Cutting-edge transformer-based deep learning for Parkinson's disease diagnosing using multimodal hand-drawing and handwriting analysis

Theyazn H.H Aldhyani; Theyazn H.H Aldhyani

doi:10.3934/Neuroscience.2026015

AIMS Neuroscience

2026, Volume 13, Issue 2: 342-370. doi: 10.3934/Neuroscience.2026015

Previous Article Next Article

Research article

Cutting-edge transformer-based deep learning for Parkinson's disease diagnosing using multimodal hand-drawing and handwriting analysis

Theyazn H.H Aldhyani ^,

Applied College, King Faisal University, P.O. Box 400, Al-Ahsa 31982, Saudi Arabia

Received: 03 March 2026 Revised: 07 May 2026 Accepted: 13 May 2026 Published: 16 June 2026

Parkinson's disease (PD) is a significant mental health condition, and patients greatly benefit from prompt diagnosis and treatment if the disease is identified early. One powerful method for diagnosing PD at an early stage is the analysis of hand-drawing and handwriting samples from individuals with PD. The novelty of this research lies in developing a handwriting and hand-drawing Parkinson's Disease (HwdPD) framework for detecting PD. This framework utilizes hand-drawing and Arabic handwriting samples, which have been observed to be effective in detecting PD. Based on VGG19 and transformer, the proposed framework was tested using a real standard dataset containing 63 hand-drawn images, namely spiral, wave, and ellipse samples, with 30 samples from PD and 33 from healthy patients. The dataset also contains Arabic handwriting samples, namely “eight” and hello (“لو”). These images were processed by using augmentation to enhance the performance of the HwdPD framework. This enhancement of the image area was fed to the classification algorithm (transformer ViT-B16 and VGG19). ViT-B16 scored high accuracy, with 100% in spiral, wave, and ellipse images. In handwriting samples (“eight”), the system successfully achieved a high percentage of 100%. This system emphasizes the potential to improve diagnostic accuracy and assist clinical decision-making by evaluating its performance on these datasets. The HwdPD framework demonstrates potential for identifying PD biomarkers, which may lead to improved diagnostics.
- Parkinson's diagnosis,
- prediction,
- transformers,
- deep learning,
- artificial intelligence
Citation: Theyazn H.H Aldhyani. Cutting-edge transformer-based deep learning for Parkinson's disease diagnosing using multimodal hand-drawing and handwriting analysis[J]. AIMS Neuroscience, 2026, 13(2): 342-370. doi: 10.3934/Neuroscience.2026015

Related Papers:

Abstract

Parkinson's disease (PD) is a significant mental health condition, and patients greatly benefit from prompt diagnosis and treatment if the disease is identified early. One powerful method for diagnosing PD at an early stage is the analysis of hand-drawing and handwriting samples from individuals with PD. The novelty of this research lies in developing a handwriting and hand-drawing Parkinson's Disease (HwdPD) framework for detecting PD. This framework utilizes hand-drawing and Arabic handwriting samples, which have been observed to be effective in detecting PD. Based on VGG19 and transformer, the proposed framework was tested using a real standard dataset containing 63 hand-drawn images, namely spiral, wave, and ellipse samples, with 30 samples from PD and 33 from healthy patients. The dataset also contains Arabic handwriting samples, namely “eight” and hello (“لو”). These images were processed by using augmentation to enhance the performance of the HwdPD framework. This enhancement of the image area was fed to the classification algorithm (transformer ViT-B16 and VGG19). ViT-B16 scored high accuracy, with 100% in spiral, wave, and ellipse images. In handwriting samples (“eight”), the system successfully achieved a high percentage of 100%. This system emphasizes the potential to improve diagnostic accuracy and assist clinical decision-making by evaluating its performance on these datasets. The HwdPD framework demonstrates potential for identifying PD biomarkers, which may lead to improved diagnostics.

Acknowledgments

This work was supported by the Deanship of Scientific Research, Vice Presidency for Graduate Studies and Scientific Research, King Faisal University, Saudi Arabia [Grant No: KFU262412].

Data availability statement

The data that support the findings of this study are openly available at https://ieee-dataport.org/documents/online-offline-iraquian-hand-drawing-dataset-early-parkinsons-disease-detection.

Conflict of interest

The author declares no conflict of interest.

References

[1]	Goetz CG (2011) The history of Parkinson's disease: early clinical descriptions and neurological therapies. Cold Spring Harb Perspect Med 1: a008862. https://doi.org/10.1101/cshperspect.a008862
[2]	Vassar SD, Bordelon YM, Hays RD, et al. (2012) Confirmatory factor analysis of the motor unified Parkinson's disease rating scale. Parkinsons Dis 2012: 719167. https://doi.org/10.1155/2012/719167
[3]	World Health OrganizationParkinson disease (2023). [cited 2026 March 03]. Available from: https://www.who.int/news-room/fact-sheets/detail/parkinson-disease
[4]	National Institute of Environmental Health SciencesNeurodegenerative diseases (2022). [cited 2026 March 03]. Available from: https://www.niehs.nih.gov/research/supported/health/neurodegenerative
[5]	Mathur S, Gawas C, Ahmad IZ, et al. (2023) Neurodegenerative disorders: Assessing the impact of natural vs drug-induced treatment options. Aging Med (Milton) 6: 82-97. https://doi.org/10.1002/agm2.12243
[6]	Shaban M (2020) Deep convolutional neural network for Parkinson's disease based handwriting screening. 2020 IEEE 17th International Symposium on Biomedical Imaging Workshops (ISBI Workshops). IEEE 2020: 1-4. https://doi.org/10.1109/ISBIWorkshops50223.2020.9153407
[7]	Aldhyani THH, Al-Nefaie AH, Koundal D (2024) Modeling and diagnosis Parkinson disease by using hand drawing: deep learning model. AIMS Math 9: 6850-6877. https://doi.org/10.3934/math.2024334
[8]	Shanthappa PM, Bayari M, Abhilash GB, et al. (2025) Parkinson's disease detection using inceptionV3: A Deep learning approach. MethodsX 14: 103333. https://doi.org/10.1016/j.mex.2025.103333
[9]	Zham P, Kumar DK, Dabnichki P, et al. (2017) Distinguishing different stages of Parkinson's disease using composite index of speed and pen-pressure of sketching a spiral. Front Neurol 8: 435. https://doi.org/10.3389/fneur.2017.00435
[10]	Mercaldo F, Brunese L, Cesarelli M, et al. (2024) Spiral drawing test and explainable convolutional neural networks for Parkinson's disease detection. ICAART 2024: 443-452. https://doi.org/10.5220/0012407100003636
[11]	Sandulescu V, Ianculescu M, Valeanu L, et al. (2024) Integrating IoMT and AI for proactive healthcare: predictive models and emotion detection in neurodegenerative diseases. Algorithms 17: 376. https://doi.org/10.3390/a17090376
[12]	Impedovo D, Pirlo G (2019) Dynamic handwriting analysis for the assessment of neurodegenerative diseases: a pattern recognition perspective. IEEE Rev Biomed Eng 12: 209-220. https://doi.org/10.1109/RBME.2018.2840679
[13]	Pereira CR, Pereira DR, Rosa GH, et al. (2018) Handwritten dynamics assessment through convolutional neural networks: An application to Parkinson's disease identification. Artif Intell Med 87: 67-77. https://doi.org/10.1016/j.artmed.2018.04.001
[14]	Gil-Martín M, Montero JM, San-Segundo R (2019) Parkinson's disease detection from drawing movements using convolutional neural networks. Electronics 8: 907. https://doi.org/10.3390/electronics8080907
[15]	Kamran I, Naz S, Razzak I, et al. (2021) Handwriting dynamics assessment using deep neural network for early identification of Parkinson's disease. Futur Gener Comp Syst 117: 234-244. https://doi.org/10.1016/j.future.2020.11.020
[16]	Drotár P, Mekyska J, Rektorová I, et al. (2016) Evaluation of handwriting kinematics and pressure for differential diagnosis of Parkinson's disease. Artif Intell Med 67: 39-46. https://doi.org/10.1016/j.artmed.2016.01.004
[17]	Pereira CR, Pereira DR, Silva FA, et al. (2016) A new computer vision-based approach to aid the diagnosis of Parkinson's disease. Comput Methods Programs Biomed 136: 79-88. https://doi.org/10.1016/j.cmpb.2016.08.005
[18]	Li Z, Yang J, Wang Y, et al. (2022) Early diagnosis of Parkinson's disease using continuous convolution network: handwriting recognition based on off-line hand drawing without template. J Biomed Inform 130: 104085. https://doi.org/10.1016/j.jbi.2022.104085
[19]	Galaz Z, Drotar P, Mekyska J, et al. (2022) Comparison of CNN-learned vs. handcrafted features for detection of Parkinson's disease dysgraphia in a multilingual dataset. Front Neuroinform 16: 877139. https://doi.org/10.3389/fninf.2022.877139
[20]	Gazda M, Hireš M, Drotár P (2021) Multiple-fine-tuned convolutional neural networks for Parkinson's disease diagnosis from offline handwriting. IEEE Trans Syst Man Cybern Syst 52: 78-89. https://doi.org/10.1109/TSMC.2020.3048892
[21]	Islam MS, Parvez I, Deng H, et al. (2014) Performance comparison of heterogeneous classifiers for detection of Parkinson's disease using voice disorder (dysphonia). 2014 International Conference on Informatics, Electronics & Vision (ICIEV). IEEE 2014: 1-7. https://doi.org/10.1109/ICIEV.2014.6850849
[22]	Senturk ZK (2020) Early diagnosis of Parkinson's disease using machine learning algorithms. Med Hypotheses 138: 109603. https://doi.org/10.1016/j.mehy.2020.109603
[23]	Polat K (2012) Classification of Parkinson's disease using feature weighting method on the basis of fuzzy C-means clustering. Int J Sys Sci 43: 597-609. https://doi.org/10.1080/00207721.2011.581395
[24]	Pereira CR, Weber SAT, Hook C, et al. (2016) Deep learning-aided Parkinson's disease diagnosis from handwritten dynamics. 2016 29th SIBGRAPI conference on graphics, patterns and images (SIBGRAPI). IEEE 2016: 340-346. https://doi.org/10.1109/SIBGRAPI.2016.054
[25]	Taleb C, Khachab M, Mokbel C, et al. (2017) Feature selection for an improved Parkinson's disease identification based on handwriting. 2017 1st International Workshop on Arabic Script Analysis and Recognition (ASAR). IEEE 2017: 52-56. https://doi.org/10.1109/ASAR.2017.8067759
[26]	Xu S, Pan Z (2020) A novel ensemble of random forest for assisting diagnosis of Parkinson's disease on small handwritten dynamics dataset. Int J Med Inform 144: 104283. https://doi.org/10.1016/j.ijmedinf.2020.104283
[27]	Oh SL, Hagiwara Y, Raghavendra U, et al. (2020) A deep learning approach for Parkinson's disease diagnosis from EEG signals. Neural Comput Applic 32: 10927-10933. https://doi.org/10.1007/s00521-018-3689-5
[28]	Zahid L, Maqsood M, Durrani MY, et al. (2020) A spectrogram-based deep feature assisted computer-aided diagnostic system for Parkinson's disease. IEEE Access 8: 35482-35495. https://doi.org/10.1109/ACCESS.2020.2974008
[29]	Solana-Lavalle G, Rosas-Romero R (2021) Classification of PPMI MRI scans with voxel-based morphometry and machine learning to assist in the diagnosis of Parkinson's disease. Comput Methods Progr Biomed 198: 105793. https://doi.org/10.1016/j.cmpb.2020.105793
[30]	Shamrat FJM, Asaduzzaman M, Rahman AS, et al. (2019) A comparative analysis of parkinson disease prediction using machine learning approaches. Int J Sci Technol Res 8: 2576-2580.
[31]	Lee SH, Lim JS (2012) Parkinson's disease classification using gait characteristics and wavelet-based feature extraction. Expert Syst Appl 39: 7338-7344. https://doi.org/10.1016/j.eswa.2012.01.084
[32]	Babu GS, Suresh S, Mahanand BS (2014) A novel PBL-McRBFN-RFE approach for identification of critical brain regions responsible for Parkinson's disease. Expert Syst Appl 41: 478-488. https://doi.org/10.1016/j.eswa.2013.07.073
[33]	Hemmerling D, Sztaho D (2019) Parkinson's disease classification based on vowel sound. Models and Analysis of Vocal Emissions for Biomedical Applications .
[34]	Moetesum M, Siddiqi I, Vincent N, et al. (2019) Assessing visual attributes of handwriting for prediction of neurological disorders—A case study on Parkinson's disease. Pattern Recognit Lett 121: 19-27. https://doi.org/10.1016/j.patrec.2018.04.008
[35]	Diaz M, Ferrer MA, Impedovo D, et al. (2019) Dynamically enhanced static handwriting representation for Parkinson's disease detection. Pattern Recognit Lett 128: 204-210. https://doi.org/10.1016/j.patrec.2019.08.018
[36]	Ali L, Zhu C, Golilarz NA, et al. (2019) Reliable Parkinson's disease detection by analyzing handwritten drawings: construction of an unbiased cascaded learning system based on feature selection and adaptive boosting model. IEEE Access 7: 116480-116489. https://doi.org/10.1109/ACCESS.2019.2932037
[37]	Gupta D, Julka A, Jain S, et al. (2018) Optimized cuttlefish algorithm for diagnosis of Parkinson's disease. Cognit Syst Res 52: 36-48. https://doi.org/10.1016/j.cogsys.2018.06.006
[38]	Allebawi MF, Dhieb T, Neji M, et al. (2024) Parkinson's disease detection from online handwriting based on beta-elliptical approach and fuzzy perceptual detector. IEEE Access 12: 56936-56950. https://doi.org/10.1109/ACCESS.2024.3387342

Reader Comments

Your name:*

Email:*
© 2026 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)