Parkinson's disease is one of the most common neurodegenerative diseases that affect persons over 65. Because this sickness is progressive, if it is not identified early and monitored at different times, individuals will experience severe health problems and higher healthcare expenses. One kind of neurological condition that usually affects older persons is Parkinson's disease. Parkinson's disease (PD) affects about 1% of the global population, and many of its victims struggle with intricate movement and cognitive issues. As the illness worsens, cognitive and behavioral symptoms such a variety of personality changes, depressive disorders, memory issues, and emotion dysregulation may manifest. Additionally, the movement-related symptoms intensify along with the sickness. Early diagnosis of dementia is essential for implementing appropriate therapeutic strategies to slow cognitive decline. Parkinson's disease (PD) is typically diagnosed by clinicians based on characteristic symptoms, including muscle stiffness, tremors, slowed movements, and difficulties with balance and coordination. However, the symptoms and progression rate can vary significantly among individuals, making diagnosis challenging. Currently, no specific blood test or biomarker exists to reliably confirm PD or monitor the underlying pathological changes as the disease advances. For more than three decades, magnetic resonance imaging (MRI) has been widely used to differentiate PD from other neurological conditions and aid in diagnosis. Recent studies have shown that state-of-the-art convolutional neural networks (CNNs) can achieve diagnostic accuracy comparable to that of human experts in medical imaging tasks. A key factor in medical image processing is effective feature representation. Unlike traditional machine learning methods, deep learning approaches such as CNNs can automatically extract complex, hidden features from imaging data, enabling more accurate classification. In this project, an automated system has been developed and trained on features extracted from MRI scans of both PD patients and healthy individuals. This system is designed to evaluate disease severity across different stages and distinguish PD patients from healthy controls based on neuroimaging data.
Keywords: Convolutional Neural Network (CNN); Dense Convolutional Neural Network (DCNN); Parkinson's Disease (PD) Classification; Magnetic Resonance Imaging (MRI); Neuroimaging (NI).
How to cite this article: Pragadeesh S, Kanthalakshmi S, Hybrid DenseNet-CNN Framework for Early Detection of Parkinson's Disease Using Multimodal Data. Int J Drug Deliv Technol. 2026;16(2s): 300-305; DOI: 10.25258/ijddt.16.300-305