Diabetes patients risk developing foot ulcers that can lead to amputation, which is a dire concern. According to a 2015 study, 6% of the 22.3 million American with diabetes develop foot ulcers each year. Overall, one in four can expect to suffer foot ulcers in their lifetime. The best defense against amputation is early detection followed by proper treatment.
Manchester Metropolitan University researchers Dr. Moi Hoon Yap, a Senior Lecturer in Computer Science, and Professor Neil Reeves, Professor of Musculoskeletal Biomechanics, have developed FootSnap. The FootSnap app helps medical personnel use Apple iPads to detect and monitor foot ulcers in diabetic patients. Yap and Reeves employ algorithms with deep learning, a form of artificial intelligence and machine learning. The program analyzes FootSnap images in the context of a large data set of images of diabetic foot ulcers obtained from the Lancashire Teaching Hospitals NHS Foundation Trust. The team has worked to standardize image distance, orientation, and lighting to improve qualitative image comparisons. The app guides the medical professional capturing the image to orient and align patient’s foot in a manner that helps build a portfolio of uniform images to assist with diagnosis and treatment monitoring.
The app should be available for download soon. Next steps are to modify the program for use with smartphones and to enable less-trained operators. A report published in the Journal of Diabetes Science and Technology documents initial FootSnap proof-of-concept testing success. In the future, diabetics may be able to screen for potential foot ulcers with a smartphone photo.