TREMOR CLASSIFICATION USING WEARABLE IOT BASED SENSORS

Abstract

Tremors, a significant symptom of movement disorder, affects a part of the body ranging from slight to severe. These Tremors are symptoms of various neurological diseases such as Parkinson’s disease (PD), Essential tremors (ET), Physiological tremors (PT), Cerebellar tremor, Dystonic tremor, Psychogenic tremor, and many more. Tremor features and types differ for various neurological disorders. During the early stages of the disease, clinical examination of tremors plays a significant role in diagnose management. This work aims to develop a wearable assistive system with an Inertial Measurement Unit (IMU) sensor to classify the tremor of three different neurological disorders based on the tremor position and frequency. This research has been carried out in SRM Medical college and Research Centre with 15 patients. The type of neurodegenerative disease of the subject with tremor is evaluated based on the tremor position and tremor frequency level. The data is collected, transmitted, and processed using the IMU sensor with Internet of things (IoT) and Node MCU board. The decision tree algorithm is used for the classification of tremors. ET, PD, and PT tremors are classified based on the tremor frequency and tremor position. A high rate of accuracy is achieved for the developed system when compared with the Neurologist results. The proposed device quantitatively classified the tremor based on the frequency and position among the three different neurological disorders, i.e., ET, PD, and PT tremors.