OzCHI 2017, Nov 28 - Dec 1, Brisbane, Australia

Human - Nature Short Papers

MyFootCare: A Mobile Self-tracking Tool to Promote

Self-care Amongst People with Diabetic Foot Ulcers
Ross Brown, Bernd Ploderer, Leonard Si Da Seng, Peter Lazzarini, Jaap van Netten
Queensland University of Technology (QUT)
Brisbane, Australia
[email protected], [email protected], [email protected], [email protected],
[email protected]
ABSTRACT There is a growing interest in HCI in self-tracking to support
positive health outcomes. Several studies explored how people
We present the design of MyFootCare, a mobile app to support collect, reflect, and act on step data from wearable fitness
people with diabetic foot ulcers in their self-care. Self-care is a trackers (e.g., Fitbit, Garmin), showing that people derive useful
critical component of care for people with a diabetic foot ulcer as information and motivation from these data [5, 13]. Many fitness
most of their ulcer care is provided away from the clinic. To trackers also provide insights into how long and deep we sleep,
promote better self-care, we designed a mobile application which can help individuals to make adjustments to their
‘MyFootCare’ that harnesses visual analytics and self-report to environment and lifestyle factors that may impact their sleep
provide feedback about the healing process. MyFootCare [15, 19].
encourages people to take a photo of their ulcer with their More recently, HCI researchers have started to explore self-
mobile phone each time they change their wound dressing. tracking in clinical contexts where patients and healthcare
Based on computer vision techniques, users receive graphical professionals collaboratively examine the data. For example,
feedback on changes in ulcer size over time to objectively track Mentis et al., [16] examined collaboration between patients with
the healing progress. Additionally, MyFootCare seeks to foster Parkinson’s disease and their clinicians around physical activity
self-care through personal goals, diaries, and reminders to enact data from Fitbit devices. Ploderer et al., [17] explored
care. Feedback from three people with chronic ulcers shows that opportunities for stroke survivors and their therapists based on
the app builds on existing practices of taking wound photos and upper limb movement data. Finally, Smith et al. [22] explored
that it is seen as useful to track progress and to facilitate how people with diabetes and their doctors discuss diet and
dialogue with clinicians. More work is underway to evaluate the glucose levels tracked in daily life.
use of MyFootCare in a deeper field study. This research builds on the growing body of self-tracking
technologies in medical settings. In particular, our aim is to
CCS CONCEPTS develop self-tracking technologies that support people with
• Human-centered computing → Ubiquitous and mobile diabetes in the care of their diabetic foot ulcer (DFU). DFUs are a
computing; significant complication of diabetes mellitus and without good
quality care lead to lower-extremity amputation. The most
frequent underlying cause is loss of sensation from diabetes in
KEYWORDS combination with unchecked foot trauma [7]. Many people with
Self-monitoring, Personal Informatics, Diabetic Foot Ulcer, diabetes are already familiar with self-tracking, for example via
1
Mobile Devices diet monitoring, to proactively manage their diabetes outcomes
[22]. This work seeks to build on and support existing practices
ACM Reference format: to encourage self-management of wound-care.
R. Brown, B. Ploderer, L. Si Da Seng, P. Lazzarini, J. van Netten. In this paper, we present the design and early evaluation of
2017. MyFootCare: A Mobile Self-tracking Tool to Promote Self- ‘MyFootCare’, a self-tracking app to encourage self-care
care Amongst People with Diabetic Foot Ulcers. In Proceedings of amongst people with DFU. The design has been inspired by
the 29th Australian Conference on Human-Computer Interaction, theories of behaviour change [4], which suggest that people need
Brisbane, QLD, Australia, November 2017 (OzCHI 2017), 5 pages. the ability, motivation, and triggers to enact a desirable
https://doi.org/10.1145/3152771.3156158 behaviour. Early pilot feedback from three people with a DFU
shows that users consider MyFootCare easy to use. They have an
1 INTRODUCTION existing interest in using wound photos, appreciate seeing
healing progress, and they want to share results with clinicians.

Permission to make digital or hard copies of all or part of this work for personal or
classroom use is granted without fee provided that copies are not made or
2 CHALLENGES OF SELF-CARE IN DFU
distributed for profit or commercial advantage and that copies bear this notice and Each day in Australia 50,000 people suffer with a DFU, 1,000 are
the full citation on the first page. Copyrights for components of this work owned
by others than ACM must be honored. Abstracting with credit is permitted. To
hospitalised, 12 have an amputation, and four die because of a
copy otherwise, or republish, to post on servers or to redistribute to lists, requires DFU; at an estimated annual cost of $1 billion [11-13]. DFUs take
prior specific permission and/or a fee. Request permissions from a toll on the physical and mental quality of life of patients,
[email protected]. OzCHI '17, November 28-December 1, 2017, Brisbane, QLD, partners, family, workplaces and their communities [11].
Australia © 2017 Association for Computing Machinery. ACM ISBN 978-1-4503-
5379-3/17/11…$15.00 https://doi.org/10.1145/3152771.3156158 Queensland has implemented best practice clinical care to
patients with DFU and demonstrated 40% reductions in

462
 OzCHI 2017, Nov 28 - Dec 1, Brisbane, Australia
Short Papers Human - Nature

hospitalisation and amputation rates across the state [12].

However, to achieve further reductions in the large DFU burden,
experts suggest new strategies that engage patients in self-care
[1].
There are seven key elements [20], that support DFU clinical
treatment: "1. relief of pressure and protection of the ulcer; 2.
restoration of skin perfusion; 3. treatment of infection; 4.
metabolic control and treatment of co-morbidity; 5. local wound
care; 6. education for patient and relatives; and 7. prevention of
recurrence."
All these elements are only fully optimized with patient
engagement in self-care practices. Thus, engaging patients away
from clinics remains a global challenge [20]. To date, providing
patients with education and treatment information has not
improved DFU healing outcomes [3]. Experts instead suggest
simpler actionable tasks need to be provided for the user to
support self-care [18].
Figure 1: Fogg’s behaviour model for persuasive
DFU treatment experts have recommended using pictures
technology [4].
wherever possible as a motivational tool and catalyst for
improved communication [18]. Most recently, however, van
Netten et al. [23] reported that digital images alone were not 1. Trigger - prompt the user to action on a regular basis
valid or reliable for monitoring DFU outcomes and towards success. MyFootcare provides regular
recommended the need to augment digital images with thermal reminders to write in the application diary.
images, self-reported symptom diaries and computerised 2. Motivation - ensure compliance with helpful objective
machine learning algorithms. However, none of this previous actions via motivational feedback. ‘Live’ objective
work has, to our knowledge, integrated motivational self- wound tracking is used to support user motivation by
tracking, data visualisation, image capture and image processing visualising to the user their progress in healing the
to provide multi-faceted support for DFU self-care. wound.
3. Ability - promote adherence by provision of intuitive
software systems to support ease of use. A foot image
3 THEORIES GUIDING THE DESIGN OF capture mechanism has been developed that supports
MYFOOTCARE easier image capture.
This project aims to empower people with DFU in their self-care 4. Using the above elements, we have derived application
through a novel mobile application called ‘MyFootCare’. functionality to encourage self-collection of relevant
MyFootCare promotes self-care through a focus on personal treatment care data to improve DFU self-care
goals, reminders to enact care and novel visual analytics, outcomes.
including digital images and self-reported symptoms, to allow
patients and their clinicians to track their healing progress. 4 MYFOOTCARE PROTOTYPE
From self-tracking theory, we derive key components to drive Overall the aim is to provide a diary-based approach to enhance
the user requirements to meet motivational theory variables. We patient-clinician communication with self-tracking data.
observe from [14, 4] the need to support people in collecting Another aim is to provide visual analytics technology to monitor
information for self-reflection, motivation and future action. The progress in wound healing: MyFootCare applies computer vision
major focus is to provide mobile software technology that techniques to extract information from digital colour images
supports this process, especially the collection phase. We note taken through the app, when the user changes their wound
that the acquisition of DFU care skills are considered out of dressing.
scope for this research and we instead focus on objective data A prototype application was developed for Android devices
gathering and motivation. using Java frameworks and OpenCV (www.opencv.org) for
From Fogg’s behaviour model [4] (see Fig. 1) we see that wound segmentation and analysis. Our prototype was
successful behaviour change is derived from two variables, viz. developed on a Samsung Galaxy S4 mobile phone.
ability and motivation. That is, easy to do tasks do not require From engagement with podiatrists and previously described
much motivation, compared to harder tasks. Enhancing ability persuasive technology theory models, the following features
(making tasks easy to do) and motivation (providing useful have been placed into MyFootcare. A mapping to the motivation
feedback) along with judicious use of triggers, should support model factor and example figure reference is included in
the user in transition to a successful set of self-care behaviours. parentheses:
Using Fogg's model, we enumerate the phases required to
produce behaviour change in DFU self-care: 1. Goal Setting Picture (Motivation, Figure 2.a) – the
home screen shows an image captured by the user to
visualise the goal they wish to achieve when fully fit
(e.g. an image of forest shows a place they wish to
walk when healthy).

463
OzCHI 2017, Nov 28 - Dec 1, Brisbane, Australia
Human - Nature Short Papers

We conducted a pilot qualitative study with three people with a

DFU to evaluate MyFootCare and improve its utility. The
evaluation occurred in parallel with the software development
based on a high-fidelity prototype of the application, which
allowed participants to click through screen mock-ups of all
previously mentioned features. While not implemented, image
capture concepts were presented for evaluation by the
particpants, giving insight into possible future implementations.
The aims of this evaluation were to (1) obtain feedback from
people with a DFU to see if the application has the potential to
engage them in the care of their ulcer, and (2) to gain feedback
on the utility of the application, to improve its design and
functionality.

Figure 2: (a) Goal Setting Picture (App Feature 1.) on

the home screen to visualise what the user wants to
achieve; (b) Image Capture with voice feedback (App
Feature 2.) to centre the foot over the camera.
2. Image Capture (Ability, Figure 2.b) – it is difficult to
take photos of the foot without assistance from
another person. Hence, we devised a mechanism to
allow individuals to capture foot images by placing the
phone on the floor and hovering the foot over it. The
app guides the user through voice feedback (based on
OpenCV image analysis) to centre the foot over the
camera at an appropriate distance, and to take a photo.
3. Wound Analysis (Ability, Figure 3.) - an easy to use
wound analysis method was implemented using
OpenCV that automatically captures the size of the
wound after segmentation from the image by drawing Figure 4: (a) Diary Reminders (left) - showing emotional
inside and outside the affected area. mood and of
The aims text
thisentry (App were
evaluation Feature
to (1)4.)obtain
initiated by
feedback
4. Diary Reminders (Trigger, Figure 4.a) - daily reminders reminders.
from (b) Wound
people with a DFU Size
to seeTracking
if the app(right)
has the - showing
potential
to prompt users to record daily actions and mood values
to engagedrawn
themfrom
in thewound analysis
care of their asand
ulcer, a graph (App
(2) to gain
levels. Feature
The aims5.).
of this evaluation were to (1) obtain feedback
We recruited three people with a DFU who received foot
therapy at the Chermside Diabetic Foot Service (Metro North
Hospital and Health Service, Brisbane, Australia). All
participants were male and 44, 56, and 77 years old. All three
participants had type 2 diabetes and foot ulcers for extended
periods: participant 1 had recurrent ulcers over the last 13 years
(and 6 toes amputated as a result), participant 2 had one ulcer for
12 months, and participant 3 had two recurrent ulcers for 4
years. Participant 1 cared for the ulcer himself, participants 2 and
3 had spouses who enacted the wound care with them. All three
participants had smartphones and received weekly clinical care
from the diabetic foot clinic.
Participants were interviewed by author Si Da Seng at
Figure 3: Wound Analysis process - take image (left), Chermside Community Health Centre (Metro North Hospital
and Health Service, Brisbane, Australia). Standardised
outline inside and outside of wound (middle),
background questions were asked about their ulcer history,
segmentation for wound size estimation (right). clinical care and self-care, and what applications they used on
5. Wound Size Tracking (Motivation, Figure 4.b) - live their smartphone, followed by a discussion of the MyFootCare
wound size graph visualizations, derived from prototype using a talk-aloud protocol. In discussing the
segmented wound photos in Feature 2., providing prototype, we asked about their first impressions, followed by
reports for self-tracking and clinician communication. questions about if and how they could integrate the various
features into their care practices, closing with suggestions for
improving the app for their personal circumstances and needs.
5 MYFOOTCARE EVALUATION Example questions included, "What do you think about each
individual feature?", "You mentioned that you would consider
trying out this smartphone application. Why is that?" Interviews

464
 OzCHI 2017, Nov 28 - Dec 1, Brisbane, Australia
Short Papers Human - Nature

lasted between 20 and 30 minutes. We transcribed all interviews All participants found the MyFootCare prototype easy to use.
and analysed them thematically through several rounds of However, they pointed out several minor usability issues to
reading and coding [2]. improve its utility. First, the goal setting feature needs to
We acknowledge that the findings here are limited by the highlight that users are encouraged to set their own goals by
small number of participants in this pilot study and the uploading their own photos. Second, the diary and photo section
discussion of potential use and utility, rather than actual use should not only capture the date but also the time of the day. As
while caring for their DFU at home. Despite these limitations, pointed out by participant 1, the time of the day provides
however, the evaluation served our aims of understanding and important contextual information, e.g., “you might get out of bed
enhancing the utility of MyFootCare. and feel excellent.” Finally, taking photos of the underside of the
foot was not tested in our evaluation. We talked through the
5.1 Potential to Support Care proposed process of laying the phone on the floor and
Overall, participants 1 and 3 were very positive about the positioning the foot over it, but as pointed out by participant 1,
MyFootCare prototype and saw its potential to support their “you’d have to use it to know.”
care. Participant 3 indicated that the app has potential to Participant 3 suggested adding a feature to contact a health
empower him in his interactions with podiatrists: “it would make professional for advice based on the photos and graphs. He
me more involved, because that’s the stuff that you guys recommended “a section, like a messenger, where you get online
[podiatrists] do. That would make me feel like I’m more involved help if you’ve got a question; for example, ‘I noticed a different
and my thoughts and opinions are more valued because I’m typing colour ooze coming out of the wound.’ You can share the photo and
in my diary how I feel.” Participant 2, on the other hand, stated ask, ‘should I contact my podiatrist or can it wait to the next
“I’m too old for that” and elaborated that he rarely uses apps appointment?’”
because he finds it hard to touch and type on a phone: “I can take
photos, but it would take me forever to type.” 6 CONCLUSIONS
All three participants found photos of their ulcers important The main contribution of this paper is a prototype self-care
to support their self-care. In fact, all three participants had taken mobile application with novel image capture interaction
photos of their own ulcer in the past with their own phone. techniques, wound analysis, diary features with reminders,
Participant 1 stated “it’s a record, a memory” and similarly integrated together into a self-care service.
participant 3 elaborated that “no one wants to see a photo of a Feedback from people with chronic DFUs showed that the
chronic ulcer, but for me it shows where I’ve come from, what it application has the potential to support care. It builds on existing
looked like then, and what it looks like now.” Even participant 2, practices of tracking diet [22] as well as taking photos of DFUs.
who rarely used applications on his phone, had ulcer photos The self-tracking idea was perceived well, as it makes progress
taken by his wife. objective and visible, and is thus a value add to their present
Wound size tracking graphs were seen as an easy and practices. We also acknowledge mixed results on diary and goal
objective way to monitor progress. Participant 2 stated that photos; indicating the need to clarify the results through an
graphs are “simple to read; you can see if it’s going down – just actual trial study where people with DFUs can try out the app
like what I have on my glucose monitor.” Participant 1 highlighted over several weeks.
that graphs present information in an objective manner: “it’s not The photo-taking process where users place the phone on the
gonna lie. is it [the wound] bigger? Yes or no?” Participant 3 floor and receive audio feedback also needs further evaluation.
pointed out the opportunity for sharing this information with Not all people with chronic DFUs have partners who support
healthcare professionals: “I like that because then I can show my wound care and who can take photos of ulcers. Hence this
doctor that this is the progress that is happening. It’s a nice and feature will be crucial to allow individuals to take photos of the
easy way to look at it.” underside of their foot. This feature constitutes an interesting
Goal setting via motivational images received mixed interaction design challenge, which can add to the emerging area
feedback. Participant 1 found the goal setting useful, because “it of foot-based interaction techniques [24].
shows what you want to achieve, like being able to shower [without The findings indicate that the application can empower
a covered wound dressing].” Participants 2 and 3 misunderstood individuals with a DFU in their interactions with clinicians.
this feature. They thought that the images would come from the Currently only clinicians (such as podiatrists) track wound size
app, and they missed the point that the app encourages them to during consultation, whereas the application provides users with
upload photos that represent their own personal goals. the opportunity to monitor in daily life. The integration of image
Participant 2 said that this is “not useful for me”, whereas capture, processing, monitoring and diary writing is novel and
participant 3 stated that he would prefer text rather than potentially powerful tool for motivating best practice self-care.
pictures “I prefer something like ‘don’t beat yourself up – next Furthermore, these features adhere to recommendations from
week will be better’ – you see it on Facebook all the time.” related recent studies [23, 18, 10, 9].
Similarly, the diary received mixed feedback. Participant 1 Concluding, the MyFootCare prototype has received
said that he is “not a journal person, that’s why I like the graph.” promising feedback that self-tracking based on analytics of
Participant 2 pointed out that he finds it very difficult to type on wound images can add value to the care of foot ulcers. The next
a smartphone. Participant 3, on the other hand, found the diary step is to evaluate with a more mature application in the field.
useful to add personal information to the photo and graph to
“check the fact.”
ACKNOWLEDGMENTS
5.1 Enhancing the Utility of MyFootCare This research was supported by a QUT School of Clinical Science
Seeding Grant and a QUT Engagement Innovation Grant.

465
OzCHI 2017, Nov 28 - Dec 1, Brisbane, Australia
Human - Nature Short Papers

REFERENCES with diabetic foot ulcers compared to controls in their free-living

environments. International Wound Journal, (2017) In press 13th
[1] Bus SA, van Netten JJ: A shift in priority in diabetic foot care and July 2017:doi: 10.1111/iwj.12782.
research: 75% of foot ulcers are preventable. Diabetes/Metabolism [22] Smith, B.K., Frost, J., Albayrak, M. and Sudhakar, R. Integrating
Research and Reviews 32 (2016), 195-200. glucometers and digital photography as experience capture tools to
[2] Braun, V. and Clarke, V. Using thematic analysis in psychology. enhance patient understanding and communication of diabetes self-
Qualitative research in psychology 3, 2 (2006), 77-101. management practices. Personal and Ubiquitous Computing 11, 4
[3] Dorresteijn JAN, Kriegsman DMW, Assendelft WJJ, Valk GD: (2007)
Patient education for preventing diabetic foot ulceration. Cochrane [23] Van Netten JJ, Clark D, Lazzarini PA, Janda M, Reed L: The validity
Database of Systematic Reviews 12 (2014). and reliability of remote diabetic foot ulcer assessment using mobile
[4] Fogg, B. A behavior model for persuasive design. In Proceedings of phone images. Scientific Reports 2017, In press (August 1 2017).
the 4th International Conference on Persuasive [24] Velloso, E., Schmidt, D., Alexander, J., Gellersen, H., & Bulling, A. .
Technology (Persuasive '09). ACM, New York, NY, USA, Article 40, The feet in human--computer interaction: A survey of foot-based
(2009) 1-7. interaction. ACM Computing Surveys (CSUR) 48, 2 (2015) 21:1-35.
[5] Frost, J., & Smith, B. K. (2003, June). Visualizing health: imagery in
diabetes education. In Proceedings of the 2003 conference on
Designing for user experiences (pp. 1-14). ACM.
[6] Fritz, T., Huang, E.M., Murphy, G.C. and Zimmermann, T.
Persuasive technology in the real world: a study of long-term use of
activity sensing devices for fitness. In Proc. CHI 2014, ACM (2014),
487-496.
[7] Frykberg, R. Diabetic Foot Ulcers: Pathogenesis and Management,
American Fam Physician. 66 (2002) 1655-1663.
[8] Goyal, M. Reeves, N. Rajbhandari, S. Spragg, J. and Yap, M. Fully
Convolutional Networks for Diabetic Foot Ulcer Segmentation.
arXiv preprint arXiv:1708.01928 (2017).
[9] Hazenberg, C. E., van Netten, J. J., van Baal, S. G., & Bus, S. A.
Assessment of signs of foot infection in diabetes patients using
photographic foot imaging and infrared thermography. Diabetes
technology & therapeutics, 16 (2014), 370-377.
[10] Kanazawa, T., Nakagami, G., Goto, T., Noguchi, H., Oe, M.,
Miyagaki, T., … & Sanada, H. Use of smartphone attached mobile
thermography assessing subclinical inflammation: a pilot
study. Journal of wound care, 25 (2016) 177-182.
[11] Lazzarini PA, Gurr JM, Rogers JR, Schox A, Bergin SM: Diabetes
foot disease: the Cinderella of Australian diabetes management?
Journal of Foot and Ankle Research, 5 (2012) 24-24.
[12] Lazzarini PA, O’Rourke SR, Russell AW, Derhy PH, Kamp MC:
Reduced Incidence of Foot-Related Hospitalisation and Amputation
amongst Persons with Diabetes in Queensland, Australia. PLoS
ONE, 10 (2015 ).
[13] Lazzarini PA, Hurn SE, Kuys SS, Kamp MC, Ng V, Thomas C, Jen S,
Kinnear EM, d'Emden MC, Reed L: Direct inpatient burden caused
by foot-related conditions: a multisite point-prevalence study. BMJ
Open 2016, 6:e010811.
[14] Li, I., Dey, A. and Forlizzi, J. A stage-based model of personal
informatics systems. In Proc. CHI 2010, ACM (2010), 557-566.
[15] Liu, W., Ploderer, B. and Hoang, T. In bed with technology:
challenges and opportunities for sleep tracking. In Proc. OzCHI
2015, ACM (2015), 142-151.
[16] Mentis, H.M., Komlodi, A., Schrader, K., Phipps, M., Gruber-Baldini,
A., Yarbrough, K. and Shulman, L. Crafting a View of Self-Tracking
Data in the Clinical Visit. In Proc. CHI 2017, ACM (2017), 5800-5812.
[17] Ploderer, B., Fong, J., Withana, A., Klaic, M., Nair, S., Crocher, V.,
Vetere, F. and Nanayakkara, S. ArmSleeve: a patient monitoring
system to support occupational therapists in stroke rehabilitation.
In Proc. DIS 2016, ACM (2016), 700-711.
[18] Price, P. "How can we improve adherence?" Diabetes/metabolism
research and reviews 32, S1 (2016), 201-205.
[19] Ravichandran, R., Sien, S.-W., Patel, S.N., Kientz, J.A. and Pina, L.R.
Making Sense of Sleep Sensors: How Sleep Sensing Technologies
Support and Undermine Sleep Health. In Proc. CHI 2017, ACM
(2017), 6864-6875.
[20] Schaper NC, Van Netten JJ, Apelqvist J, Lipsky BA, Bakker K.
Prevention and management of foot problems in diabetes: A
Summary Guidance for Daily Practice 2015, based on the IWGDF
guidance documents. Diab Metab Res Rev 32 S1 (2016), 7-15.
[21] Sheahan H, Canning K, Refausse N, Kinnear EM, Jorgensen G,
Walsh JR, Lazzarini PA: Differences in the daily activity of patients

466