Dental students marking more panoramic regions on AI-assisted reads scored lower

In 266 dental students using an artificial intelligence-supported panoramic radiograph platform, marking >8.5 regions was tied to lower diagnostic performance scores (p<0.001); region frequency versus AI-supported diagnosis had AUC 0.414 (95% CI 0.342-0.486, p=0.018).

• The study used a confusion-matrix framework to calculate precision and sensitivity for 266 fourth-year dental students evaluating five panoramic radiographs on an AI-supported lesion-detection platform.
• AI-supported performance scores showed no significant correlation with traditional summative examination outcomes, suggesting the tool measures applied diagnostic reasoning distinct from written exams.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.