdžĞƵĐƚŝŽŶĂŶĚDŽŶŝƚŽƌŝŶŐ
ǀĂůƵĂƚŝŽŶ
džĞĐƵƚŝŽŶ DŽŶŝƚŽƌŝŶŐ
ĂƚĂďĂƐĞΘKZD
^ĞŶƐŽƌ&ƌĂŵĞǁŽƌŬdŽ
/ŶƚĞŐƌĂƚĞ,ĂƌĚǁĂƌĞ
ƵƐƚŽŵŝnjĞĚdžĞĐƵƚĂďůĞ
ŽŵƉŽŶĞŶƚƐ
h/'ĞŶĞƌĂƚŽƌǁŝƚŚ
ƵƐƚŽŵŽŶƚƌŽůůĞŵĞŶƚƐ
Figure 1. Providing Flexibility in the Lifecycle of Mobile Data Collection Applications
sensors) as well as the corresponding UI is realized using ex-
ecutable components. The Execution module of the process
engine dynamically loads the components needed in order
to adjust functionality depending on the application scenario
during run time. Second, an advanced UI generator combines
all UI fragments created by the executable components and
displays them to the user. In particular, platform-specific
guidelines needed to be considered as well. Furthermore,
custom control elements must be implemented to meet
scenario-specific requirements. Third, a sophisticated sensor
framework allows connecting both external and internal
sensors (e.g., camera, pulse-measuring devices) with the
smart mobile application.
The combined use of the various technologies enables
flexibility for all phases of mobile data collection applica-
tions (cf. Table I). More specifically, they foster the con-
tinuous development of mobile data collection applications
as new releases can be easily created by domain experts
themselves without the involvement of IT experts.
III. RELATED WORK
An approach supporting researchers in collecting data
with smart mobile devices is presented in [2]. The platform,
however, is specifically tailored to mental health research
End-User Programming
Process Model
Process Engine
Executable Components
Sensor Framework
Model-Based Approach • •
Complex Navigation • • • •
Different Releases • •
Flexible Execution • • •
Monitoring & Analysis • • •
Sensors • •
Multilingualism • •
UI Generator •
Evolution • • • • •
Table I
COMBINED TECHNIQUES ENABLING FLEXIBILITY
and cannot be simply adjusted to other domains. Despite use-
ful features, like interval-based questionnaires, the approach
does not allow for an automatic UI generation, which is
indispensable for any flexible approach. Another framework
for mobile data collection is presented in [3]. It provides
a limited configurator component allowing users to create
data collection applications. This approach, however, lacks
flexibility concerning user navigation and sensor integration.
IV. SUMMARY AND OUTLOOK
Smart mobile devices offer promising perspectives for col-
lecting high quality data in healthcare scenarios. In particu-
lar, domain experts shall be empowered to create mobile data
collection applications themselves. This paper introduced a
framework, providing flexibility along different phases of the
lifecycle of mobile data collection applications. The proof-
of-concept prototypes indicate promising results combining
end-user programming with process management technology
to enhance flexibility in mobile data collection applications.
Still further research is needed to foster usability of the
developed components. Therefore, a study is planned to eval-
uate the approach as well as to measure mental efforts spent
when designing such mobile data collection applications.
Finally, techniques providing flexibility for other phases of
the lifecycle (e.g., evaluation of the data collected) need to
be introduced and evaluated.
REFERENCES
[1] J. Schobel, M. Schickler, R. Pryss, and M. Reichert, “Process-
driven data collection with smart mobile devices,” in 10th Int’l
Conf on Web Information Systems and Technologies (Revised
Selected Papers), ser. LNBIP. Springer, 2015, no. 226.
[2] A. Gaggioli, G. Pioggia, G. Tartarisco, G. Baldus, D. Corda,
P. Cipresso, and G. Riva, “A mobile data collection platform
for mental health research,” Personal and Ubiquitous Comput-
ing, vol. 17, no. 2, 2013.
[3] S. Kim, J. Mankoff, and E. Paulos, “Sensr: Evaluating a Flex-
ible Framework for Authoring Mobile Data-Collection Tools
for Citizen Science,” in Proceedings of the 2013 Conference
on Computer Supported Cooperative Work. ACM, 2013.
