Fig. 2

Consistency analysis and scoring results of the subjective scoring before and after artifact processing. a shows the results of doctor A’s two evaluations before and after the removal of the artifact, P> 0.05, suggesting that the differences between the two evaluations are small. b shows that the results of Doctor B are the same as those of Doctor A. c shows no significant difference was found between the three subjective scores between before and after artifact removal (P > 0.05), indicating good inter-observer consistency. Because of this consistency, we grouped the two doctors’ scores into a single dataset