Deep Learning-Based Stage-Wise Risk Stratification for Early Lung Adenocarcinoma in CT Images: A Multi-Center Study.

Jing Gong,Weijun Peng,Tong Tong,Tingdan Hu,Xianwu Xia,Tingting Wang,Shengping Wang,Menglei Li,Yajia Gu,Haiming Li,Hui Zhu,Xianfang Hu,Jiyu Liu

doi:10.3390/cancers13133300

Abstract

Simple SummaryPrediction of the malignancy and invasiveness of ground glass nodules (GGNs) from computed tomography images is a crucial task for radiologists in risk stratification of early-stage lung adenocarcinoma. In order to solve this challenge, a two-stage deep neural network (DNN) was developed based on the images collected from four centers. A multi-reader multi-case observer study was conducted to evaluate the model capability. The performance of our model was comparable or even more accurate than that of senior radiologists, with average area under the curve values of 0.76 and 0.95 for two tasks, respectively. Findings suggest (1) a positive trend between the diagnostic performance and radiologist’s experience, (2) DNN yielded equivalent or even higher performance in comparison with senior radiologists, and (3) low image resolution reduced the model performance in predicting the risks of GGNs.This study aims to develop a deep neural network (DNN)-based two-stage risk stratification model for early lung adenocarcinomas in CT images, and investigate the performance compared with practicing radiologists. A total of 2393 GGNs were retrospectively collected from 2105 patients in four centers. All the pathologic results of GGNs were obtained from surgically resected specimens. A two-stage deep neural network was developed based on the 3D residual network and atrous convolution module to diagnose benign and malignant GGNs (Task1) and classify between invasive adenocarcinoma (IA) and non-IA for these malignant GGNs (Task2). A multi-reader multi-case observer study with six board-certified radiologists’ (average experience 11 years, range 2–28 years) participation was conducted to evaluate the model capability. DNN yielded area under the receiver operating characteristic curve (AUC) values of 0.76 ± 0.03 (95% confidence interval (CI): (0.69, 0.82)) and 0.96 ± 0.02 (95% CI: (0.92, 0.98)) for Task1 and Task2, which were equivalent to or higher than radiologists in the senior group with average AUC values of 0.76 and 0.95, respectively (p > 0.05). With the CT image slice thickness increasing from 1.15 mm ± 0.36 to 1.73 mm ± 0.64, DNN performance decreased 0.08 and 0.22 for the two tasks. The results demonstrated (1) a positive trend between the diagnostic performance and radiologist’s experience, (2) the DNN yielded equivalent or even higher performance in comparison with senior radiologists, and (3) low image resolution decreased model performance in predicting the risks of GGNs. Once tested prospectively in clinical practice, the DNN could have the potential to assist doctors in precision diagnosis and treatment of early lung adenocarcinoma.

Highlights

Lung cancer is the leading cause of cancer-related deaths globally, with almost onequarter of all cancer deaths [1]
Our contributions can be summarized as follows: (1) In this study, we proposed and developed a deep neural network (DNN) model to stratify the risk of early lung adenocarcinoma by using computed tomography (CT) images
The two-stage model classified between benign and malignant ground glass nodules (GGNs), and predicted the invasiveness of malignant tumors by differing invasive adenocarcinoma (IA) from non-IA. (2) By conducting an multi-reader multi-case (MRMC) observer study, our result demonstrates that the deep learning model performed equivalent to or even better than senior radiologists in predicting the risk of GGNs. (3) Analyzing the DNN performance changes on CT images with different resolutions, we found that the low resolution of CT images decreased the model performance

Summary

Introduction

Lung cancer is the leading cause of cancer-related deaths globally, with almost onequarter of all cancer deaths [1]. Lung cancer screening through detection and diagnosis of pulmonary nodules on CT scans is an essential and effective method. A large fraction of ground glass nodules (GGNs) are detected on the screening of CT images. As the biopsy of GGNs is a difficult task for interventional physicians, CT imaging is one of the optimal diagnosis measures for GGNs, especially for small ones. Most malignant GGNs are histopathologically confirmed as early-stage lung adenocarcinomas. According to the classification of the International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society, earlystage lung adenocarcinomas consist of pre-invasive lesions involving atypical adenomatous hyperplasia and adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), and invasive adenocarcinoma (IA) [3]. A precise diagnosis of GGNs facilitates the classification of lowand high-risk individuals (i.e., patients with benign and malignant GGNs, respectively), thereby avoiding overdiagnosis or overtreatment for early lung adenocarcinoma [5]. It is possible to make a personalized clinical care plan and select the optimal surgical treatment for patients with different pathological types (i.e., IA and non-IA patients)

Objectives

Methods

Results

Discussion

Conclusion