Abstract
Background: People living with diabetes are at risk of longterm complications including foot ulceration, infection and
amputation resulting in adverse psychosocial and economic
impact. However, through identification of those most at risk, proactive preventative care and engaged self-management, most diabetes complications can be prevented.
Aim: To develop a clinician-facing digital tool with features for risk stratification for foot ulceration and amputation, alerts for high risk and patient-centred advice. The patient-facing site enables self-management through input of symptoms, test results and photos, instructions for and results of touch-the-toes tests performed at home.
Method: Predictive models for identifying those at risk of foot-related complications were developed using machine learning techniques applied to a large phenotypic database; SCIdiabetes, the electronic health record for all people in Scotland with diabetes. We used a user-centric agile rapid cycle design process, starting with a user needs assessment and the prototype platform was user tested with clinician and patient users, involving ‘think aloud’ and task-based methods.
Results: While the tool was well accepted during testing, issues over the way risk is calculated, the use of AI in healthcare and patients’ subjective assessments during tests performed at home were noted throughout development. These were addressed during development.
Conclusion: The MyWay Clinical tool is ready for technical integration into and testing within the Scottish diabetes ecosystem, e.g., SCI-diabetes. This will contribute to predicting future foot risk and flexible and remote care delivery.
amputation resulting in adverse psychosocial and economic
impact. However, through identification of those most at risk, proactive preventative care and engaged self-management, most diabetes complications can be prevented.
Aim: To develop a clinician-facing digital tool with features for risk stratification for foot ulceration and amputation, alerts for high risk and patient-centred advice. The patient-facing site enables self-management through input of symptoms, test results and photos, instructions for and results of touch-the-toes tests performed at home.
Method: Predictive models for identifying those at risk of foot-related complications were developed using machine learning techniques applied to a large phenotypic database; SCIdiabetes, the electronic health record for all people in Scotland with diabetes. We used a user-centric agile rapid cycle design process, starting with a user needs assessment and the prototype platform was user tested with clinician and patient users, involving ‘think aloud’ and task-based methods.
Results: While the tool was well accepted during testing, issues over the way risk is calculated, the use of AI in healthcare and patients’ subjective assessments during tests performed at home were noted throughout development. These were addressed during development.
Conclusion: The MyWay Clinical tool is ready for technical integration into and testing within the Scottish diabetes ecosystem, e.g., SCI-diabetes. This will contribute to predicting future foot risk and flexible and remote care delivery.
Original language | English |
---|---|
Pages | 14-14 |
Number of pages | 1 |
Publication status | Published - 2024 |
Event | IDF Virtual Congress 2023: Diabetes Complications and Diabetes in Crises - Online Duration: 4 Dec 2023 → 7 Dec 2023 https://idf.org/events/idf-congress/ |
Conference
Conference | IDF Virtual Congress 2023 |
---|---|
Period | 4/12/23 → 7/12/23 |
Internet address |