MedNeXt for accurate medical image classification and segmentation: A lightweight transformer-style convolutional neural network
PLOS One · 05.01.2026 · 21 (1) · e0340108
DOI: 10.1371/journal.pone.0340108 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, online erschienen am 05.01.2026. Verfasst von Ziqing Xue und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Clinical Medicine, Hebei University, Baoding, Hebei, P.R. China. Hauptthemen laut MEDLINE: Neural Networks, Computer, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 58 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Beijing Natural Science Foundation.
Abstract (Originalsprache)
Transformer-based deep learning architectures have achieved notable success across various medical image analysis tasks, driven by the global modeling capabilities of the self-attention mechanism. However, Transformer-based methods exhibit significant computational complexity and a large number of parameters, rendering them challenging to apply effectively in practical medical scenarios. Compared with Transformers, large-kernel Convolutional Neural Networks (CNNs) and Multi-Layer Perceptrons (MLPs) offer more efficient inference while retaining global contextual awareness. Therefore, we rethink the role of large-kernel CNNs and MLPs in medical image analysis and leverage them to replace the heavy self-attention operation, to strike a better balance between performance and efficiency. Specifically, we propose backbone models for medical image classification and segmentation, featured by three lightweight modules: Linear Attention Feed Forward Network (FFN) for enhancing lesion features, Spatial Encoding Module for integrating multi-scale lesion information, and Smooth Depth-Wise Convolution (DwConv) FFN for efficient interaction of channel features. Composed solely of lightweight convolutional and MLP operations, our method achieves a better balance between performance and efficiency, validated by the superior performances on five datasets with varying data scales and diseases, with 98.39% on SARS-COV2-CT-Scan, 98.12% on Monkeypox Skin Lesion Dataset, 98.58% on Large COVID-19-CT scan slice, 79.45% on Synapse and 91.28% on ACDC. The low computational cost, high-performance with limited training data, and generalizability to various of medical tasks make the proposed method a promising and practical solution for medical image classification and segmentation.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- School of Clinical Medicine, Hebei University, Baoding, Hebei, P.R. China.
- Affiliated Hospital of Hebei University, Baoding, Hebei, P.R. China.
- School of Intelligence Science and Technology, University of Science and Technology Beijing, Beijing, P.R. China.
- Department of Neurosurgery, Jiangxi Provincial People's Hospital, The First Affiliated Hospital of Nanchang Medical College, Nanchang, Jiangxi, P.R. China.
- Department of Toxicology and Sanitary Chemistry, School of Public Health, Capital Medical University, Beijing, P.R. China.
ORCID-Profile: Ziyi Liu
Förderung
- National Natural Science Foundation of China
- Beijing Natural Science Foundation
Zitieren
APA
Xue, Z., Pi, P., Liu, Z., Zeng, Z., & Sun, Z. (2026). MedNeXt for accurate medical image classification and segmentation: A lightweight transformer-style convolutional neural network. PLOS One, 21(1), e0340108. https://doi.org/10.1371/journal.pone.0340108
Vancouver
Xue Z, Pi P, Liu Z, Zeng Z, Sun Z. MedNeXt for accurate medical image classification and segmentation: A lightweight transformer-style convolutional neural network. PLOS One. 2026;21(1):e0340108. doi:10.1371/journal.pone.0340108
BibTeX
@article{Xue2026,
author = {Xue, Ziqing and Pi, Pengpeng and Liu, Ziyi and Zeng, Zhaomu and Sun, Zhiwei},
title = {MedNeXt for accurate medical image classification and segmentation: A lightweight transformer-style convolutional neural network},
journal = {PLOS One},
year = {2026},
volume = {21},
number = {1},
pages = {e0340108},
doi = {10.1371/journal.pone.0340108}
}
EUVIMED