Decreasing risk for developing diabetic foot ulceration with personalised feedback on daily physical activity

Background

Diabetic foot ulcers are one of the most feared complications of diabetes. These ulcers usually develop in people with diabetic neuropathy with loss of protective sensation in combination with biomechanical abnormalities. Abnormal loading of tissue of the foot is the result of factors such as foot deformities, poorly fitting shoes and lack of sensory feedback. When during a period of walking subsequently tissue damage occurs, the patient will continue to walk due to lack of pain, resulting in ulceration. Therefore, prevention is very important. In particular in people with an history of ulceration, the risk of recurrence is high.

Why prevention fails in these people is still largely unclear. Special footwear to reduce the abnormal loading is frequently prescribed, but unfortunately patients do not always wear these. In addition, the duration of loading, i.e. number of steps taken, probably also plays an essential role. However, quantitative data on physical activity are mostly lacking, and therefore it is not possible to provide people at risk of ulceration concrete advice on safe activity-rich days in order to guide personalised prevention. Within this context, smart (wearable) technology, and the application of artificial intelligence (AI) solutions are essential tools to help in measuring, qualifying and modifying behaviour in ulcer prevention, allowing for gender, age, and cultural differences in physical activity to also be considered. In this DIALECT project, we aim to assess, validate, and implement machine learning techniques to better understand how the quantity and quality of daily physical activity could inform personalized strategies for foot ulcer prevention in people with type 2 diabetes.

Approach

The doctoral candidate will primarily use data from The Maastricht Study, and in addition, clinical observational data will be obtained. The Maastricht Study is an extensive, unique, phenotyping study that focuses on the etiology of type 2 diabetes, its classic complications (cardiovascular disease, nephropathy, neuropathy and retinopathy), and its emerging comorbidities, including cognitive decline, depression, and gastrointestinal, respiratory and musculoskeletal diseases. The study uses advanced state-of-the-art imaging techniques and extensive biobanking to determine health status in a population-based cohort of 9200 individuals that is enriched with type 2 diabetes participants. Detailed physical behaviour (sleep, sitting, standing, physical activity) data have been acquired with a thigh-worn accelerometer (ActivPAL3).  Advanced statistical analyses, which may include group based trajectory modelling and AI approaches, will be used in this project to identify physical activity patterns. Depending on the other parts of the project, setting up an observational, prospective study will be considered. The outcomes of these studies are to be translated using co-creation design involving end-users, digital solutions and communication strategies, to provide personalised feedback to people at risk of ulceration.

Secondments will take place at Amsterdam UMC, to analyse data from an existing prospective observational cohort study, and for additional clinical placement; and at Novel, for an industry experience and to link the activity profiles with the smart technology under development at Novel.

Our Research Team

Research at UM is characterized by a multidisciplinary and thematic approach, and this is visible in this project. The supervisory team of the candidate is multidisciplinary, with emphasis on movement sciences, epidemiology, and clinical (scientific) expertise on diabetic foot ulceration from the academic hospital. Moreover, our team has ample expertise on the measurement and analyses of accelerometer data. The human movement sciences group focuses on physical activity behavior and energetics of human movement. Epidemiological know-how and coordination of the Maastricht Study will be provided from the department of Social Medicine.

External supervision in this project comes from DIALECT partners at Amsterdam UMC (Amsterdam, Netherlands) and Novel GmbH (Munich, Germany). Amsterdam UMC is a leading institute on clinical, biomechanical and radiological research on diabetic foot disease, in particular the prevention of foot ulceration and amputation. Novel GmbH is the global leader in accurate and reliable load distribution measurement systems and will ensure a link with the smart technology under development as part of DIALECT.

Maastricht University

Maastricht University is renowned for its unique, innovative, problem-based learning system, which is characterized by a small-scale and student-oriented approach. Research at UM is characterized by a multidisciplinary and thematic approach, and is concentrated in research institutes and schools. Maastricht University has around 22,000 students and about 5,000 employees. Reflecting the university’s strong international profile, a fair amount of both students and staff are from abroad. The Faculty of Health, Medicine and Life Sciences, with its long history of mobility and diabetic foot disease research, is one of Maastricht’s 6 faculties: Faculty of Law, School of Business and Economics, Faculty of Science and Engineering, Faculty of Arts and Social Sciences, Faculty of Psychology and Neuroscience. For more information, visit www.maastrichtuniversity.nl.