Parkinson's Disease Assist Device Using Machine Learning and Internet of Things

Abstract

The main purpose of the paper is to construct a prototype of a spoon which senses the hand motion of an individual suffering from Parkinson's disease. It gives a counter motion to the trembling actions of a patient's hand, in order that they don't spill their food. The movement of the hand acts like feedback to a microcontroller that operates two servos that provide counter-motion. There are Inertial Measurement Units positioned on the spoon that measure the deviations in yaw, pitch, and roll of the spoon and thus govern the counter balancing action of the servos. A PID controller is used to maintain a tilt to the spoon's angle and as a consequence the spoon is kept at a level plane. Features such as IoT and Machine Learning have been integrated into the system, that enables the product to monitor changes. IoT is used in the device to monitor the sensor reading remotely and analyze the progress of the patient. Machine Learning is used to detect any anomalies in the sensor data from the device. If an anomaly is detected the device send a mail alert to the well-wishers of the patient.