Published online Sep 14, 2020. doi: 10.3748/wjg.v26.i34.5156
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
Pancreatic cancer is a highly lethal malignancy with a very poor prognosis. With promising achievements in deep neural networks and increasing medical needs, computer-aided diagnosis systems have become a new research focus.
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.
To identify pancreatic cancer in computed tomography (CT) images automatically by constructing a convolutional neural network (CNN) classifier.
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 built three datasets from our images according to the image phases, evaluated our approach in terms of binary classification and ternary classification using 10-fold cross validation, and measured the effectiveness of the model with regard to the accuracy, sensitivity, and specificity.
In the binary classifiers, the performance of plain, arterial and venous phase showed no difference. Considering that plain phase had relatively 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%. In the ternary classifier, the arterial phase had the highest sensitivity in detecting cancer in the head of the pancreas among three phases, but it achieved only moderate performances.
In this study, we developed a deep learning-based, computer-aided pancreatic ductal adenocarcinoma classifier trained on medium-sized CT images. It was suitable for screening purposes in pancreatic cancer detection.
Further improvement in the performance of models would be required before it could be integrated into a clinical strategy.