Innovative Sensor-Based Approach for Predicting Foot Ulcers in Diabetic Patients Using Machine Learning Models

Foot ulcers, particularly in diabetic patients, are a significant health concern that can lead to severe complications if left untreated. Early detection of foot ulcers is critical for preventing infections, reducing hospitalizations, and improving the quality of life for at-risk individuals. This project aims to develop an automatic detection system for foot ulcers using an integrated approach that combines hardware sensors and predictive analytics software. The hardware consists of a foot temperature sensor and a pressure sensor to monitor the conditions that contribute to ulcer formation. The software, powered by machine learning algorithms, collects, preprocesses, and analyzes the data from these sensors to build a predictive model capable of detecting early signs of foot ulcers. By leveraging this integrated system, the solution provides a non-invasive, real-time detection method that can be deployed in clinical settings or as part of home monitoring systems for high-risk patients. This technology has the potential to enhance early diagnosis, reduce complications, and lower healthcare costs by enabling timely interventions. In this various predictive model like support vector machine, naive bayes, decision tree, and random forest has been implemented successfully and the results are compared. This technology has the potential to enhance early diagnosis, reduce complications, and lower healthcare costs by enabling timely interventions.