Parkinson's disease (PD) is a progressive neurodegenerative disorder characterized by motor and non-motor symptoms, with handwriting often serving as a critical marker for early diagnosis. This study presents a novel system for early PD detection using offline hand-drawing analysis. Following data pre-processing and augmentation, we defined and trained a deep learning model based on the CNN-VGG16 architecture for PD detection. To validate the approach, we developed a publicly accessible Iraquian hand-drawing dataset, available at https://ieee-dataport.org/documents/online-offline-iraquian-hand-drawing-dataset-early-parkinsons-disease-detection . The dataset comprises hand-drawing data from 30 healthy individuals and 30 PD patients from Marjan Hospital in Hilla, Iraq, including tasks such as repetitive ellipses, spirals, digits, and Arabic word writing. The results demonstrate the system’s effectiveness in detecting PD at early stages.