FALL DETECTION IN ALZHEIMER'S DISEASE PATIENTS USING MACHINE LEARNING, INTEGRATED WITH A WRIST-WRAP DESIGN AND MOBILE APP

Based on information collected from various research papers [1] about the blood pressure values and daily check-up details of patients suffering from Alzheimer's disease (AD), a machine learning algorithm is applied to predict the fall of the patients. A wrist wrap is designed to detect the fluctuations in blood pressure, pulse rate and muscle activity of the patient and will help in predicting the fall of the patient with precise location [2]. The logistic regression model is trained using the dataset on two different platforms, i.e., Peltarion and Weka. Finally, the fluctuation in blood pressure and fall detection was verified using Arduino based tool and appropriate C# coding. The accuracy of the test was nearly 66%.