Development and validation of a deep learning model for morphological assessment of myeloproliferative neoplasms using clinical data and digital pathology.
Rong Wang, Zhongxun Shi, Yuan Zhang, Liangmin Wei, Minghui Duan, Min Xiao, Jin Wang, Suning Chen, Qian Wang, Jianyao Huang, Xiaomei Hu, Jinhong Mei, Jieyu He, Feng Chen, Lei Fan, Guanyu Yang, Wenyi Shen, Yongyue Wei, Jianyong Li
Author Information
Rong Wang: Department of Haematology, Collaborative Innovation Center for Cancer Personalized Medicine, Jiangsu Province Hospital, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China.
Zhongxun Shi: Department of Haematology, Collaborative Innovation Center for Cancer Personalized Medicine, Jiangsu Province Hospital, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China. ORCID
Yuan Zhang: Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications (Southeast University), Ministry of Education, Nanjing, China.
Liangmin Wei: Department of Biostatistics, Center for Global Health, School of Public Health, Nanjing Medical University, Nanjing, China.
Minghui Duan: Department of Haematology, Peking Union Medical College Hospital, Beijing, China. ORCID
Min Xiao: Department of Haematology, Tongji Medical College, Tongji Hospital, Huazhong University of Science and Technology, Wuhan, China. ORCID
Jin Wang: Department of Haematology, Tongji Medical College, Tongji Hospital, Huazhong University of Science and Technology, Wuhan, China.
Suning Chen: NHC Key Laboratory of Thrombosis and Hemostasis, National Clinical Research Center for Haematologic Diseases, Jiangsu Institute of Haematology, The First Affiliated Hospital of Soochow University, Suzhou, China. ORCID
Qian Wang: NHC Key Laboratory of Thrombosis and Hemostasis, National Clinical Research Center for Haematologic Diseases, Jiangsu Institute of Haematology, The First Affiliated Hospital of Soochow University, Suzhou, China. ORCID
Jianyao Huang: Department of Haematology, The First Affiliated Hospital of Anhui Medical University, Hefei, China.
Xiaomei Hu: Department of Pathology, Fujian Medical University Union Hospital, Fuzhou, China.
Jinhong Mei: The First Affiliated Hospital of Nanchang University, Nanchang, China.
Jieyu He: Department of Biostatistics, Center for Global Health, School of Public Health, Nanjing Medical University, Nanjing, China.
Feng Chen: Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications (Southeast University), Ministry of Education, Nanjing, China.
Lei Fan: Department of Haematology, Collaborative Innovation Center for Cancer Personalized Medicine, Jiangsu Province Hospital, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China. ORCID
Guanyu Yang: Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications (Southeast University), Ministry of Education, Nanjing, China.
Wenyi Shen: Department of Haematology, Collaborative Innovation Center for Cancer Personalized Medicine, Jiangsu Province Hospital, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China. ORCID
Yongyue Wei: Center for Public Health and Epidemic Preparedness & Response, Key Laboratory of Epidemiology of Major Diseases (Ministry of Education), School of Public Health, Peking University, Beijing, China. ORCID
Jianyong Li: Department of Haematology, Collaborative Innovation Center for Cancer Personalized Medicine, Jiangsu Province Hospital, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China. ORCID
The subjectivity of morphological assessment and the overlapping pathological features of different subtypes of myeloproliferative neoplasms (MPNs) make accurate diagnosis challenging. To improve the pathological assessment of MPNs, we developed a diagnosis model (fusion model) based on the combination of bone marrow whole-slide images (deep learning [DL] model) and clinical parameters (clinical model). Thousand and fifty-one MPN and non-MPNpatients were divided into the training, internal testing and one internal and two external validation cohorts (the combined validation cohort). In the combined validation cohort, fusion model achieved higher areas under curve (AUCs) than clinical or DL model or both for MPNs and subtype identification. Compared with haematopathologists with different experience, clinical model achieved AUC which was comparable to seniors and higher than juniors (p���=���0.0208) for polycythaemia vera. The AUCs of fusion model were comparable to seniors and higher than juniors for essential thrombocytosis (p���=���0.0141), prefibrotic primary myelofibrosis (p���=���0.0085) and overt primary myelofibrosis (p���=���0.0330) identification. In conclusion, the performances of our proposed models are equivalent to senior haematopathologists and better than juniors, providing a new perspective on the utilization of DL algorithms in MPN morphological assessment.
