Retrospective Study
Copyright ©The Author(s) 2020. Published by Baishideng Publishing Group Inc. All rights reserved.
World J Gastroenterol. Sep 14, 2020; 26(34): 5156-5168
Published online Sep 14, 2020. doi: 10.3748/wjg.v26.i34.5156
Construction of a convolutional neural network classifier developed by computed tomography images for pancreatic cancer diagnosis
Han Ma, Zhong-Xin Liu, Jing-Jing Zhang, Feng-Tian Wu, Cheng-Fu Xu, Zhe Shen, Chao-Hui Yu, You-Ming Li
Han Ma, Jing-Jing Zhang, Cheng-Fu Xu, Zhe Shen, You-Ming Li, Department of Gastroenterology, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, Zhejiang Province, China
Zhong-Xin Liu, College of Computer Science and Technology, Zhejiang University, Hangzhou 310058, Zhejiang Province, China
Feng-Tian Wu, State Key Laboratory for Diagnosis and Treatment of Infectious Disease, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, Zhejiang Province, China
Chao-Hui Yu, Department of Gastroenterology, Zhejiang Provincial Key Laboratory of Pancreatic Disease, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, Zhejiang Province, China
Author contributions: Yu CH and Li YM conceived this project, and acted as guarantors; Ma H managed the data acquisition, data labeling, and contributed to the data analysis and manuscript drafting and editing, and experiments; Liu ZX developed the technology, implemented the deep-learning architectures, conducted the main experiments, and contributed to the data analysis and manuscript drafting and editing; Wu FT and Zhang JJ collected the data; Xu CF and Shen Z reviewed CT images and data labeling; Yu CH supervised the experiments, analyzed the results; all authors approved the final version of the manuscript.
Supported by the National Natural Science Foundation of China, No. 81900509; Fundamental Research Funds for the Central Universities, No. 2018XZZX002-10; and High-Level Talents Special Support Plan of Zhejiang Province (known as the Ten Thousand Talents Plan), No. ZJWR0108008.
Institutional review board statement: The study was reviewed and approved by the review board of Zhejiang University School of Medicine, Zhejiang Province, China.
Informed consent statement: Patients were not required to give written informed consent to the study because the analysis used anonymous clinical data that were obtained after each patient agreed to treatment by written consent.
Conflict-of-interest statement: We have no financial relationships to disclose.
Data sharing statement: Consent was not obtained, but the presented data are anonymized and risk of identification is low.
Open-Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Corresponding author: Chao-Hui Yu, MD, PhD, Chief Doctor, Professor, Department of Gastroenterology, Zhejiang Provincial Key Laboratory of Pancreatic Disease, The First Affiliated Hospital, Zhejiang University School of Medicine, No. 79 Qingchun Road, Hangzhou 310003, Zhejiang Province, China. zyyyych@zju.edu.cn
Received: April 20, 2020
Peer-review started: April 20, 2020
First decision: May 1, 2020
Revised: May 19, 2020
Accepted: August 26, 2020
Article in press: August 26, 2020
Published online: September 14, 2020
Processing time: 142 Days and 0.5 Hours
Abstract
BACKGROUND

Efforts should be made to develop a deep-learning diagnosis system to distinguish pancreatic cancer from benign tissue due to the high morbidity of pancreatic cancer.

AIM

To identify pancreatic cancer in computed tomography (CT) images automatically by constructing a convolutional neural network (CNN) classifier.

METHODS

A CNN model was constructed using a dataset of 3494 CT images obtained from 222 patients with pathologically confirmed pancreatic cancer and 3751 CT images from 190 patients with normal pancreas from June 2017 to June 2018. We established three datasets from these images according to the image phases, evaluated the approach in terms of binary classification (i.e., cancer or not) and ternary classification (i.e., no cancer, cancer at tail/body, cancer at head/neck of the pancreas) using 10-fold cross validation, and measured the effectiveness of the model with regard to the accuracy, sensitivity, and specificity.

RESULTS

The overall diagnostic accuracy of the trained binary classifier was 95.47%, 95.76%, 95.15% on the plain scan, arterial phase, and venous phase, respectively. The sensitivity was 91.58%, 94.08%, 92.28% on three phases, with no significant differences (χ2 = 0.914, P = 0.633). Considering that the plain phase had same sensitivity, easier access, and lower radiation compared with arterial phase and venous phase , it is more sufficient for the binary classifier. Its accuracy on plain scans was 95.47%, sensitivity was 91.58%, and specificity was 98.27%. The CNN and board-certified gastroenterologists achieved higher accuracies than trainees on plain scan diagnosis (χ2 = 21.534, P < 0.001; χ2 = 9.524, P < 0.05; respectively). However, the difference between CNN and gastroenterologists was not significant (χ2 = 0.759, P = 0.384). In the trained ternary classifier, the overall diagnostic accuracy of the ternary classifier CNN was 82.06%, 79.06%, and 78.80% on plain phase, arterial phase, and venous phase, respectively. The sensitivity scores for detecting cancers in the tail were 52.51%, 41.10% and, 36.03%, while sensitivity for cancers in the head was 46.21%, 85.24% and 72.87% on three phases, respectively. Difference in sensitivity for cancers in the head among the three phases was significant (χ2 = 16.651, P < 0.001), with arterial phase having the highest sensitivity.

CONCLUSION

We proposed a deep learning-based pancreatic cancer classifier trained on medium-sized datasets of CT images. It was suitable for screening purposes in pancreatic cancer detection.

Keywords: Deep learning; Convolutional neural networks; Pancreatic cancer; Computed tomography

Core Tip: We developed a deep learning-based, computer-aided pancreatic ductal adenocarcinoma model trained on computed tomography images with pathologically confirmed pancreatic cancer in this retrospective study. We evaluated the approach used on the datasets in terms of both binary and ternary classifier, with the purposes of detecting and localizing masses, respectively. In the binary classifier, the performance of plain, arterial and venous phase had no difference. Its accuracy on plain scan was 95.47%, sensitivity was 91.58%, and specificity was 98.27%. In the ternary classifier, the arterial phase had the highest sensitivity in detecting cancer in the head of the pancreas among the three phases. Our model is suitable for screening purposes in pancreatic cancer detection.