Unobtrusive Detection and Monitoring of Tremors using Non-Contact Radar Sensor

People experiencing tremors, find it difficult to perform the activities of daily life. Simple yet essential everyday tasks such as eating, reaching out and grasping an object prove to be a challenge for them. Clinical assessment proves to be subjective and may not provide a true picture of how one's tremor changes throughout the day or varies from one activity to another. Hence, remote monitoring is vital. A few remote assessment approaches do exist in the literature, however, these are based on wearables. These solutions require the continuous wearing of the device, by the individual. This work presents a solution that can not only remotely and unobtrusively detect but also monitor tremors while individuals perform their daily life activities. A Frequency Modulated Continuous Wave (FMCW) radar sensor, designed for the automotive industry, has been configured to monitor these action tremors. Moreover, signal processing has been applied to convert temporal data acquired by the radar to time-frequency data. The results thus generated are used to extract useful clinical information regarding the peculiarities of tremor such as frequency, amplitude and time duration.