Detecting Parkinsonian Symptoms using Data Analysis

Parkinson's Disease (PD) is one of the most common neurodegenerative diseases. In this paper, we have researched the various methods used to detect major symptoms of Parkinson's Disease (PD), namely resting tremors, impaired gait and vocal impairments. We have conducted an extensive literature review on the automated detection of each of the aforementioned symptoms. Each symptom is discussed in a separate section which covers the data mining and classification techniques for the respective symptom and weighs the advantages and disadvantages of each method. These techniques are then compared to find the most useful, convenient and easy-to-use process which could be easily implemented in a normal medical checkup. A comparison between these processes has been presented with respect to their accuracies, precisions and other relevant metrics. This comparison reveals that testing the extent of tremors can detect PD with an accuracy of 92.19% while examining gait results in 92.71% accuracy and voice-based analysis provides 97.7% accuracy.