Automatic recognition of bladder tumors using deep learning technology and its clinical application.

Bladder cancer is a kind of tumor with a high recurrence rate. Improvement of the cure rate and prognosis of bladder tumors depends on the accurate recognition of bladder tumors under a cystoscope. In this study, 1200 cystoscopic cancer images from 224 patients with bladder cancer and 1150 cystoscopic images from 221 patients with no bladder cancer were collected. Three convolutional neural networks (LeNet, AlexNet, and GoogLeNet), and the EasyDL deep learning platform were used to train deep learning models to distinguish images of bladder cancer. The efficiency of EasyDL was the highest, and the accuracy was 96.9%. The efficiency of GoogLeNet was the second highest, and the accuracy was 92.54%. The EasyDL model was loaded into a deep learning calculation card (EdgeBoard), and it was applied in clinical practice. The accuracy of the deep learning system was compared with that of clinical experts. Among the 33 bladder cancer nodes and 11 no bladder cancer nodes, the accuracy of the neural network was 83.36% and that of medical experts was 84.09% (P > 0.05). This study confirmed that deep learning technology can effectively differentiate bladder tumors in real-time, with an accuracy rate close to the urologist, which might improve the diagnosis and treatment level of bladder tumors. This article is protected by copyright. All rights reserved.