Abstract: Leveraging DenseNet architecture, our novel approach to Parkinson's disease detection focuses on analyzing spiral and wave images derived from handwriting samples, a method proven to capture subtle motor abnormalities characteristic of the condition. By training the model on a dataset comprising annotated samples from individuals with clinically confirmed diagnoses, our system learns to discern distinctive patterns indicative of Parkinson's disease. Through the integration of traditional image processing techniques for preprocessing, we enhance the model's ability to extract relevant features from handwriting patterns. The multi-label classification enables not only the identification of Parkinson's disease presence but also offers insights into its severity and progression. This comprehensive approach empowers clinicians with a reliable tool for early diagnosis and personalized treatment planning, ultimately improving patient outcomes and quality of life.
Keywords: DenseNet architecture, Parkinson's disease detection, Spiral and wave images, Handwriting samples, Motor abnormalities, Early diagnosis
Cite:
Shikha Ballal, Sourabha Jain, Sweedle Suares, Gururaj, Dr.Rejeesh Rayaroth,"Detection and Risk assessment of Parkinson’s disease : A Machine Learning Approach ", IJARCCE International Journal of Advanced Research in Computer and Communication Engineering, vol. 13, no. 3, 2024, Crossref https://doi.org/10.17148/IJARCCE.2024.13370.
|
DOI:
10.17148/IJARCCE.2024.13370
