An accurate prediction of the origin for bone metastatic cancer using deep learning on digital pathological imagesResearch in context
Lianghui Zhu,
Huijuan Shi,
Huiting Wei,
Chengjiang Wang,
Shanshan Shi,
Fenfen Zhang,
Renao Yan,
Yiqing Liu,
Tingting He,
Liyuan Wang,
Junru Cheng,
Hufei Duan,
Hong Du,
Fengjiao Meng,
Wenli Zhao,
Xia Gu,
Linlang Guo,
Yingpeng Ni,
Yonghong He,
Tian Guan,
Anjia Han
Affiliations
Lianghui Zhu
Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen, Guangdong, China
Huijuan Shi
Department of Pathology, the First Affiliated Hospital, Sun Yat-sen University, Guangzhou, China
Huiting Wei
Department of Pathology, the First Affiliated Hospital, Sun Yat-sen University, Guangzhou, China
Chengjiang Wang
Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen, Guangdong, China
Shanshan Shi
Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen, Guangdong, China
Fenfen Zhang
Department of Pathology, the First Affiliated Hospital, Sun Yat-sen University, Guangzhou, China
Renao Yan
Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen, Guangdong, China
Yiqing Liu
Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen, Guangdong, China
Tingting He
Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen, Guangdong, China
Liyuan Wang
Department of Pathology, the First Affiliated Hospital, Sun Yat-sen University, Guangzhou, China
Junru Cheng
Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen, Guangdong, China
Hufei Duan
Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen, Guangdong, China
Hong Du
Department of Pathology, Guangzhou First People's Hospital, School of Medicine, South China University of Technology, Guangzhou, China
Fengjiao Meng
Department of Pathology, Zhongshan People's Hospital, Zhongshan, China
Wenli Zhao
Department of Pathology, The First People's Hospital of Huizhou, Huizhou, China
Xia Gu
Department of Pathology, the First Affiliated Hospital of Guangzhou Medical University, Guangzhou, China
Linlang Guo
Department of Pathology, Zhujiang Hospital, Southern Medical University, Guangzhou, China
Yingpeng Ni
Department of Pathology, Jieyang People's Hospital (Jieyang Affiliated Hospital, Sun Yat-Sen University), Jieyang, China
Yonghong He
Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen, Guangdong, China; Corresponding author.
Tian Guan
Institute of Biopharmaceutical and Health Engineering, Tsinghua Shenzhen International Graduate School, Shenzhen, Guangdong, China; Corresponding author.
Anjia Han
Department of Pathology, the First Affiliated Hospital, Sun Yat-sen University, Guangzhou, China; Corresponding author.
Summary: Background: Determining the origin of bone metastatic cancer (OBMC) is of great significance to clinical therapeutics. It is challenging for pathologists to determine the OBMC with limited clinical information and bone biopsy. Methods: We designed a regional multiple-instance learning algorithm to predict the OBMC based on hematoxylin-eosin (H&E) staining slides alone. We collected 1041 cases from eight different hospitals and labeled 26,431 regions of interest to train the model. The performance of the model was assessed by ten-fold cross validation and external validation. Under the guidance of top3 predictions, we conducted an IHC test on 175 cases of unknown origins to compare the consistency of the results predicted by the model and indicated by the IHC markers. We also applied the model to identify whether there was tumor or not in a region, as well as distinguishing squamous cell carcinoma, adenocarcinoma, and neuroendocrine tumor. Findings: In the within-cohort, our model achieved a top1-accuracy of 91.35% and a top3-accuracy of 97.75%. In the external cohort, our model displayed a good generalizability with a top3-accuracy of 97.44%. The top1 consistency between the results of the model and the immunohistochemistry markers was 83.90% and the top3 consistency was 94.33%. The model obtained an accuracy of 98.98% to identify whether there was tumor or not and an accuracy of 93.85% to differentiate three types of cancers. Interpretation: Our model demonstrated good performance to predict the OBMC from routine histology and had great potential for assisting pathologists with determining the OBMC accurately. Funding: National Science Foundation of China (61875102 and 61975089), Natural Science Foundation of Guangdong province (2021A15-15012379 and 2022A1515 012550), Science and Technology Research Program of Shenzhen City (JCYJ20200109110606054 and WDZC20200821141349001), and Tsinghua University Spring Breeze Fund (2020Z99CFZ023).
