A classifier procedure to identify a Parkinson’s disease by using handwriting of Archimedean spiral

The costs associated with the diagnosis of Parkinson’s disease and its medications are considerably high. One of the main problems with Parkinson’s disease is the difficulty with writing, and so one can use an Archimedean spiral to assess that. In this paper, a classification procedure is developed to categorize Parkinson’s patients as compared to non-Parkinson patients. A threshold, determined by bootstrap method, is used to classify if the spiral break could be due to Parkinson’s disease. A statistical test, using a parametric bootstrap method, is used for this purpose. We demonstrate that the proposed approach works well by means of Monte Carlo simulations as well as real data sets involving Parkinson’s patients’ and healthy individuals’ handwritings.