Handwriting Analysis for Alzheimer’s Diagnosis: Feature Enrichment and Machine Learning Classification

Alzheimer’s is an incurable neurodegenerative disease that causes cognitive impairment. Various treatment methods are applied to slow the progression of the disease and improve the quality of patient’s life. However, early diagnosis and treatment are of great importance. Due to motor coordination difficulties and decline in cognitive functions, abnormalities can be observed in the handwriting of Alzheimer’s patients. Therefore, handwriting analysis is considered as an important tool for the early diagnosis of the disease. This study aims to detect Alzheimer’s disease through handwriting analysis. Features extracted from different handwriting tasks of both healthy and diseased individuals were classified using machine learning methods. In addition to the existing features in the original dataset, two features were added, and the impact of these fetaures on the classification performance was evaluated. The results of the experiments and analyses revealed that the new features significantly improved the classification performance. The highest accuracy rate, $85.86 \%$, was achieved using the Logistic Regression method.