Patient-Centric Hand-Gesture Driven Wheelchair Control Using Convolutional Neural Networks for Enhanced Healthcare Mobility

This work presents a smart wheelchair system whereby users may operate it in an intuitive and hands-free manner by using a Convolutional Neural Network (CNN) for hand gesture recognition in real-time. With programmed hand signals, the smart wheelchair substitutes for conventional button-operated controllers, so translating users' mobility and freedom despite restricted physical capability. This system's basic components are an L298 motor driver to regulate wheelchair movement, a Raspberry Pi for real-time image processing, and an 8x8 LED matrix visual feedback model. Data preprocessing, supervised learning model training, and testing in many environmental settings improves the gesture recognition accuracy. Testing reveals this system offers fast response times, accurate gesture detection, and consistent control. This work supports assistive technology by giving wheelchair users a flexible, adaptive, user-friendly control mechanism. Future research should look at expanding the gesture variety, improving CNN performance for faster real-time processing, and integrating sophisticated safety elements-such as obstacle detection and voice commands.