Gastric Section Correlation Network for Gastric Precancerous Lesion Diagnosis
<italic>Goal:</italic> Diagnosing the corpus-predominant gastritis index (CGI) which is an early precancerous lesion in the stomach has been shown its effectiveness in identifying high gastric cancer risk patients for preventive healthcare. However, invasive biopsies and time-consuming p...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Open Journal of Engineering in Medicine and Biology |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10128879/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832582289949720576 |
|---|---|
| author | Jyun-Yao Jhang Yu-Ching Tsai Tzu-Chun Hsu Chun-Rong Huang Hsiu-Chi Cheng Bor-Shyang Sheu |
| author_facet | Jyun-Yao Jhang Yu-Ching Tsai Tzu-Chun Hsu Chun-Rong Huang Hsiu-Chi Cheng Bor-Shyang Sheu |
| author_sort | Jyun-Yao Jhang |
| collection | DOAJ |
| description | <italic>Goal:</italic> Diagnosing the corpus-predominant gastritis index (CGI) which is an early precancerous lesion in the stomach has been shown its effectiveness in identifying high gastric cancer risk patients for preventive healthcare. However, invasive biopsies and time-consuming pathological analysis are required for the CGI diagnosis. <italic>Methods:</italic> We propose a novel gastric section correlation network (GSCNet) for the CGI diagnosis from endoscopic images of three dominant gastric sections, the antrum, body and cardia. The proposed network consists of two dominant modules including the scaling feature fusion module and section correlation module. The front one aims to extract scaling fusion features which can effectively represent the mucosa under variant viewing angles and scale changes for each gastric section. The latter one aims to apply the medical prior knowledge with three section correlation losses to model the correlations of different gastric sections for the CGI diagnosis. <italic>Results:</italic> The proposed method outperforms competing deep learning methods and achieves high testing accuracy, sensitivity, and specificity of 0.957, 0.938 and 0.962, respectively. <italic>Conclusions:</italic> The proposed method is the first method to identify high gastric cancer risk patients with CGI from endoscopic images without invasive biopsies and time-consuming pathological analysis. |
| format | Article |
| id | doaj-art-21acf23abce141a789ae996b12c2367c |
| institution | Kabale University |
| issn | 2644-1276 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Open Journal of Engineering in Medicine and Biology |
| spelling | doaj-art-21acf23abce141a789ae996b12c2367c2025-01-30T00:03:48ZengIEEEIEEE Open Journal of Engineering in Medicine and Biology2644-12762024-01-01543444210.1109/OJEMB.2023.327721910128879Gastric Section Correlation Network for Gastric Precancerous Lesion DiagnosisJyun-Yao Jhang0Yu-Ching Tsai1Tzu-Chun Hsu2Chun-Rong Huang3https://orcid.org/0000-0003-2372-5429Hsiu-Chi Cheng4https://orcid.org/0000-0003-0954-0647Bor-Shyang Sheu5https://orcid.org/0000-0002-1500-6929Department of Computer Science and Engineering, National Chung Hsing University, Taichung, TaiwanDepartment of Internal Medicine, Tainan Hospital, Ministry of Health and Welfare, Tainan, TaiwanDepartment of Computer Science and Engineering, National Chung Hsing University, Taichung, TaiwanCross College Elite Program, and Academy of Innovative Semiconductor and Sustainable Manufacturing, National Cheng Kung University, Tainan, TaiwanDepartment of Internal Medicine, Institute of Clinical Medicine and Molecular Medicine, National Cheng Kung University, Tainan, TaiwanInstitute of Clinical Medicine and Department of Internal Medicine, National Cheng Kung University Hospital, National Cheng Kung University, Tainan, Taiwan<italic>Goal:</italic> Diagnosing the corpus-predominant gastritis index (CGI) which is an early precancerous lesion in the stomach has been shown its effectiveness in identifying high gastric cancer risk patients for preventive healthcare. However, invasive biopsies and time-consuming pathological analysis are required for the CGI diagnosis. <italic>Methods:</italic> We propose a novel gastric section correlation network (GSCNet) for the CGI diagnosis from endoscopic images of three dominant gastric sections, the antrum, body and cardia. The proposed network consists of two dominant modules including the scaling feature fusion module and section correlation module. The front one aims to extract scaling fusion features which can effectively represent the mucosa under variant viewing angles and scale changes for each gastric section. The latter one aims to apply the medical prior knowledge with three section correlation losses to model the correlations of different gastric sections for the CGI diagnosis. <italic>Results:</italic> The proposed method outperforms competing deep learning methods and achieves high testing accuracy, sensitivity, and specificity of 0.957, 0.938 and 0.962, respectively. <italic>Conclusions:</italic> The proposed method is the first method to identify high gastric cancer risk patients with CGI from endoscopic images without invasive biopsies and time-consuming pathological analysis.https://ieeexplore.ieee.org/document/10128879/Corpus-predominant gastritis indexdeep learningprecancerous lesion classificationgastric endoscopy |
| spellingShingle | Jyun-Yao Jhang Yu-Ching Tsai Tzu-Chun Hsu Chun-Rong Huang Hsiu-Chi Cheng Bor-Shyang Sheu Gastric Section Correlation Network for Gastric Precancerous Lesion Diagnosis IEEE Open Journal of Engineering in Medicine and Biology Corpus-predominant gastritis index deep learning precancerous lesion classification gastric endoscopy |
| title | Gastric Section Correlation Network for Gastric Precancerous Lesion Diagnosis |
| title_full | Gastric Section Correlation Network for Gastric Precancerous Lesion Diagnosis |
| title_fullStr | Gastric Section Correlation Network for Gastric Precancerous Lesion Diagnosis |
| title_full_unstemmed | Gastric Section Correlation Network for Gastric Precancerous Lesion Diagnosis |
| title_short | Gastric Section Correlation Network for Gastric Precancerous Lesion Diagnosis |
| title_sort | gastric section correlation network for gastric precancerous lesion diagnosis |
| topic | Corpus-predominant gastritis index deep learning precancerous lesion classification gastric endoscopy |
| url | https://ieeexplore.ieee.org/document/10128879/ |
| work_keys_str_mv | AT jyunyaojhang gastricsectioncorrelationnetworkforgastricprecancerouslesiondiagnosis AT yuchingtsai gastricsectioncorrelationnetworkforgastricprecancerouslesiondiagnosis AT tzuchunhsu gastricsectioncorrelationnetworkforgastricprecancerouslesiondiagnosis AT chunronghuang gastricsectioncorrelationnetworkforgastricprecancerouslesiondiagnosis AT hsiuchicheng gastricsectioncorrelationnetworkforgastricprecancerouslesiondiagnosis AT borshyangsheu gastricsectioncorrelationnetworkforgastricprecancerouslesiondiagnosis |