Movement analysis for essential tremor

Tremor is a neurological disorder, affecting several millions of people around the world, an involuntary rhythmic movement of one or more of the body parts which can happen during rest or movement. Essential tremor (ET) is one of the most common neurological disorders, partially due to genetic components and potentially due to heterogeneous pathophysiological conditions affecting one or more body segments. ET is a postural and action tremor involving predominantly the upper limbs (hand), in a few cases, it also affects the voice and head.  There have been several studies undertaken to understand the cause and severity of ET. The research reported on smart technology such as wearable sensors, IMU sensors, pressure sensors and wireless sensors for tremor assessment are based on objective and quantitative diagnosis using signal processing, statistical models, and machine learning algorithms for an alternative to clinical assessment, as they apply various tremor assessment tasks that are widely and commonly used for the evaluation in neurology departments. There has been no study conducted to phenotype the family of essential tremors. To overcome this research gap, this study proposes a novel method of using digital drawing analysis by collecting data from wearable IMU sensors for essential tremor prediction, classification, and severity estimation. A novel multimodal sensor approach of combining accelerometer and gyroscope information and analysing the inter-rater, inter-segment and inter-quadrant analysis of the digital drawings was performed. Using the features of wearable IMU sensors an estimate of the clinical Fahn-Tolosa Marin (FTM) rating score was determined.