Accepted for/Published in: Journal of Medical Internet Research
Date Submitted: Jul 23, 2024
Date Accepted: Dec 27, 2024
Enhancing Diagnostic Accuracy of Lung Nodules in Chest CT Scans Using Artificial Intelligence: A Retrospective Analysis
ABSTRACT
Background:
Artificial Intelligence (AI) systems are increasingly being integrated into medical imaging to enhance the accuracy and efficiency of diagnostic procedures. These technologies are designed to assist radiologists in the precise identification and measurement of abnormalities, such as lung nodules, in chest Computed Tomography (CT) scans.
Objective:
This study evaluates how an AI-assisted diagnostic system influenced the identification and measurement of lung nodules by radiologists.
Methods:
Conducted at two tertiary hospitals in Beijing from April 2018 to March 2022, this retrospective study assessed the influence of the AI system on the accuracy of detecting and measuring lung nodules in chest CT scans. Changes in report modification rates pre- and post-AI implementation were compared, using senior radiologists’ evaluations as a reference. The statistical analyses included descriptive statistics, Chi-squared, the Cochran-Armitage, and the Mann-Kendall tests.
Results:
A total of 12,889 patients were included in this study. Patients had encounters with the hospitals before, right after, or a few years after the implementation of the AI system. An increase in the modification rate of diagnostic reports was noted for the detection of lung nodules post-AI system implementation, particularly at Hospital A (p<.001). Both hospitals showed increased lung nodule detection rates post-implementation: Hospital C from 46.19% to 53.45%, and Hospital A from 39.29% to 55.22%. At Hospital A, a reduction in the false negative rate was observed (8.40% to 5.16%, p=.002), although it was accompanied by an increase in the false positive rate (2.36% to 9.77%, p=.014).
Conclusions:
The AI system enhanced lung nodule detection, offering the possibility of earlier disease identification and timelier intervention. Nevertheless, the initial reduction in accuracy underscores the need for standardized diagnostic criteria and comprehensive training for radiologists to maximize the effectiveness of AI-enable diagnostic systems.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.