Qualitative Histopathological Classification of Primary Bone Tumors Using Deep Learning: A Pilot Study

Frontiers in Oncology

Yu Chen¹  Yuzhang Tao¹  Wu Yang¹  Xiao Huang¹  Hongwei Wang¹  Boyu Tang¹  Minkang Guo¹  Aiguo Zhou¹  Jian Zhang¹  Weiqian Jiang¹  Chenglong Wang²  Jing Luo³  Youde Cao³  Yiwen Tan⁴  Mengli Yao⁵  Tingmei Chen⁵  Kangrong Gao⁶  Zhi Liu⁶  Chengsi Luo⁷ 

[1] Department of Orthopaedics, The First Affiliated Hospital of Chongqing Medical University, Chongqing, China;Department of Pathology, Chongqing Hospital of Traditional Chinese Medicine, Chongqing, China;Department of Pathology, College of Basic Medicine, Chongqing Medical University, Chongqing, China;Department of Pathology, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, China;Key Laboratory of Clinical Laboratory Diagnostics (Ministry of Education), College of Laboratory Medicine, Chongqing Medical University, Chongqing, China;Research and Development Department, Chongqing Defang Information Technology Co., Ltd, Chongqing, China;School of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu, China;

关键词: primary bone tumors;    deep learning;    histopathological classification;    convolutional neural network (CNN);    diagnosis;   

DOI  :  10.3389/fonc.2021.735739

来源: DOAJ

【 摘 要 】

BackgroundHistopathological diagnosis of bone tumors is challenging for pathologists. We aim to classify bone tumors histopathologically in terms of aggressiveness using deep learning (DL) and compare performance with pathologists.MethodsA total of 427 pathological slides of bone tumors were produced and scanned as whole slide imaging (WSI). Tumor area of WSI was annotated by pathologists and cropped into 716,838 image patches of 256 × 256 pixels for training. After six DL models were trained and validated in patch level, performance was evaluated on testing dataset for binary classification (benign vs. non-benign) and ternary classification (benign vs. intermediate vs. malignant) in patch-level and slide-level prediction. The performance of four pathologists with different experiences was compared to the best-performing models. The gradient-weighted class activation mapping was used to visualize patch’s important area.ResultsVGG-16 and Inception V3 performed better than other models in patch-level binary and ternary classification. For slide-level prediction, VGG-16 and Inception V3 had area under curve of 0.962 and 0.971 for binary classification and Cohen’s kappa score (CKS) of 0.731 and 0.802 for ternary classification. The senior pathologist had CKS of 0.685 comparable to both models (p = 0.688 and p = 0.287) while attending and junior pathologists showed lower CKS than the best model (each p < 0.05). Visualization showed that the DL model depended on pathological features to make predictions.ConclusionDL can effectively classify bone tumors histopathologically in terms of aggressiveness with performance similar to senior pathologists. Our results are promising and would help expedite the future application of DL-assisted histopathological diagnosis for bone tumors.

【 授权许可】

Unknown