Alzheimer's Patient Support System Based on IoT and ML

Abstract

Alzheimer's disease poses significant challenges as it progressively erodes memory and identity, severely impacting daily functioning. Patients often experience disorientation, wandering, and are at risk of falls, leading to heightened concerns for caregivers. These difficulties can result in a loss of independence and increased caregiver burden. In response to these challenges, this study introduces an innovative assistive system designed to enhance the safety and quality of life for Alzheimer's patients. The system comprises of two main components: a smart arm band and a facial recognition system. The smart arm band is equipped with a suite of sensors including GPS, accelerometer, and heart rate sensor. These sensors enable real-time monitoring of the patient's location, movement, and physiological parameters. By leveraging these data streams, caregivers can track the patient's activities, detect falls or emergencies, and provide timely assistance when needed. The facial recognition system employs state-of-the-art machine learning techniques, specifically the CAFFE and Local Binary Patterns Histograms (LBPH), to recognize familiar faces in the patient's environment. This capability promotes social interaction and enhances the patient's sense of familiarity and security. Through rigorous testing, the facial recognition system achieves an impressive accuracy of 97% with a low error rate of 3%, validating its effectiveness in real-world scenarios. Overall, the integrative assistive system presented in this study offers a promising solution to address the multifaceted challenges associated with Alzheimer's disease. This system provides caregivers with invaluable support in ensuring the safety and well-being of Alzheimer's patients while fostering social engagement and autonomy.