Image-based Handwriting Analysis for Disease Diagnosis

Handwriting movement is an activity dominated and controlled by human advanced nervous system, which depends on coordination and cooperation of multiple parts of the whole body, and the analysis of handwriting movement has been applied in many fields. In disease diagnosis, handwriting tremor is common in neurodegenerative diseases(NDs), such as Parkinson's disease(PD) and Alzheimer's disease(AD). In this paper, the image-based handwriting trajectories of mild and severe Parkinson's disease patients and healthy people with no tremor are collected, and the features of images are extracted. Utilizing corner detection method to judge the fluctuation of trajectory; 2-dimensional Discrete Fourier Transform (2D-DFT) is applied to transform it into frequency-domain space, and texture features of amplitude spectrum are extracted based on gray level co-occurrence matrix (GLCM). Finally, the machine learning method is used to classify the above features, so as to realize the diagnosis of diseases.