An influence of Radon Transform Technique on Handwriting Task for the Detection of Parkinson's disease

• Published in: 2021 16th International Conference on Computer Engineering and Systems (ICCES) pp. 1 - 5
• Main Authors: Elazazy, Sara A., Eldesoky, Mohamed A., El-Wakad, Mohamed T., Soliman, Ahmed M.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 15-12-2021
• Subjects: Handwriting drawings; K-Nearest Neighbor; Logistic Regression; Magnetic resonance imaging; Parkinson's disease; Parkinson's Disease Tremor; Radon; Radon transform; Random Forest; Spirals; Support Vector Machine; Support vector machines; Training; Transforms
• DOI: 10.1109/ICCES54031.2021.9686160

Tremors, stiffness, bradykinesia, and rigidity are all symptoms of Parkinson's disease, which is caused by a lack of dopamine. To detect Parkinson's disease, several diagnostic techniques were used, including MRI scans and EEG data. The lack of standard methods causes findings to take a long time to appear, and these approaches are typically expensive. The innovative notion for detecting Parkinson's illness is based on handwriting image processing. Scientists discovered in prior investigations that the drawing rate becomes stable in healthy situations and becomes unstable in patient cases. We employed the Radon Transform to improve the training of machine learning classifiers (KNN, LR, RF, and SVM) and improve their ability to detect Parkinson's disease in this article. SVM classifier has the best accuracy of 92.45 per cent, with 70.38 % specificity, 88.98 % F1 score, and precision of 87.68 %.