Monday, December 1 | 2:00 p.m.-2:10 p.m. | M6-SSMK03-4 | Room E450A
AI-assisted structured reporting for knee MRI streamlines radiology workflow by decreasing reporting time and cognitive workload, according to research results to be presented Monday afternoon.
A team led by presenter Benoit Rizk, MD, of 3R - Réseau Radiologique Romand in Sion, Switzerland, found that integrating AI-generated findings into structured knee MRI reports reduced reporting time, particularly for general radiologists.
Rizk and colleagues evaluated the impact of AI on knee MRI reporting time by comparing three different workflows: reporting without the help of AI, partial integration of commercial AI software that helps radiologists assess cartilage/menisci/ligaments, and full AI integration with AI findings incorporated into structured report templates.
The group conducted a study that included 10 private radiology centers in Switzerland, eight radiologists (4 musculoskeletal subspecialists and 4 generalists), and 1,285 knee MRI exams. The team tracked reporting time via the facilities' radiology information systems.
The group found that fully integrated AI-assisted reporting reduced overall knee MRI reporting time by 13.4% compared with standard reporting, with a p-value of < 0.01, and that partial AI integration into the reporting process did not produce workflow time savings. It also reported that 75% of the participating radiologists stated that AI "facilitated their work," and that 62.5% declared that structured report integration was beneficial.
"Effective integration into existing reporting systems is crucial to realizing these benefits in routine clinical practice," Rizk and colleagues concluded.
