Incidence Lung Cancer after a Negative CT Screening in the National Lung Screening Trial: Deep Learning-Based Detection of Missed Lung Cancers.

Kyung Won Lee,Kyung Hee Lee,Jungheum Cho,Chang Mo Nam,Junghoon Kim,Hwayoung Song,Jihang Kim,Sung Hyun Yoon,Ye Ra Choi,Kyong Joon Lee

doi:10.3390/jcm9123908

Abstract

We aimed to analyse the CT examinations of the previous screening round (CTprev) in NLST participants with incidence lung cancer and evaluate the value of DL-CAD in detection of missed lung cancers. Thoracic radiologists reviewed CTprev in participants with incidence lung cancer, and a DL-CAD analysed CTprev according to NLST criteria and the lung CT screening reporting & data system (Lung-RADS) classification. We calculated patient-wise and lesion-wise sensitivities of the DL-CAD in detection of missed lung cancers. As per the NLST criteria, 88% (100/113) of CTprev were positive and 74 of them had missed lung cancers. The DL-CAD reported 98% (98/100) of the positive screens as positive and detected 95% (70/74) of the missed lung cancers. As per the Lung-RADS classification, 82% (93/113) of CTprev were positive and 60 of them had missed lung cancers. The DL-CAD reported 97% (90/93) of the positive screens as positive and detected 98% (59/60) of the missed lung cancers. The DL-CAD made false positive calls in 10.3% (27/263) of controls, with 0.16 false positive nodules per scan (41/263). In conclusion, the majority of CTprev in participants with incidence lung cancers had missed lung cancers, and the DL-CAD detected them with high sensitivity and a limited false positive rate.

Highlights

The National Lung Screening Trial (NLST) demonstrated that three rounds of annual low-dose computed tomography (CT) screening reduced lung cancer mortality rates by 20% among asymptomatic high-risk participants [1]
We reviewed de-identified CT screenings of the NLST participants who were diagnosed with incidence lung cancer, defined as cancer diagnosed through incidence screening after previous verifiable, negative CT screenings
The deep learning (DL)-computer-aided detection system (CAD) reviewed CT examinations of the previous screening round (CTprev) in participants diagnosed with incidence lung cancer

Summary

Introduction

The National Lung Screening Trial (NLST) demonstrated that three rounds of annual low-dose computed tomography (CT) screening (rounds T0, T1, and T2) reduced lung cancer mortality rates by 20% among asymptomatic high-risk participants [1]. Lung cancers in the NLST can be categorised into prevalent, interval, and incidence [2]. Prevalence of cancer refers to cancer detected on the first screening (T0), and interval cancer refers to cancer detected before the scheduled screening (between T0 and T1 or between T1 and T2). Incidence cancer refers to cancer diagnosed through a scheduled incidence screening test, after a previous round of negative screening (detected through T1 or T2). A previous study [3], which retrospectively analysed the CT examinations of 44 participants who were subsequently diagnosed with interval lung cancer, reported a higher number of false negative screens in the interval lung cancer group compared to the control group. To the best of our knowledge, no previous study has investigated the previous round of CT screenings in participants who were subsequently diagnosed with incidence lung cancer. Liang et al [4] reported that the computer-aided detection system (CAD) might be used as an effective second reader, which could detect up to 70% of the nodules missed by radiologists, and Ardila et al [5] reported the possibility of end-to-end lung screening by means of the deep learning (DL)-based CAD

Objectives

Results

Discussion

Conclusion