EUVIMEDEUVIMED entdecken
Crossref

ViM-PD: A Bidirectional Vision Mamba Framework for Multimodal Parkinson's Disease Detection

S. Santhi, R. P. Shermy, N. Saranya, K Saranya

International Journal of Drug Delivery Technology · 30.05.2026 · 16 (37s)

DOI: 10.25258/ijddt.16.37s.54 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Drug Delivery Technology, online erschienen am 30.05.2026. Verfasst von S. Santhi und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

It is vital in the early diagnosis of parkinson's disease for the appropriate management of neurodegenerative diseases. However, traditional clinical examinations and methods may not be sensitive enough in capturing subtle changes in early stages of Parkinson's disease. A new classification of Parkinson's disease, referred to as ViMPD, has been developed. ViM-PD is a new and innovative system of classification of parkinson's. For dynamic spiral/wave drawing analysis, we use the PaHaW and NewHandPD datasets, whereas acoustic biomarker extraction is done with the UCI Parkinson's Speech dataset. In contrast to typical CNN which focus on local textures our model uses a Bidirectional Selective Scan (SS2D) approach to capture long-range global dependencies in handwriting geometry and voice prosody with linear computational cost. According to the experimental data, our multimodal ViM-PD framework achieves a peak classification accuracy of 98.2%, beating the baseline ResNet and Vision Transformer models. This excellent accuracy is due to the model's capacity to cross-correlate fine-motor tremors in handwriting and vocal dysphonia. Furthermore, the design has a 42% shorter inference latency and memory footprint, making it appropriate for use in resource-constrained telehealth situations. This work confirms ViM as a better, high-efficiency option for reliable, non-invasive PD screening.

Zitieren

APA

Santhi, S., Shermy, R. P., Saranya, N., & Saranya, K. (2026). ViM-PD: A Bidirectional Vision Mamba Framework for Multimodal Parkinson's Disease Detection. International Journal of Drug Delivery Technology, 16(37s). https://doi.org/10.25258/ijddt.16.37s.54

Vancouver

Santhi S, Shermy RP, Saranya N, Saranya K. ViM-PD: A Bidirectional Vision Mamba Framework for Multimodal Parkinson's Disease Detection. International Journal of Drug Delivery Technology. 2026;16(37s). doi:10.25258/ijddt.16.37s.54

BibTeX
@article{Santhi2026,
  author = {Santhi, S. and Shermy, R. P. and Saranya, N. and Saranya, K},
  title = {ViM-PD: A Bidirectional Vision Mamba Framework for Multimodal Parkinson's Disease Detection},
  journal = {International Journal of Drug Delivery Technology},
  year = {2026},
  volume = {16},
  number = {37s},
  doi = {10.25258/ijddt.16.37s.54}
}
Kostenlos starten →