An Image Analysis Based Method for the Quantification Of Tremor

Abstract

In th is study, we introduce a simple and cost-effective method based on image analysis that can be used in the objective assessment/measurement of tremor. Tremor can be defined as the involuntary rhythmic movement of body parts (usually hands). Types of tremor include, but not limited to essential tremor, athetose, chorea, and flapping tremor. Objective assessment of tremor is crucial in the diagnosis of many neuromuscular d iseases. Velocity and acceleration sensors utilizing semiconductor technology a re widely used for measurement of tremor. However, one major weakness of this type of sensors is that they cannot sa tisfactorily track slow or constant speed motion, because of their poor response in the low-frequency region, where some of tremors may have some significant spectral components. Therefore, accurate measurement of tremor for the diagnosis or prognosis of movement disorders is an important issue. The method that we devise consists of an in fra-red LED that is attached to the body part whose motion to be measured and a simple CCD camera that is a ttached to a personal computer. The frames acquired by th e system first saved into a movie file, and then analyzed using a software tool that we have developed. We present the use and advantages of the devised system on a sample artificial data se t.