Abstract:
This research aims at filling the existing gap of Diabetic Foot Ulcers (DFUs) through deploying a mobile application involving machine learning to enhance early detection and intervention. The application trained the arising model to predict DFUs with high accuracy simply by merging not only the features elicited from medical image analysis, but also other features from patient’s demographics. Flipping through several machine learning models, the most optimal one is used for classification of DFU. The system incorporates a registration screen for creating the account, a login screen for accesses, and detailed diagnosis screen for uploading medical images with classification and prediction on possibility of DFUs. To communicate with the mobile application and the built machine learning model, Flask web-interface is employed for the development of Application Programming Interface (API) that provides real time predictions and suggestions. This integration benefits healthcare professionals and diabetics by giving warnings about DFU development and seeking treatment immediately. To close the gap between theoretical models and actual applications in the management of diabetic foot, the project will be of importance in promoting knowledge regarding available technological applications for optimal diabetic foot complications management and ultimately, the improvement of patient outcomes in diabetic foot.
