EUVIMEDEUVIMED entdecken
Crossref

A Vision Transformer-Based Approach for Human Organ Identification across Multiple Medical Imaging Modalities

Sheik Imran, N Pradeep

International Journal of Computing · 30.06.2026 · 324-333

DOI: 10.47839/ijc.25.2.4658 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Computing, online erschienen am 30.06.2026. Verfasst von Sheik Imran und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Modern healthcare relies extensively on medical imaging, which helps with disease monitoring, diagnosis, and therapy planning. In many medical conditions, diseases start in one organ and spread to multiple organs (e.g., metastasis). Organ identification ensures that each organ is independently analyzed for associated abnormalities, offering a comprehensive diagnostic solution. In medical research and diagnostic procedures, the categorization of medical images is of utmost importance, especially for organs like the brain, breast, lungs, and liver are frequently mentioned in cancer reports. Computed Tomography (CT), Magnetic Resonance Imaging (MRI), Breast Mammograms (MIAS), Brain (Br35H), Lung (IQ-OTHNCCD), and Liver (LiTs) are among the datasets used. ViT demonstrates self-attention, allowing each part of the image to weigh its importance relative to other parts and capture long-range dependency between input sequences. The model utilizes the Adam optimizer with a learning rate of 0.0001, achieving strong performance across multiple imaging types. In validation of the aforementioned datasets, the model’s accuracy for the brain, breast, lung, and liver is 96%, 94%, 93%, and 92%, respectively. The results indicate that ViT can minimize handcrafted feature extraction and achieve state-of-the-art performance by applying its attention mechanism to medical imaging applications.

Zitieren

APA

Imran, S., & Pradeep, N. (2026). A Vision Transformer-Based Approach for Human Organ Identification across Multiple Medical Imaging Modalities. International Journal of Computing, 324-333. https://doi.org/10.47839/ijc.25.2.4658

Vancouver

Imran S, Pradeep N. A Vision Transformer-Based Approach for Human Organ Identification across Multiple Medical Imaging Modalities. International Journal of Computing. 2026:324-333. doi:10.47839/ijc.25.2.4658

BibTeX
@article{Imran2026,
  author = {Imran, Sheik and Pradeep, N},
  title = {A Vision Transformer-Based Approach for Human Organ Identification across Multiple Medical Imaging Modalities},
  journal = {International Journal of Computing},
  year = {2026},
  pages = {324--333},
  doi = {10.47839/ijc.25.2.4658}
}
Kostenlos starten →