A software tool has been developed to support the objective diagnosis of patients
with Parkinson's Disease (PD). Patients completed hand exercises using a personal
computer mouse and data has been gathered for further studies. We have analyzed different
parameters and suggest using a particular parameter vector containing median and standard
deviation values for tracking the daily changes in PD patients' status. Using the
classification abilities of self-organizing feature maps (SOFM) we were able to provide
support for the diagnostic process.