THE EFFECT OF DEEP LEARNING-BASED COMPUTER-AIDED DETECTION ON THE DETECTION RATE OF PULMONARY NODULES IN MRI SCANS

Abstract

Computer-aided detection (CAD) systems have been developed to aid in the detection of pulmonary nodules on MRI scans. However, the diagnostic accuracy of these systems varies widely, and their impact on the detection rate of pulmonary nodules in clinical practice is uncertain. The aim was to evaluate the effect of a deep learning-based CAD system on the detection rate of pulmonary nodules in MRI scans. This randomized controlled trial was conducted in participants undergoing lung cancer screening with low-dose MRI scans. 
Participants were randomly assigned to either the intervention or control group in a 1:1 ratio. The primary outcome, the detection rate of pulmonary nodules, was significantly higher in the intervention group compared to the control group (70% vs. 50%, p=0.03). The use of a deep learning-based CAD system in addition to standard radiologist interpretation improves the detection rate of pulmonary nodules on low-dose MRI scans in participants undergoing lung cancer screening.