Commercial AI cuts pulmonary nodule assessment times by up to 25%
The use of artificial intelligence-enabled software could significantly reduce interpretation times for chest CT scans.
A new paper in RSNA’s flagship journal Radiology details the utility of a commercial AI tool designed to support radiologists when evaluating pulmonary nodules—a common finding on chest CTs that often requires additional workup.
“Chest CT has become a primary tool for identifying pulmonary nodules, yet interpreting these scans remains time-intensive and cognitively demanding, even for experienced radiologists,” Jacob J. Visser, MD, PhD, with the department of radiology and nuclear medicine at Erasmus Medical Center in the Netherlands, and colleagues noted. “With increasing imaging volumes and growing pressure on radiology departments, efficiency and accuracy in lung nodule detection and analysis are more important than ever.”
The retrospective study included 19,433 patients who underwent a total of 39,323 chest CT exams at a tertiary medical center between September 2021 and May 2024. Researchers compared reporting times before and after implementation of a commercial AI software for pulmonary nodule assessment.
Following AI implementation, the adjusted median reporting time decreased from 21.3 minutes to 18.2 minutes—a 14.6% reduction. Overall, AI implementation was associated with a 17% increase in the rate of report completion after adjustment for reader function, examination type, patient location and requesting specialty.
They observed significant variation in the effect of AI across clinical subgroups; the largest reduction in reporting time occurred for ECG-gated CT thorax examinations, where median reporting time decreased by 41.1%. Thoracic radiologists also experienced a substantial reduction of 25%.
However, AI did not shorten reporting times in every setting. Scans requested from the emergency department were associated with a 7.1% increase in median reporting time following implementation. This could suggest that AI's effect on radiologist efficiency may depend on the radiologist’s interpreting environment.
“A plausible explanation is that emergency imaging workflows are typically driven by acute diagnostic priorities, in which pulmonary nodule assessment is often not the primary clinical focus,” the authors explained. “In such settings, thoracic nodule-specific AI support may offer less practical workflow benefit than in elective or subspecialized care pathways.”
The group acknowledged the promise of their findings before suggesting that future research should assess generalizability across different healthcare settings. They also emphasized the importance of determining whether efficiency gains are maintained long-term and whether they improve diagnostic performance.
Read more here.
In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She began covering the medical imaging industry for Innovate Healthcare in 2021.