Figure 1 | Scientific Reports

Figure 1

From: The degradation of performance of a state-of-the-art skin image classifier when applied to patient-driven internet search

Figure 1

Binary classification for determining suspected malignancy using the Internet community (RD) dataset. (a) TEST = RD dataset (1,282 images). (b) TEST = RDadequate subset dataset (787 adequate images). (c) TEST = RDinadequate subset dataset (497 inadequate images), Red dot (TH1)-The algorithm at the high sensitivity threshold, Blue dot (TH2)-The algorithm at the high specificity threshold, Black dots-the six general physicians, Green dots-the layperson (cluster), × − Sensitivity and specificity derived from the Top-3 diagnoses of the participants, + − Sensitivity and specificity derived from the Top-1 diagnoses of the participants, The shaded area indicates the 95% confidence interval.

Back to article page