Fig. 4: Inter-pathologist and DLS-pathologist concordance for slide-level component scoring.
From: Deep learning models for histologic grading of breast cancer and association with disease prognosis
Analyses for mitotic count (A), nuclear pleomorphism (B), and tubule formation (C) are shown in the individual panels. Each blue bar represents the agreement (quadratic-weighted kappa) between a single pathologist and the other available pathologist scores on the same cases. The yellow bar represents the agreement of the DLS-provided component score with all available pathologist’ scores on the matched set of cases. Error bars represent 95% confidence intervals computed via bootstrap. Average values in legend represent quadratic-weighted kappa or the average of all blue bars and yellow bars, respectively. DLS deep learning system, Paths pathologists.
