IRO Journals
Journal of Information Technology and Digital World

Multimodel Deep Learning Framework for Early Detection of Parkinson's Disease

Open Access
Published: 05 June, 2026
Pages: 140-154
  • Sheebha S.N. , Sheebha S.N.
    Assistant Professor, Department of Computer Science and Engineering, Dr.G.U.Pope College of Engineering, Thoothukudi, India
    Assistant Professor, Department of Computer Science and Engineering, Dr.G.U.Pope College of Engineering, Thoothukudi, India
  • Harini S. , Harini S.
    Student, Department of Computer Science and Engineering, Dr.G.U.Pope College of Engineering, Thoothukudi, India
    Student, Department of Computer Science and Engineering, Dr.G.U.Pope College of Engineering, Thoothukudi, India
  • Agnes Belsia W. Agnes Belsia W.
    Student, Department of Computer Science and Engineering, Dr.G.U.Pope College of Engineering, Thoothukudi, India
    Student, Department of Computer Science and Engineering, Dr.G.U.Pope College of Engineering, Thoothukudi, India
view PDF
view PDF

How to Cite

S.N., Sheebha, Harini S., and Agnes Belsia W. 2026. “Multimodel Deep Learning Framework for Early Detection of Parkinson’s Disease”. Journal of Information Technology and Digital World 8 (3): 140-54. https://doi.org/10.36548/jitdw.2026.3.001.
ACM ACS APA ABNT Chicago Harvard IEEE MLA Turabian Vancouver AMA

Download Citation

Endnote/Zotero/Mendeley (RIS) BibTeX

Keywords

Parkinson’s Disease
MRI Analysis
Spiral Drawing
Multimodal Fusion
VGG19
SVM
AdaBoost
Random Forest
Feature Extraction
Deep Learning
Early Detection

Abstract

Parkinson’s Disease (PD) is a progressively occurring neurodegenerative disease that impacts motor functions and coordination. However, early diagnosis of PD is extremely important, and the current diagnosis methods available do not have any level of comprehensiveness or objectivity. This study explores the use of a multimodal approach to detect Parkinson’s disease at its early stages through deep learning methods. Magnetic Resonance Imaging (MRI) images are extracted using Sobel filter, K-Means segmentation, and VGG19 feature extraction, while Spiral drawings are processed through edge detection using Canny edge detection and contour-based features. Some of the models used include SVM, AdaBoost, Random Forest, and logistic regression, and the final model uses a combination of these classifiers and a score-level fusion model of both modalities. The proposed stacked ensemble model achieved 96.41% accuracy, 96% Precision, 96% recall, 96% f1-score, and a ROC-AUC of 0.99, which is significantly better than single modality baselines.

References

  1. Jankovic, Joseph. "Parkinson’s Disease: Clinical Features and Diagnosis." Journal of neurology, neurosurgery & psychiatry 2008, vol. 79, no. 4: 368-376.
  2. Kalia, Lorraine V., and Anthony E. Lang. "Parkinson's Disease." The lancet 2015, vol. 386, no. 9996: 896-912.
  3. Pereira, Clayton R., Danillo R. Pereira, Joao P. Papa, Gustavo H. Rosa, and Xin-She Yang. "Convolutional Neural Networks Applied for Parkinson’s Disease Identification." In Machine Learning for Health Informatics: State-Of-The-Art and Future Challenges, Cham: Springer International Publishing 2016, 377-390.
  4. Prashanth, R., Sumantra Dutta Roy, Pravat K. Mandal, and Shantanu Ghosh. "High-Accuracy Classification of Parkinson's Disease Through Shape Analysis and Surface Fitting In 123I-Ioflupane SPECT Imaging." IEEE journal of biomedical and health informatics 2016, vol. 21, no. 3: 794-802.
  5. Sivaranjini, S., and C. M. Sujatha. "Deep Learning Based Diagnosis of Parkinson’s Disease Using Convolutional Neural Network." Multimedia tools and applications 2020, vol. 79, no. 21: 15467-15479.
  6. Nandpuru, Hari Babu, S. S. Salankar, and V. R. Bora. "MRI Brain Cancer Classification Using Support Vector Machine." In 2014 IEEE Students' Conference on Electrical, Electronics and Computer Science, 1-6.
  7. Little, Max, Patrick McSharry, Eric Hunter, Jennifer Spielman, and Lorraine Ramig. "Suitability of Dysphonia Measurements for Telemonitoring of Parkinson’s Disease." Nature Precedings 2008, 1-1.
  8. Das, Resul. "A Comparison of Multiple Classification Methods for Diagnosis of Parkinson Disease." Expert Systems with Applications 2010, vol. 37, no. 2: 1568-1572.
  9. Tsanas, Athanasios, Max Little, Patrick McSharry, and Lorraine Ramig. "Accurate Telemonitoring of Parkinson’s Disease Progression by Non-Invasive Speech Tests." Nature Precedings 2009, 1-1.
  10. Sakar, Betul Erdogdu, M. Erdem Isenkul, C. Okan Sakar, Ahmet Sertbas, Fikret Gurgen, Sakir Delil, Hulya Apaydin, and Olcay Kursun. "Collection and Analysis of a Parkinson Speech Dataset with Multiple Types of Sound Recordings." IEEE journal of biomedical and health informatics 2013, vol. 17, no. 4: 828-834.
  11. Dataset - https://www.kaggle.com/datasets/trainingdatapro/brain-anomaly-detection