Breast Cancer Molecular Subtype Prediction on Pathological Images with Discriminative Patch Selection and Multi-Instance Learning
Molecular subtypes of breast cancer are important references to personalized clinical treatment. For cost and labor savings, only one of the patient’s paraffin blocks is usually selected for subsequent immunohistochemistry (IHC) to obtain molecular subtypes. Inevitable block sampling error is risky...
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-04-01
|
| Series: | Frontiers in Oncology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fonc.2022.858453/full |
| _version_ | 1818061652452442112 |
|---|---|
| author | Hong Liu Wen-Dong Xu Wen-Dong Xu Zi-Hao Shang Zi-Hao Shang Xiang-Dong Wang Hai-Yan Zhou Ke-Wen Ma Huan Zhou Jia-Lin Qi Jia-Rui Jiang Li-Lan Tan Hui-Min Zeng Hui-Juan Cai Kuan-Song Wang Kuan-Song Wang Yue-Liang Qian |
| author_facet | Hong Liu Wen-Dong Xu Wen-Dong Xu Zi-Hao Shang Zi-Hao Shang Xiang-Dong Wang Hai-Yan Zhou Ke-Wen Ma Huan Zhou Jia-Lin Qi Jia-Rui Jiang Li-Lan Tan Hui-Min Zeng Hui-Juan Cai Kuan-Song Wang Kuan-Song Wang Yue-Liang Qian |
| author_sort | Hong Liu |
| collection | DOAJ |
| description | Molecular subtypes of breast cancer are important references to personalized clinical treatment. For cost and labor savings, only one of the patient’s paraffin blocks is usually selected for subsequent immunohistochemistry (IHC) to obtain molecular subtypes. Inevitable block sampling error is risky due to the tumor heterogeneity and could result in a delay in treatment. Molecular subtype prediction from conventional H&E pathological whole slide images (WSI) using the AI method is useful and critical to assist pathologists to pre-screen proper paraffin block for IHC. It is a challenging task since only WSI-level labels of molecular subtypes from IHC can be obtained without detailed local region information. Gigapixel WSIs are divided into a huge amount of patches to be computationally feasible for deep learning, while with coarse slide-level labels, patch-based methods may suffer from abundant noise patches, such as folds, overstained regions, or non-tumor tissues. A weakly supervised learning framework based on discriminative patch selection and multi-instance learning was proposed for breast cancer molecular subtype prediction from H&E WSIs. Firstly, co-teaching strategy using two networks was adopted to learn molecular subtype representations and filter out some noise patches. Then, a balanced sampling strategy was used to handle the imbalance in subtypes in the dataset. In addition, a noise patch filtering algorithm that used local outlier factor based on cluster centers was proposed to further select discriminative patches. Finally, a loss function integrating local patch with global slide constraint information was used to fine-tune MIL framework on obtained discriminative patches and further improve the prediction performance of molecular subtyping. The experimental results confirmed the effectiveness of the proposed AI method and our models outperformed even senior pathologists, which has the potential to assist pathologists to pre-screen paraffin blocks for IHC in clinic. |
| first_indexed | 2024-12-10T13:51:43Z |
| format | Article |
| id | doaj.art-2a042e3177e34a58b5b3f1de3587663e |
| institution | Directory Open Access Journal |
| issn | 2234-943X |
| language | English |
| last_indexed | 2024-12-10T13:51:43Z |
| publishDate | 2022-04-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Oncology |
| spelling | doaj.art-2a042e3177e34a58b5b3f1de3587663e2022-12-22T01:46:07ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2022-04-011210.3389/fonc.2022.858453858453Breast Cancer Molecular Subtype Prediction on Pathological Images with Discriminative Patch Selection and Multi-Instance LearningHong Liu0Wen-Dong Xu1Wen-Dong Xu2Zi-Hao Shang3Zi-Hao Shang4Xiang-Dong Wang5Hai-Yan Zhou6Ke-Wen Ma7Huan Zhou8Jia-Lin Qi9Jia-Rui Jiang10Li-Lan Tan11Hui-Min Zeng12Hui-Juan Cai13Kuan-Song Wang14Kuan-Song Wang15Yue-Liang Qian16Beijing Key Laboratory of Mobile Computing and Pervasive Device, Institute of Computing Technology, Chinese Academy of Sciences, Beijing, ChinaBeijing Key Laboratory of Mobile Computing and Pervasive Device, Institute of Computing Technology, Chinese Academy of Sciences, Beijing, ChinaUniversity of Chinese Academy of Sciences, Beijing, ChinaBeijing Key Laboratory of Mobile Computing and Pervasive Device, Institute of Computing Technology, Chinese Academy of Sciences, Beijing, ChinaUniversity of Chinese Academy of Sciences, Beijing, ChinaBeijing Key Laboratory of Mobile Computing and Pervasive Device, Institute of Computing Technology, Chinese Academy of Sciences, Beijing, ChinaDepartment of Pathology, Xiangya Hospital, Central South University, Changsha, ChinaDepartment of Pathology, Xiangya Hospital, Central South University, Changsha, ChinaDepartment of Pathology, Xiangya Hospital, Central South University, Changsha, ChinaDepartment of Pathology, Xiangya Hospital, Central South University, Changsha, ChinaDepartment of Pathology, Xiangya Hospital, Central South University, Changsha, ChinaDepartment of Pathology, Xiangya Hospital, Central South University, Changsha, ChinaDepartment of Pathology, Xiangya Hospital, Central South University, Changsha, ChinaDepartment of Pathology, Xiangya Hospital, Central South University, Changsha, ChinaDepartment of Pathology, Xiangya Hospital, Central South University, Changsha, ChinaSchool of Basic Medical Science, Central South University, Changsha, ChinaBeijing Key Laboratory of Mobile Computing and Pervasive Device, Institute of Computing Technology, Chinese Academy of Sciences, Beijing, ChinaMolecular subtypes of breast cancer are important references to personalized clinical treatment. For cost and labor savings, only one of the patient’s paraffin blocks is usually selected for subsequent immunohistochemistry (IHC) to obtain molecular subtypes. Inevitable block sampling error is risky due to the tumor heterogeneity and could result in a delay in treatment. Molecular subtype prediction from conventional H&E pathological whole slide images (WSI) using the AI method is useful and critical to assist pathologists to pre-screen proper paraffin block for IHC. It is a challenging task since only WSI-level labels of molecular subtypes from IHC can be obtained without detailed local region information. Gigapixel WSIs are divided into a huge amount of patches to be computationally feasible for deep learning, while with coarse slide-level labels, patch-based methods may suffer from abundant noise patches, such as folds, overstained regions, or non-tumor tissues. A weakly supervised learning framework based on discriminative patch selection and multi-instance learning was proposed for breast cancer molecular subtype prediction from H&E WSIs. Firstly, co-teaching strategy using two networks was adopted to learn molecular subtype representations and filter out some noise patches. Then, a balanced sampling strategy was used to handle the imbalance in subtypes in the dataset. In addition, a noise patch filtering algorithm that used local outlier factor based on cluster centers was proposed to further select discriminative patches. Finally, a loss function integrating local patch with global slide constraint information was used to fine-tune MIL framework on obtained discriminative patches and further improve the prediction performance of molecular subtyping. The experimental results confirmed the effectiveness of the proposed AI method and our models outperformed even senior pathologists, which has the potential to assist pathologists to pre-screen paraffin blocks for IHC in clinic.https://www.frontiersin.org/articles/10.3389/fonc.2022.858453/fullpathological imageweakly supervised learningmolecular subtypebreast cancerH&E |
| spellingShingle | Hong Liu Wen-Dong Xu Wen-Dong Xu Zi-Hao Shang Zi-Hao Shang Xiang-Dong Wang Hai-Yan Zhou Ke-Wen Ma Huan Zhou Jia-Lin Qi Jia-Rui Jiang Li-Lan Tan Hui-Min Zeng Hui-Juan Cai Kuan-Song Wang Kuan-Song Wang Yue-Liang Qian Breast Cancer Molecular Subtype Prediction on Pathological Images with Discriminative Patch Selection and Multi-Instance Learning Frontiers in Oncology pathological image weakly supervised learning molecular subtype breast cancer H&E |
| title | Breast Cancer Molecular Subtype Prediction on Pathological Images with Discriminative Patch Selection and Multi-Instance Learning |
| title_full | Breast Cancer Molecular Subtype Prediction on Pathological Images with Discriminative Patch Selection and Multi-Instance Learning |
| title_fullStr | Breast Cancer Molecular Subtype Prediction on Pathological Images with Discriminative Patch Selection and Multi-Instance Learning |
| title_full_unstemmed | Breast Cancer Molecular Subtype Prediction on Pathological Images with Discriminative Patch Selection and Multi-Instance Learning |
| title_short | Breast Cancer Molecular Subtype Prediction on Pathological Images with Discriminative Patch Selection and Multi-Instance Learning |
| title_sort | breast cancer molecular subtype prediction on pathological images with discriminative patch selection and multi instance learning |
| topic | pathological image weakly supervised learning molecular subtype breast cancer H&E |
| url | https://www.frontiersin.org/articles/10.3389/fonc.2022.858453/full |
| work_keys_str_mv | AT hongliu breastcancermolecularsubtypepredictiononpathologicalimageswithdiscriminativepatchselectionandmultiinstancelearning AT wendongxu breastcancermolecularsubtypepredictiononpathologicalimageswithdiscriminativepatchselectionandmultiinstancelearning AT wendongxu breastcancermolecularsubtypepredictiononpathologicalimageswithdiscriminativepatchselectionandmultiinstancelearning AT zihaoshang breastcancermolecularsubtypepredictiononpathologicalimageswithdiscriminativepatchselectionandmultiinstancelearning AT zihaoshang breastcancermolecularsubtypepredictiononpathologicalimageswithdiscriminativepatchselectionandmultiinstancelearning AT xiangdongwang breastcancermolecularsubtypepredictiononpathologicalimageswithdiscriminativepatchselectionandmultiinstancelearning AT haiyanzhou breastcancermolecularsubtypepredictiononpathologicalimageswithdiscriminativepatchselectionandmultiinstancelearning AT kewenma breastcancermolecularsubtypepredictiononpathologicalimageswithdiscriminativepatchselectionandmultiinstancelearning AT huanzhou breastcancermolecularsubtypepredictiononpathologicalimageswithdiscriminativepatchselectionandmultiinstancelearning AT jialinqi breastcancermolecularsubtypepredictiononpathologicalimageswithdiscriminativepatchselectionandmultiinstancelearning AT jiaruijiang breastcancermolecularsubtypepredictiononpathologicalimageswithdiscriminativepatchselectionandmultiinstancelearning AT lilantan breastcancermolecularsubtypepredictiononpathologicalimageswithdiscriminativepatchselectionandmultiinstancelearning AT huiminzeng breastcancermolecularsubtypepredictiononpathologicalimageswithdiscriminativepatchselectionandmultiinstancelearning AT huijuancai breastcancermolecularsubtypepredictiononpathologicalimageswithdiscriminativepatchselectionandmultiinstancelearning AT kuansongwang breastcancermolecularsubtypepredictiononpathologicalimageswithdiscriminativepatchselectionandmultiinstancelearning AT kuansongwang breastcancermolecularsubtypepredictiononpathologicalimageswithdiscriminativepatchselectionandmultiinstancelearning AT yueliangqian breastcancermolecularsubtypepredictiononpathologicalimageswithdiscriminativepatchselectionandmultiinstancelearning |