Design and Implementation of a Machine Learning-Based Drop-Foot Management System

This work presents the design and implementation of a machine learning‑based drop‑foot management system that delivers personalized rehabilitation recommendations using patient‑reported symptoms, pain characteristics, severity levels, and assistive device usage. The system compares Logistic Regression, Random Forest, and LightGBM models, achieving up to 90% classification accuracy and 0.99 ROC‑AUC, and is deployed via a Streamlit web application to improve accessibility for patients and clinicians.Preprint of undergraduate thesis manuscript on a machine learning‑based drop‑foot management system, currently under review with supervising professors at Afe Babalola University.