EnNuSegNet: Enhancing Weakly Supervised Nucleus Segmentation through Feature Preservation and Edge Refinement
Nucleus segmentation plays a crucial role in tissue pathology image analysis. Despite significant progress in cell nucleus image segmentation algorithms based on fully supervised learning, the large number and small size of cell nuclei pose a considerable challenge in terms of the substantial worklo...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-01-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/13/3/504 |
| _version_ | 1797318853475172352 |
|---|---|
| author | Xiaohui Chen Qisheng Ruan Lingjun Chen Guanqun Sheng Peng Chen |
| author_facet | Xiaohui Chen Qisheng Ruan Lingjun Chen Guanqun Sheng Peng Chen |
| author_sort | Xiaohui Chen |
| collection | DOAJ |
| description | Nucleus segmentation plays a crucial role in tissue pathology image analysis. Despite significant progress in cell nucleus image segmentation algorithms based on fully supervised learning, the large number and small size of cell nuclei pose a considerable challenge in terms of the substantial workload required for label annotation. This difficulty in acquiring datasets has become exceptionally challenging. This paper proposes a novel weakly supervised nucleus segmentation method that only requires point annotations of the nuclei. The technique is an encoder–decoder network which enhances the weakly supervised nucleus segmentation performance (EnNuSegNet). Firstly, we introduce the Feature Preservation Module (FPM) in both encoder and decoder, which preserves more low-level features from the shallow layers of the network during the early stages of training while enhancing the network’s expressive capability. Secondly, we incorporate a Scale-Aware Module (SAM) in the bottleneck part of the network to improve the model’s perception of cell nuclei at different scales. Lastly, we propose a training strategy for nucleus edge regression (NER), which guides the model to optimize the segmented edges during training, effectively compensating for the loss of nucleus edge information and achieving higher-quality nucleus segmentation. Experimental results on two publicly available datasets demonstrate that our proposed method outperforms state-of-the-art approaches, with improvements of 2.02%, 1.41%, and 1.59% in terms of F1 score, Dice coefficient, and Average Jaccard Index (AJI), respectively. This indicates the effectiveness of our method in improving segmentation performance. |
| first_indexed | 2024-03-08T03:58:25Z |
| format | Article |
| id | doaj.art-2c142e9afcaa477a9a535e1949a7f069 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-08T03:58:25Z |
| publishDate | 2024-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-2c142e9afcaa477a9a535e1949a7f0692024-02-09T15:10:26ZengMDPI AGElectronics2079-92922024-01-0113350410.3390/electronics13030504EnNuSegNet: Enhancing Weakly Supervised Nucleus Segmentation through Feature Preservation and Edge RefinementXiaohui Chen0Qisheng Ruan1Lingjun Chen2Guanqun Sheng3Peng Chen4Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering, China Three Gorges University, Yichang 443002, ChinaHubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering, China Three Gorges University, Yichang 443002, ChinaYichang Power Supply Company, State Grid Hubei Electric Power Co., Ltd., Yichang 443002, ChinaHubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering, China Three Gorges University, Yichang 443002, ChinaHubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering, China Three Gorges University, Yichang 443002, ChinaNucleus segmentation plays a crucial role in tissue pathology image analysis. Despite significant progress in cell nucleus image segmentation algorithms based on fully supervised learning, the large number and small size of cell nuclei pose a considerable challenge in terms of the substantial workload required for label annotation. This difficulty in acquiring datasets has become exceptionally challenging. This paper proposes a novel weakly supervised nucleus segmentation method that only requires point annotations of the nuclei. The technique is an encoder–decoder network which enhances the weakly supervised nucleus segmentation performance (EnNuSegNet). Firstly, we introduce the Feature Preservation Module (FPM) in both encoder and decoder, which preserves more low-level features from the shallow layers of the network during the early stages of training while enhancing the network’s expressive capability. Secondly, we incorporate a Scale-Aware Module (SAM) in the bottleneck part of the network to improve the model’s perception of cell nuclei at different scales. Lastly, we propose a training strategy for nucleus edge regression (NER), which guides the model to optimize the segmented edges during training, effectively compensating for the loss of nucleus edge information and achieving higher-quality nucleus segmentation. Experimental results on two publicly available datasets demonstrate that our proposed method outperforms state-of-the-art approaches, with improvements of 2.02%, 1.41%, and 1.59% in terms of F1 score, Dice coefficient, and Average Jaccard Index (AJI), respectively. This indicates the effectiveness of our method in improving segmentation performance.https://www.mdpi.com/2079-9292/13/3/504weakly supervisednucleus segmentationpoint annotationcomputational pathology |
| spellingShingle | Xiaohui Chen Qisheng Ruan Lingjun Chen Guanqun Sheng Peng Chen EnNuSegNet: Enhancing Weakly Supervised Nucleus Segmentation through Feature Preservation and Edge Refinement Electronics weakly supervised nucleus segmentation point annotation computational pathology |
| title | EnNuSegNet: Enhancing Weakly Supervised Nucleus Segmentation through Feature Preservation and Edge Refinement |
| title_full | EnNuSegNet: Enhancing Weakly Supervised Nucleus Segmentation through Feature Preservation and Edge Refinement |
| title_fullStr | EnNuSegNet: Enhancing Weakly Supervised Nucleus Segmentation through Feature Preservation and Edge Refinement |
| title_full_unstemmed | EnNuSegNet: Enhancing Weakly Supervised Nucleus Segmentation through Feature Preservation and Edge Refinement |
| title_short | EnNuSegNet: Enhancing Weakly Supervised Nucleus Segmentation through Feature Preservation and Edge Refinement |
| title_sort | ennusegnet enhancing weakly supervised nucleus segmentation through feature preservation and edge refinement |
| topic | weakly supervised nucleus segmentation point annotation computational pathology |
| url | https://www.mdpi.com/2079-9292/13/3/504 |
| work_keys_str_mv | AT xiaohuichen ennusegnetenhancingweaklysupervisednucleussegmentationthroughfeaturepreservationandedgerefinement AT qishengruan ennusegnetenhancingweaklysupervisednucleussegmentationthroughfeaturepreservationandedgerefinement AT lingjunchen ennusegnetenhancingweaklysupervisednucleussegmentationthroughfeaturepreservationandedgerefinement AT guanqunsheng ennusegnetenhancingweaklysupervisednucleussegmentationthroughfeaturepreservationandedgerefinement AT pengchen ennusegnetenhancingweaklysupervisednucleussegmentationthroughfeaturepreservationandedgerefinement |