A Smartphone-Based Clinical Decision Support System for Tremor Assessment

摘要

Tremor severity assessment is an important element for the diagnosis and treatment decision-making process of patients suffering from Essential Tremor (ET). Classically, questionnaires like the ETRS and QUEST surveys have been used to assess tremor severity. Recently, attention around computerized tremor analysis has grown. In this study, we use regression trees to map the relationship between tremor data that is collected using the TREMOR12 smartphone application with ETRS and QUEST scores. We aim to develop a model that is able to automatically assess tremor severity of patients suffering from Essential Tremor without the use of more subjective questionnaires. This study shows that tremor data gathered using the TREMOR12 application is useful for constructing machine learning models that can be used to support the diagnosis and monitoring of patients who suffer from Essential Tremor.