Enhancing medical image segmentation with HCE-Net: A heterogeneous feature collaborative extraction network
- 1School of Computer Science, Chengdu University, Chengdu, China
- 2Key Laboratory of Pattern Recognition and Intelligent Information Processing, Institutions of Higher Education of Sichuan Province, Chengdu University, Chengdu, China
- 3Key Laboratory of Digital Innovation of Tianfu Culture, Sichuan Provincial Department of Culture and Tourism, Chengdu University, Chengdu, China
Abstract
Medical image segmentation occupies a pivotal role in clinical diagnosis and therapeutic planning. Conventional segmentation techniques struggle to accommodate the intricate representational attributes of medical images, while prevailing deep learning segmentation models predominantly adopt homogeneous single encoders and linear feature fusion paradigms, thereby exhibiting inherent limitations in contextual modeling, fine-grained feature excavation, and contour segmentation accuracy. To address such dilemmas, this paper proposes a Heterogeneous Collaborative Feature Extraction Network (HCE-Net). Establishing a dual-path heterogeneous encoding architecture, this network excavates complementary feature representations in a parallel manner. An Orthogonal Fusion Module (OFM) is elaborately devised to integrate Sobel edge priors with multi-scale convolutions, enabling the precise capture of edge and contour characteristics. A Diversified Receptive Field Module (DFM) is constructed to consolidate multi-scale detail manifestations and preserve global contextual cues. Furthermore, Adaptive Aggregation Connection (AAC) is embedded within skip connection branches to supplant conventional simplistic concatenation, efficaciously mitigating the technical bottleneck of heterogeneous feature fusion. Comparative experiments conducted on three publicly accessible medical datasets, encompassing dental radiographs, pulmonary CT scans, and retinal vascular images, demonstrate that HCE-Net substantially elevates segmentation accuracy, enhances boundary integrity, and ameliorates the general robustness of the model. It manifests distinguished application prospects across diverse medical image segmentation scenarios.
Related
- MaSkel and napari-MaSkel: fast morphological skeleton feature extraction from segmentation masks
- Enhancing medical education with AI: A comparative study of various large language models in radiology
- Collaborative Outcomes Study on Health and Functioning During Infection Times (COH-FIT): Global and Risk-Group Stratified Course of Well-Being and Mental Health During the COVID-19 Pandemic in Adolescents
- Seatizen Atlas: a collaborative dataset of underwater and aerial marine imagery
- A collaborative submission model for building high-quality data resources at scale through partnership: lessons from single-cell genomics
BibTeX
@article{zhu2026enhancing,
title = {{Enhancing medical image segmentation with HCE-Net: A heterogeneous feature collaborative extraction network}},
author = {Guoqing Zhu and Hongzhi Li and Han Cui and Zhanghao Ren and Jiaxi Wang},
journal = {PLOS One},
year = {2026},
month = nov,
doi = {10.1371/journal.pone.0359398},
url = {https://doi.org/10.1371/journal.pone.0359398},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Enhancing medical image segmentation with HCE-Net: A heterogeneous feature collaborative extraction network
AU - Guoqing Zhu
AU - Hongzhi Li
AU - Han Cui
AU - Zhanghao Ren
AU - Jiaxi Wang
PY - 2026
DA - 2026/11/03
JO - PLOS One
DO - 10.1371/journal.pone.0359398
UR - https://doi.org/10.1371/journal.pone.0359398
AB - Medical image segmentation occupies a pivotal role in clinical diagnosis and therapeutic planning. Conventional segmentation techniques struggle to accommodate the intricate representational attributes of medical images, while prevailing deep learning segmentation models predominantly adopt homogeneous single encoders and linear feature fusion paradigms, thereby exhibiting inherent limitations in contextual modeling, fine-grained feature excavation, and contour segmentation accuracy. To address such dilemmas, this paper proposes a Heterogeneous Collaborative Feature Extraction Network (HCE-Net). Establishing a dual-path heterogeneous encoding architecture, this network excavates complementary feature representations in a parallel manner. An Orthogonal Fusion Module (OFM) is elaborately devised to integrate Sobel edge priors with multi-scale convolutions, enabling the precise capture of edge and contour characteristics. A Diversified Receptive Field Module (DFM) is constructed to consolidate multi-scale detail manifestations and preserve global contextual cues. Furthermore, Adaptive Aggregation Connection (AAC) is embedded within skip connection branches to supplant conventional simplistic concatenation, efficaciously mitigating the technical bottleneck of heterogeneous feature fusion. Comparative experiments conducted on three publicly accessible medical datasets, encompassing dental radiographs, pulmonary CT scans, and retinal vascular images, demonstrate that HCE-Net substantially elevates segmentation accuracy, enhances boundary integrity, and ameliorates the general robustness of the model. It manifests distinguished application prospects across diverse medical image segmentation scenarios.
SN - 1932-6203
ER -