AI Can Accurately Improve Chest X-Ray Interpretation

Background: Chest x-rays are the most commonly ordered imaging study with reportedly often missed interpretation findings. An artificial intelligence (AI) product using a computer-assisted detection (CAD) system, called Chest-CAD, has been approved by the U.S. Food and Drug Administration to assist in the interpretation of chest x-rays. Objective: To determine whether AI using the CAD system, Chest-CAD, could accurately detect chest x-ray abnormalities. Methods: Chest x-rays that met the inclusion criteria (ie, patients aged ≥22 years and 1 or 2 acceptable chest x-ray images) were analyzed. The ability of the Chest-CAD system to detect chest x-ray abnormalities was tested when compared to expert radiologists. The accuracy was also evaluated using an independent chest x-ray data set from the National Institutes of Health (NIH). The accuracy of physicians in detecting chest x-ray abnormalities with and without the AI system was compared. Results: When compared to expert radiologists more...