Automatic detection of tic activity in the Tourette Syndrome Conference Paper uri icon

abstract

  • This study presents a simple decision-support system for the detection of tic events during the Tourette Syndrome (TS). The system is based on a triaxial accelerometer placed on the patient%27s trunk. TS is a neurological disorder that emerges during childhood and that is characterized by a large spectrum of involuntary/compulsive movements and sounds. 12 subjects with chronic TS participated in the study and the tic events were both measured by the proposed device and visually classified through video recording. 3D-acceleration timeseries were combined through a module operator and their noise was eliminated by a median filter. Signal to noise ratio was improved by a nonlinear energy operator. Finally, a timevariant threshold was used to detect tic events. The automatic tic recognition showed a performance around 80 %25 in terms of sensitivity, specificity and accuracy. In conclusion, this simple, automatic and unobtrusive method offers an alternative approach to quantitatively assess the tic events in clinical and non clinical environments. This overcomes the limitations of the current motor tic evaluation which is done by clinical observation and/or video-inspection in specialized neurological centres. © 2010 IEEE.
  • This study presents a simple decision-support system for the detection of tic events during the Tourette Syndrome (TS). The system is based on a triaxial accelerometer placed on the patient's trunk. TS is a neurological disorder that emerges during childhood and that is characterized by a large spectrum of involuntary/compulsive movements and sounds. 12 subjects with chronic TS participated in the study and the tic events were both measured by the proposed device and visually classified through video recording. 3D-acceleration timeseries were combined through a module operator and their noise was eliminated by a median filter. Signal to noise ratio was improved by a nonlinear energy operator. Finally, a timevariant threshold was used to detect tic events. The automatic tic recognition showed a performance around 80 %25 in terms of sensitivity, specificity and accuracy. In conclusion, this simple, automatic and unobtrusive method offers an alternative approach to quantitatively assess the tic events in clinical and non clinical environments. This overcomes the limitations of the current motor tic evaluation which is done by clinical observation and/or video-inspection in specialized neurological centres. © 2010 IEEE.

publication date

  • 2010-01-01