A deep learning-based semiautomated workflow for triaging follow-up MR scans in treated nasopharyngeal carcinoma

Ying-Ying Huang; Yi-Shu Deng; Yang Liu; Meng-Yun Qiang; Wen-Ze Qiu; Wei-Xiong Xia; Bing-Zhong Jing; Chen-Yang Feng; Hao-Hua Chen; Xun Cao; Jia-Yu Zhou; Hao-Yang Huang; Ze-Jiang Zhan; Ying Deng; Lin-Quan Tang; Hai-Qiang Mai; Ying Sun; Chuan-Miao Xie; Xiang Guo; Liang-Ru Ke; Xing Lv; Chao-Feng Li

doi:10.1016/j.isci.2023.108347

A deep learning-based semiautomated workflow for triaging follow-up MR scans in treated nasopharyngeal carcinoma

iScience. 2023 Oct 29;26(12):108347. doi: 10.1016/j.isci.2023.108347. eCollection 2023 Dec 15.

Authors

Ying-Ying Huang^{1

2}, Yi-Shu Deng^{1

3

4}, Yang Liu^{1

5}, Meng-Yun Qiang⁶, Wen-Ze Qiu⁷, Wei-Xiong Xia^{1

8}, Bing-Zhong Jing^{1

3}, Chen-Yang Feng^{1

3}, Hao-Hua Chen^{1

3}, Xun Cao^{1

9}, Jia-Yu Zhou^{1

8}, Hao-Yang Huang^{1

8}, Ze-Jiang Zhan^{1

8}, Ying Deng^{1

8}, Lin-Quan Tang^{1

8}, Hai-Qiang Mai^{1

8}, Ying Sun^{1

5}, Chuan-Miao Xie^{1

2}, Xiang Guo^{1

8}, Liang-Ru Ke^{1

2}, Xing Lv^{1

8}, Chao-Feng Li^{1

3}

Affiliations

¹ State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangdong Key Laboratory of Nasopharyngeal Carcinoma Diagnosis and Therapy, Guangzhou 510060, China.
² Department of Radiology, Sun Yat-sen University Cancer Center, Guangzhou 510060, China.
³ Department of Information, Sun Yat-sen University Cancer Center, Guangzhou 510060, China.
⁴ School of Electronics and Information Technology, Sun Yat-sen University, Guangzhou 510006, China.
⁵ Department of Radiation Oncology, Sun Yat-sen University Cancer Center, Guangzhou 510060, China.
⁶ Department of Radiation Oncology, Cancer Hospital of The University of Chinese Academy of Sciences, Hangzhou 310005, China.
⁷ Department of Radiation Oncology, The Affiliated Cancer Hospital of Guangzhou Medical University, Guangzhou 510095, China.
⁸ Department of Nasopharyngeal Carcinoma, Sun Yat-sen University Cancer Center, Guangzhou 510060, China.
⁹ Department of Critical Care Medicine, Sun Yat-sen University Cancer Center, Guangzhou 510060, China.

Abstract

It is imperative to optimally utilize virtues and obviate defects of fully automated analysis and expert knowledge in new paradigms of healthcare. We present a deep learning-based semiautomated workflow (RAINMAN) with 12,809 follow-up scans among 2,172 patients with treated nasopharyngeal carcinoma from three centers (ChiCTR.org.cn, Chi-CTR2200056595). A boost of diagnostic performance and reduced workload was observed in RAINMAN compared with the original manual interpretations (internal vs. external: sensitivity, 2.5% [p = 0.500] vs. 3.2% [p = 0.031]; specificity, 2.9% [p < 0.001] vs. 0.3% [p = 0.302]; workload reduction, 79.3% vs. 76.2%). The workflow also yielded a triaging performance of 83.6%, with increases of 1.5% in sensitivity (p = 1.000) and 0.6%-1.3% (all p < 0.05) in specificity compared to three radiologists in the reader study. The semiautomated workflow shows its unique superiority in reducing radiologist's workload by eliminating negative scans while retaining the diagnostic performance of radiologists.

Keywords: Applied computing; Health technology.