MedSpectralNet: A lightweight convolutional neural network architecture for multi-modal image classification
PLOS One · 27.04.2026 · 21 (4) · e0346128
DOI: 10.1371/journal.pone.0346128 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, online erschienen am 27.04.2026. Verfasst von Nabilah Afrin und 7 weiteren Autor:innen. Beteiligte Einrichtung: Innovative Skills Ltd., Dhaka, Bangladesh. Hauptthemen laut MEDLINE: Image Processing, Computer-Assisted, Neural Networks, Computer. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 48 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Princess Nourah Bint Abdulrahman University.
Abstract (Originalsprache)
Medical image classification requires models that effectively capture both fine-grained local patterns and global anatomical structures while maintaining computational efficiency for clinical deployment. Although state-of-the-art models such as MedMamba utilize State-Space Models (SSMs) to balance accuracy and efficiency, their sequential operations limit parallelism and increase runtime. To overcome these limitations, we propose MedSpectralNet, a lightweight Convolutional Neural Network (CNN) architecture that approximates self-attention with linear complexity to efficiently extract multi-frequency features. The model introduces a dual-stream feature extractor that processes global and local information in parallel, and a ContextGate block that adaptively fuses multi-scale representations. MedSpectralNet is evaluated across six benchmark datasets from MedMNIST (including BloodMNIST, BreastMNIST, DermaMNIST, PneumoniaMNIST, OrganCMNIST, and OrganSMNIST), MedSpectralNet achieves an average accuracy of 93.7% on OrganCMNIST and 98.0% on BloodMNIST, showing 1–4.3% relative accuracy gains when compared to larger transformer-based models. Importantly, it delivers this performance with only 8.5 million parameters, representing approximately 60% fewer parameters than MedMamba-T, which requires 14.5 million parameters. MedSpectralNet has also achieved high AUC values up to 0.999 across multiple classes, demonstrating state-of-the-art accuracy with substantially reduced computational cost and improved parallelization, which makes MedSpectralNet well-suited for real-time and resource-constrained classification-based medical applications.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Innovative Skills Ltd., Dhaka, Bangladesh.
- Department of Communications and Computer Engineering, Faculty of Engineering, Al-Ahliyya Amman University, Amman, Jordan.
- Department of Electrical Engineering, College of Engineering, Princess Nourah bint Abdulrahman University, P.O. Box 84428, Riyadh 11671, Saudi Arabia.
- CS Department, Mutah University, Karak, Jordan.
- College of Engineering and Technology, American University of the Middle East, Egaila, Kuwait.
ORCID-Profile: Wasan Alamro, Yazan M Allawi
Förderung
- Princess Nourah Bint Abdulrahman University
Zitieren
APA
Afrin, N., Fahim, M. A. I., Alamro, W., Allawi, Y. M., Abadleh, A., Sultan, S. M., Elbasi, E., & Zreikat, A. I. (2026). MedSpectralNet: A lightweight convolutional neural network architecture for multi-modal image classification. PLOS One, 21(4), e0346128. https://doi.org/10.1371/journal.pone.0346128
Vancouver
Afrin N, Fahim MAI, Alamro W, Allawi YM, Abadleh A, Sultan SM, et al. MedSpectralNet: A lightweight convolutional neural network architecture for multi-modal image classification. PLOS One. 2026;21(4):e0346128. doi:10.1371/journal.pone.0346128
BibTeX
@article{Afrin2026,
author = {Afrin, Nabilah and Fahim, Masud An-Nur Islam and Alamro, Wasan and Allawi, Yazan M and Abadleh, Ahmad and Sultan, Salman Md and Elbasi, Ersin and Zreikat, Aymen I},
title = {MedSpectralNet: A lightweight convolutional neural network architecture for multi-modal image classification},
journal = {PLOS One},
year = {2026},
volume = {21},
number = {4},
pages = {e0346128},
doi = {10.1371/journal.pone.0346128}
}
EUVIMED