Automatic diagnosis of multi-task in essential tremor: Dynamic handwriting analysis using multi-modal fusion neural network

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

Abstract: Highlights•This study proposes a novel spatial-temporal-spectral fusion neural network (STSNet) for multi-task fine-grained assessment of tremor.•STSNet can efficiently fuse complementary information from static handwriting images and dynamic multi-sensory signals.•We validate the utility of the attention mechanism, transfer learning strategy, and the features-based prior knowledge by ablation studies.•We collected the first multi-sensing laboratory examination dataset from 147 essential tremor patients based on a rigorous clinical paradigm.•Our proposed model has optimal classification performance compared to other state-of-the-art studies.