Deep Brain Stimulation

Motion Analysis in Parkinson patients

Deep Brain Stimulation (DBS) has proved to be a highly efficient treatment of symptoms related to Parkinson’s disease (PD). Assessing the severity of tremor for example is important especially before, during and after surgery to evaluate the patient state, with and without DBS, and to quantify clinical changes. Current methods such as the Unified Parkinson’s Disease Rating Scale (UPDRS) are mainly semiquantitative (clinical rating scales relying on ordinal ratings anchored on descriptive terms such as “mild”, “moderate”, “severe”) and not objective. The objective of the present work was to develop a system based on an accelerometer fixed to the patients wrist to quantitatively evaluate the patient state.

The Freescale ZSTAR3 Sensor (digital 3D axes accelerometer) has been chosen because of the supplied software allowing multiple data storage formats and because of its wireless data transmission. To protect the sensor and to allow its fixation to patients wrist, a special case had to be designed and printed out with Rapid Prototyping. The most suitable scanning frequency was 60 Hz. Furthermore, software was developed (Java) to read and convert the measured data. For each measurement the integral (derivation between 0.6 and 16 Hz), the middle derivation and the sum of the products of each derivation and frequency were calculated.

To test the system's reactivity we first made measurements on a hydropulser, i.e. with defined movements. Second, we performed measurements on a patient prior to and after a DBS operation and we compared the results.

The designed case with its fixation proved to be suitable for these kinds of measurements. The tests showed that the sensor can be operated in a fairly straightforward way and that the written software works satisfactorily. The frequency analysis of the patient data clearly confirmed the reduction of tremor during stimulation compared to the preoperative measurements. The next step will be to refine the data analysis and to look for other quantitative parameters in order to identify further pathologic movements of Parkinson patients in addition to tremor. In the long term such a system should be used to add quantitative information about the improvement of the patient state to the clinical evaluation.

• Refinement of movement pattern analysis
• Improvement of the intraoperative patient data visualisation during the DBS implantation procedure

This article is also available as PDF-file.