Quantitative assessment of essential tremor based on machine learning methods using wearable device

Chenbin Ma, Deyu Li, Longsheng Pan, Xuemei Li, Chunyu Yin, Ailing Li, Zhengbo Zhang, Rui Zong

Abstract: Highlights•Building the first resting posture database base on patients with essential tremor, relying on rigorous screening experiments and focusing on laboratory quantification.•It used wearable devices that integrate high-dimensional information that collects data under a standardized laboratory examination.•We obtained the combined consensus scores through independent scoring by multiple neural experts.•Combining multimodal features to build machine learning models and perform comprehensive performance evaluation.•The performance of our proposed method is optimal relative to other studies of five classifications of tremor severity classification.