Optimized AI strategy for thyroid nodules benefits some radiologists

According to research results published online May 16, optimized artificial intelligence (AI) strategies can reduce diagnostic time-based costs for senior radiologists, whereas conventional all-AI strategies reduce junior radiologists’ costs. It seems to be more beneficial for physicians. JAMA network open.

Wen-Juan Tong and colleagues from Guangzhou First Affiliated Hospital, China and Sun Yat-sen University have developed an optimal AI decision support that reduces the workload of radiologists while maintaining diagnostic performance compared to traditional AI-assisted strategies. developed a streamlined integration.

Using a retrospective set of 1,754 ultrasound images of 1,048 patients with 1,754 thyroid nodules, 16 junior and senior radiologists explored how AI-assisted diagnostic outcomes with a variety of image features were evaluated. Built an optimized strategy based on what you incorporated. This optimized strategy was compared to a conventional all-AI strategy using her 300 ultrasound images of 268 patients with his 300 thyroid nodules in the prospective set of studies. .

The researchers found that the optimized strategy was associated with an increase in average task completion time for junior radiologists, while it was associated with a reduction in average work time for senior radiologists compared to all conventional AI strategies. I discovered that For readers 11–16, no significant differences in sensitivity (range, 91–100 percent) or specificity (range, 94–98 percent) were observed between the two strategies.

“We recommend that junior radiologists apply all conventional AI strategies in the management of thyroid nodules, whereas senior radiologists should apply optimized strategies,” the authors wrote. ing. “The optimized integration of these AI decision supports could help reduce the workload of radiologists by reducing diagnostic time-based costs while maintaining excellent diagnostic performance.” .”