SiameseNet based on multiple instance learning for accurate identification of the histological grade of ICC tumors
BackgroundAfter hepatocellular carcinoma (HCC), intrahepatic cholangiocarcinoma (ICC) is the second most common primary liver cancer. Timely and accurate identification of ICC histological grade is critical for guiding clinical diagnosis and treatment planning.MethodWe proposed a dual-branch deep ne...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-02-01
|
| Series: | Frontiers in Oncology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fonc.2025.1450379/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823861142100901888 |
|---|---|
| author | Zhizhan Fu Fazhi Feng Xingguang He Tongtong Li Xiansong Li Jituome Ziluo Zixing Huang Jinlin Ye |
| author_facet | Zhizhan Fu Fazhi Feng Xingguang He Tongtong Li Xiansong Li Jituome Ziluo Zixing Huang Jinlin Ye |
| author_sort | Zhizhan Fu |
| collection | DOAJ |
| description | BackgroundAfter hepatocellular carcinoma (HCC), intrahepatic cholangiocarcinoma (ICC) is the second most common primary liver cancer. Timely and accurate identification of ICC histological grade is critical for guiding clinical diagnosis and treatment planning.MethodWe proposed a dual-branch deep neural network (SiameseNet) based on multiple-instance learning and cross-attention mechanisms to address tumor heterogeneity in ICC histological grade prediction. The study included 424 ICC patients (381 in training, 43 in testing). The model integrated imaging data from two modalities through cross-attention, optimizing feature representation for grade classification.ResultsIn the testing cohort, the model achieved an accuracy of 86.0%, AUC of 86.2%, sensitivity of 84.6%, and specificity of 86.7%, demonstrating robust predictive performance.ConclusionThe proposed framework effectively mitigates performance degradation caused by tumor heterogeneity. Its high accuracy and generalizability suggest potential clinical utility in assisting histopathological assessment and personalized treatment planning for ICC patients. |
| format | Article |
| id | doaj-art-d726a165933e4f598637621b1d3f8bb8 |
| institution | Kabale University |
| issn | 2234-943X |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Oncology |
| spelling | doaj-art-d726a165933e4f598637621b1d3f8bb82025-02-10T05:16:13ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2025-02-011510.3389/fonc.2025.14503791450379SiameseNet based on multiple instance learning for accurate identification of the histological grade of ICC tumorsZhizhan Fu0Fazhi Feng1Xingguang He2Tongtong Li3Xiansong Li4Jituome Ziluo5Zixing Huang6Jinlin Ye7The Quzhou Affiliated Hospital of Wenzhou Medical University, Quzhou People’s Hospital, Quzhou, ChinaYangtze Delta Region Institute (Quzhou), University of Electronic Science and Technology of China, Quzhou, ChinaYangtze Delta Region Institute (Quzhou), University of Electronic Science and Technology of China, Quzhou, ChinaDepartment of Radiology, West China Hospital, Sichuan University, Chengdu, ChinaDepartment of Radiology, West China Hospital, Sichuan University, Chengdu, ChinaDepartment of Radiology, West China Hospital, Sichuan University, Chengdu, ChinaDepartment of Radiology, West China Hospital, Sichuan University, Chengdu, ChinaThe Quzhou Affiliated Hospital of Wenzhou Medical University, Quzhou People’s Hospital, Quzhou, ChinaBackgroundAfter hepatocellular carcinoma (HCC), intrahepatic cholangiocarcinoma (ICC) is the second most common primary liver cancer. Timely and accurate identification of ICC histological grade is critical for guiding clinical diagnosis and treatment planning.MethodWe proposed a dual-branch deep neural network (SiameseNet) based on multiple-instance learning and cross-attention mechanisms to address tumor heterogeneity in ICC histological grade prediction. The study included 424 ICC patients (381 in training, 43 in testing). The model integrated imaging data from two modalities through cross-attention, optimizing feature representation for grade classification.ResultsIn the testing cohort, the model achieved an accuracy of 86.0%, AUC of 86.2%, sensitivity of 84.6%, and specificity of 86.7%, demonstrating robust predictive performance.ConclusionThe proposed framework effectively mitigates performance degradation caused by tumor heterogeneity. Its high accuracy and generalizability suggest potential clinical utility in assisting histopathological assessment and personalized treatment planning for ICC patients.https://www.frontiersin.org/articles/10.3389/fonc.2025.1450379/fullintrahepatic cholangiocarcinomahistological grademultiple instance learningcross-attention mechanismCT-based diagnostics |
| spellingShingle | Zhizhan Fu Fazhi Feng Xingguang He Tongtong Li Xiansong Li Jituome Ziluo Zixing Huang Jinlin Ye SiameseNet based on multiple instance learning for accurate identification of the histological grade of ICC tumors Frontiers in Oncology intrahepatic cholangiocarcinoma histological grade multiple instance learning cross-attention mechanism CT-based diagnostics |
| title | SiameseNet based on multiple instance learning for accurate identification of the histological grade of ICC tumors |
| title_full | SiameseNet based on multiple instance learning for accurate identification of the histological grade of ICC tumors |
| title_fullStr | SiameseNet based on multiple instance learning for accurate identification of the histological grade of ICC tumors |
| title_full_unstemmed | SiameseNet based on multiple instance learning for accurate identification of the histological grade of ICC tumors |
| title_short | SiameseNet based on multiple instance learning for accurate identification of the histological grade of ICC tumors |
| title_sort | siamesenet based on multiple instance learning for accurate identification of the histological grade of icc tumors |
| topic | intrahepatic cholangiocarcinoma histological grade multiple instance learning cross-attention mechanism CT-based diagnostics |
| url | https://www.frontiersin.org/articles/10.3389/fonc.2025.1450379/full |
| work_keys_str_mv | AT zhizhanfu siamesenetbasedonmultipleinstancelearningforaccurateidentificationofthehistologicalgradeoficctumors AT fazhifeng siamesenetbasedonmultipleinstancelearningforaccurateidentificationofthehistologicalgradeoficctumors AT xingguanghe siamesenetbasedonmultipleinstancelearningforaccurateidentificationofthehistologicalgradeoficctumors AT tongtongli siamesenetbasedonmultipleinstancelearningforaccurateidentificationofthehistologicalgradeoficctumors AT xiansongli siamesenetbasedonmultipleinstancelearningforaccurateidentificationofthehistologicalgradeoficctumors AT jituomeziluo siamesenetbasedonmultipleinstancelearningforaccurateidentificationofthehistologicalgradeoficctumors AT zixinghuang siamesenetbasedonmultipleinstancelearningforaccurateidentificationofthehistologicalgradeoficctumors AT jinlinye siamesenetbasedonmultipleinstancelearningforaccurateidentificationofthehistologicalgradeoficctumors |