Authors :
A. Anisha, B.E., M.E; Femima Shelly. A. T; Benitta. R. K; Amala Selciya. T.L
Volume/Issue :
Volume 8 - 2023, Issue 5 - May
Google Scholar :
https://bit.ly/3TmGbDi
Scribd :
https://shorturl.at/dhjz9
DOI :
https://doi.org/10.5281/zenodo.8021545
Abstract :
Parkinson's disease is a neurological disorder
that primarily affects people over the age of 60, often
leading to motor impairment (MI) such as tremors,
rigidity, and slowness. The disease's severity has been
found to be linked to a decline in handwriting quality,
with patients exhibiting reduced speed and pressure
while writing. Biomarkers can aid in the diagnosis,
monitoring, and prediction of the disease's progression,
making it critical to accurately identify them. A
convolutional neural network (CNN) is used in this study
to analyze spiral drawing patterns from Parkinson's
patients and healthy individuals, with the aim of creating
a system that can effectively differentiate between the
two groups and predict the PD stage. The model was
trained on data from 280 patients and achieved an
overall accuracy of 94.2%. Identifying biomarkers could
provide valuable insights into the disease's causes and
lead to better diagnosis and treatment outcomes
Keywords :
Parkinson’s Disease, CNN, Deep Learning, Machine Learning, Cat Boost Classifiers, VGG-16 Model
Parkinson's disease is a neurological disorder
that primarily affects people over the age of 60, often
leading to motor impairment (MI) such as tremors,
rigidity, and slowness. The disease's severity has been
found to be linked to a decline in handwriting quality,
with patients exhibiting reduced speed and pressure
while writing. Biomarkers can aid in the diagnosis,
monitoring, and prediction of the disease's progression,
making it critical to accurately identify them. A
convolutional neural network (CNN) is used in this study
to analyze spiral drawing patterns from Parkinson's
patients and healthy individuals, with the aim of creating
a system that can effectively differentiate between the
two groups and predict the PD stage. The model was
trained on data from 280 patients and achieved an
overall accuracy of 94.2%. Identifying biomarkers could
provide valuable insights into the disease's causes and
lead to better diagnosis and treatment outcomes
Keywords :
Parkinson’s Disease, CNN, Deep Learning, Machine Learning, Cat Boost Classifiers, VGG-16 Model