Handwriting analysis for assistant diagnosis of neuromuscular disorders

Abstract

This paper presents a handwriting movement analysis approach and its application in assistant diagnosis of the neuromuscular disorders rehabilitation by measuring the movement smoothness. The time-varying primitives extraction algorithm is developed to segment the handwriting strokes from natural handwriting data. Further seven smoothness metrics are proposed to evaluate the motor control abilities of neuromuscular disorders and normal people. In experimental studies, the real world handwriting data from five neuromuscular disorders' are acquired to verify the developed algorithm as well as the proposed smoothness criteria. Comparative analysis of the experimental results demonstrates that the presented approach can work well in assisting the rehabilitation diagnosis.