I Am Not a Patient: Design Challenges

in Self-Screening Medical Devices

Abstract
Personal health informatics refers to a type of medical
systems and devices intended to be used by users
without professional medical training. They play an
important supporting role for healthy decision-making
in non-clinical environments. There are two important
categories of personal health informatics: self-
monitoring and self-screening. Self-monitoring devices
help users better manage illness and maintain wellness
through surveillance of health parameters, which has
been widely studied in recent years. However, self-
screening medical devices, of which the users are
unaware of their potential illness, are less understood.
This paper presents the first results from an
investigation into the difference between self-
monitoring devices and self-screening devices. Based
on a skin cancer self-screening mobile system, we
conduct semi-structured interviews to explore the
challenges regarding acceptance and adoption. Future
work that focuses on addressing the design challenges
evolved are also introduced.
Author Keywords
Personal health informatics; self-screening; acceptance
and adoption; interviews;
ACM Classification Keywords
H.5.m. Information interfaces and presentation (e.g.,
HCI): Miscellaneous.




Permission to make digital or hard copies of part or all of this work for
personal or classroom use is granted without fee provided that copies are
not made or distributed for profit or commercial advantage and that
copies bear this notice and the full citation on the first page. Copyrights
for third-party components of this work must be honored. For all other
uses, contact the Owner/Author.

Copyright is held by the owner/author(s).
Chen Zhao
SUTD-MIT International Design
Centre
Singapore University of
Technology and Design
[email protected]

Ricardo Sosa
Engineering Product Development
Singapore University of
Technology and Design
[email protected]

Ngai-Man Cheung
Information Systems
Singapore University of
Technology and Design
[email protected]

Dawn Chin-Ing Koh
Engineering Product Development
Singapore University of
Technology and Design
[email protected]


Introduction
As life expectancy rises, the prevalence of health
problems is expected to increase [8]. With this, the
cost of supporting the related healthcare ecosystems
would become a serious social burden. Personal health
informatics systems have the potential to help mitigate
this financial and social tension [5]. Through presenting
analytic interpretations of users health data, such
systems enable users to detect and interpret early
disease symptoms, be aware of their health condition
and support healthy behavior changes by linking to
long-term health objectives.
While researchers have paid increasingly attention to
both patients and doctors needs [2] to identify the
design requirements of personal health informatics
system, one type of such systems are less studied so
far- the systems that assist users to perform self-
screening. The users of these systems are not identified
as patients yet. This user category presents unique
acceptance and adoption challenges compared with
users of other personal health informatics devices. For
instance, it is not clear how to incentivize usage, how
to best present results, and how to sustain habit
formation for periodic long-term use.
captured
skin lesion image
Image
Processing
Image
Analysis
Risk Score
Estimation
Assessment
and Suggestion

Figure 1. System framework of the eSkinCheck
Self-monitoring Vs. Self-screening
Self-monitoring devices collect physiological parameters
through wearable or non-wearable sensors or
journaling forms [7]. The users defined in this context
tend to have received professional medical diagnosis,
being categorized as patients or at least are aware of
the problem (e.g., obesity or diabetics). By providing
this type of users own personal health information,
researchers aim to empower them to take an active
role in their health management [6]. In contrast, self-
screening devices work as an accessible pre-screening
method to help users to identify potential health
problems in the early stages of the diseases. Before
being formally diagnosed with a particular health
problem, the users have different considerations and
needs compared with patients. In addition, although
self-screening devices requires a routinely data
acquisition like self-monitoring devices, a large number
of them are usually designed without a long-term
health objective which could encourage healthy
behavior changes. Instead, these systems typically
provide a yes-no suggestion or risk assessment of
specific diseases. We argue that lacking long-term
health objective may also cause difficulty to sustain
usage of the devices. An example of self-screening
device from our prior work is briefly introduced below.
eSkinCheck System
Skin cancer is one of the most common cancers in
Caucasians. Due to the lack of patients awareness, the
presence of disease is diagnosed by doctors only at a
late stage, usually fatal by then [4]. Aiming to assist
people to diagnose the risk of skin cancer in a low cost
and timely, we develop eSkinCheck system [9], a
mobile-based self-screening application for providing
common users early detection of skin cancer.

To carry out the self-screening, users need to capture
the skin lesion area using eSkinCheck. Then the system
use image analysis to compute features that capture
the morphological and chromatic characteristics of the
skin mole. Based on these features, the system uses
machine learning based algorithms to estimate the
malignancy and to provide suggestion (Figure 1).
Acceptance and Adoption Challenges
To explore the design issues of self-screening systems,
we conducted semi-structure interviews with eight
older users (over 65 years old) using the eSkinCheck
system as an example. We illustrate the interview
insights based on the acceptance and adoption of the
technology.
Acceptance
Users raised questions about potential harmfulness as a
first concern despite our system being implemented
on a smartphone. Older adults may associate these
systems with the hospital diagnostic machines and
question about whether risky detection methods (e.g.,
X-Ray) would be used, even among the older adults
who had used smartphone for a long period.
The older adults also expressed high system
acceptance if expertise of the diagnosis could be
verified. Some interviewees expressed they may start
to use the system only after receiving the
recommendations from their peers.
A key viewpoint expressed by some interviewees is that
they intend to keep happy without being aware of any
potential disease, even if this would heavily threaten
their health in the future. The designers of self-
screening systems should investigate how to turn the
sense of negative disease detection into the notion of
positive health maintenance.
Adoption
In order to detect the potential risk of disease timely,
the self-screening device requires routinely usage. Thus,
we also explore the long-term adoption of the device in
the interviews. Apart from the easiness to use, older
adults expect the results of the diagnosis to be easy to
understand. The detailed analysis of personal health
data was considered confusing and complex.
Without a specific long-term goal like self-monitoring
device, the motivation of sustained usage is a key
challenge. Most of the interviewees reported they would
not continue to use after received negative results
several times. However, they still expressed the need
for regular formal medical check-ups. Thus, the
reminder of self-screening may help routinely data
acquisition, especially in the case that the reminder was
from a professional, as some interviewees suggested.
It would be helpful to keep the data acquisition in a
routine if the process is performed by family members
or potential care givers. However, the older adults
indicated that they only intend to trouble the family
members if they have a good relationship.
Future Work and Discussion
Our work suggests that a novel type of personal health
informatics systems need further understanding of
users needs and perceptions. From our interviews, we
identify three unique issues of self-screening devices:
the importance of less perceived harmfulness, the
difficulty of sustaining usage without a long-term goal
and the lacking of intention to learn negative personal

health information. We acknowledge that the culture
difference may exist towards the system usage.
However, we believe that the difference is more subtle
than argued regarding the issues we have explored.
Our future work will further design our current
eSkinCheck system by embracing the end users
considerations and conducting field studies. More
specifically, we will consider:
Belief - Previous research reports that acceptance of
medical ICT devices is strongly motivated by perceived
usefulness [3]. Moreover, in our context, we aim to
design self-screening systems with low perceived
harmfulness and high perceived expertise.
Beyond Yes or No - Arnrich et al. suggest shifting the
reflective disease management to a more pervasive and
assistive focus [1]. Future work will investigate how to
embed the pervasive information with a constructive
health objective into the simple diagnosis information,
thereby improving sustained motivation.
Positive Perspective - Users should not be scared by
the perceived negative disease-diagnosing behavior.
Future design will focus on how to present disease
diagnosis process in a positive health maintenance way.
Reminder from Doctors- From our interviews, reminder
helps keeping the motivation of sustained usage,
especially if the reminder message is designed in the
form of professional doctors narratives.
Acknowledgements
The material reported in this document is supported by
the SUTD-MIT International Design Centre (IDC). Any
findings, conclusions, recommendations, or opinions
expressed in this document are those of the author(s)
and do not necessary reflect the views of the IDC.
References
[1] Arnrich, B., et al. Pervasive healthcare: paving the
way for a pervasive, user-centered and preventive
healthcare model. Methods of information in medicine,
49(2010), 67.
[2] Julie, M. and M. Chalmers. Self-monitoring, self-
awareness, and self-determination in cardiac
rehabilitation. Proc. CHI 2010, (2010), 12131222.
[3] Heart, T. and Kalderon, E. Older adults: Are they
ready to adopt health-related ICT? International Journal
of Medical Informatics, 2013, 82(11), 209-231.
[4] Hauryueh, L., et al., Melanoma: Differences
between Asian and Caucasian Patients, Ann Acad Med
Singapore 2012; 41: 17-20.
[5] Michael, M., et al., ICT-based health information
services for elderly people: Past experiences, current
trends, and future strategies, Medical Informatics & the
Internet in Medicine 32 (4) (2007) 251261.
[6] Steele, R. and Lo, A. Future Personal Health
Records as a Foundation for Computational Health.
ICCSA 2009, Springer (2009), 719-733.
[7] Tentori, M., Hayes, G.R., and Reddy, M. Pervasive
Computing for Hospital, Chronic, and Preventive Care.
Found. Trends Hum.-Comput. Interact., 5(2012), 1-95.
[8] CNN our mobile society, Your phone will know
you're sick before you do.
http://edition.cnn.com/2013/02/26/tech/opinion-
health-mobile-curtis/
[9] Thanh-Toan Do et al., "Design of a Mobile Imaging
System for Early Diagnosis of Skin Cancer," IEEE Life
Sciences Grand Challenges Conference 2013.