Computer-Aided Pathologic Diagnosis of Nasopharyngeal Carcinoma Based on Deep Learning

Diao, Songhui1,3; Hou, Jiaxin1,4; Yu, Hong2; Zhao, Xia2; Sun, Yikang2; Lambo, Ricardo Lewis1; Xie, Yaoqin1; Liu, Lei2; Qin, Wenjian1; Luo, Weiren2

2020-08

10.1016/j.ajpath.2020.04.008

AMERICAN JOURNAL OF PATHOLOGY   影响因子和分区

0002-9440

1525-2191

The pathologic diagnosis of nasopharyngeal carcinoma (NPC) by different pathologists is often inefficient and inconsistent. We have therefore introduced a deep learning algorithm into this process and compared the performance of the model with that of three pathologists with different levels of experience to demonstrate its clinical value. In this retrospective study, a total of 1970 whole slide images of 731 cases were collected and divided into training, validation, and testing sets. Inception-v3, which is a state-of-the-art convolutional neural network, was trained to classify images into three categories: chronic nasopharyngeal inflammation, lymphoid hyperplasia, and NPC. The mean area under the curve (AUC) of the deep learning model is 0.936 based on the testing set, and its AUCs for the three image categories are 0.905, 0.972, and 0.930, respectively. In the comparison with the three pathologists, the model outperforms the junior and intermediate pathologists, and has only a slightly lower performance than the senior pathologist when considered in terms of accuracy, specificity, sensitivity, AUC, and consistency. To our knowledge, this is the first study about the application of deep learning to NPC pathologic diagnosis. In clinical practice, the deep learning model can potentially assist pathologists by providing a second opinion on their NPC diagnoses.

SCI

英语

通讯

National Natural Science Foundation of China[61901463][81872202] ; Shenzhen Science and Technology Program of China[JCYJ20170818160306270] ; Natural Science Foundation of Guangdong Province[2015A030313263][2018A030313778] ; Guangdong Provincial Key Laboratory of Cell Microenvironment and Disease Research[2017B030301018]

Pathology

Pathology

WOS:000552675100011

ELSEVIER SCIENCE INC

CLINICAL MEDICINE

Web of Science

1.Chinese Acad Sci, Shenzhen Inst Adv Technol, 1068 Xueyuan Ave, Shenzhen 518055, Peoples R China
2.Southern Univ Sci & Technol, Shenzhen Peoples Hosp 3, Affiliated Hosp 2, Natl Clin Res Ctr Infect Dis,Canc Res Inst,Dept P, 29 Bulan Rd, Shenzhen 518112, Peoples R China
3.Univ Chinese Acad Sci, Shenzhen Coll Adv Technol, Shenzhen, Peoples R China
4.Northeastern Univ, Coll Med & Biol Informat Engn, Shenyang, Peoples R China

Diao, Songhui,Hou, Jiaxin,Yu, Hong,et al. Computer-Aided Pathologic Diagnosis of Nasopharyngeal Carcinoma Based on Deep Learning[J]. AMERICAN JOURNAL OF PATHOLOGY,2020,190(8):1691-1700.

Diao, Songhui.,Hou, Jiaxin.,Yu, Hong.,Zhao, Xia.,Sun, Yikang.,...&Luo, Weiren.(2020).Computer-Aided Pathologic Diagnosis of Nasopharyngeal Carcinoma Based on Deep Learning.AMERICAN JOURNAL OF PATHOLOGY,190(8),1691-1700.

Diao, Songhui,et al."Computer-Aided Pathologic Diagnosis of Nasopharyngeal Carcinoma Based on Deep Learning".AMERICAN JOURNAL OF PATHOLOGY 190.8(2020):1691-1700.