QUANTITATIVE ANALYSIS OF ESSENTIAL TREMOR DURING CLINICAL SPIRAL DRAWING TASK USING GYRO SENSORS

The spiral drawing task is a representative clinical test used to assess essential tremor. Many clinicians have used spirography to assess through visual rating of the drawings. We quantitatively investigated the clinical characteristics of action tremors during spiral drawing tasks using 3-axis gyro sensors. Eighteen essential tremor patients with sensors attached to their forefinger, dorsum of the hand, and forearm participated in a spiral drawing test. As quantitative variables, the average speed, maximum frequency and maximum power were derived from the gyro sensor signals. The average speed and maximum power were also logarithmically transformed. Two-way repeated measures analysis of variance (ANOVA) with tremor direction and joint (attachment location) as independent factors was performed to evaluate the main and interactive effects of direction and joint. Also, Spearman's correlation analysis was performed to evaluate the relationship between the quantitative variables and visual rating scores. The average speed showed a significant main effect of direction (P<0.01). Logarithmic variables showed interactive effects (P<0.01) as well as main effects (P<0.01). The interactive effects were attributed to no significant difference among the joints particularly in the roll direction (P>0.05). The logarithmic variables of roll direction were significantly greater than that of the other directions (P<0.001). Also, the quantitative variables were moderately correlated with visual rating scores (r=0.48-0.73, P<0.05). These results indicate that kinetic tremor during spiral drawing mainly appears in the roll direction and may contribute to the determination of sensor location and direction for continuous monitoring of patients with essential tremor.