Skip to main content
Fig. 4 | Breast Cancer Research

Fig. 4

From: A deep learning image-based intrinsic molecular subtype classifier of breast tumors reveals tumor heterogeneity that may affect survival

Fig. 4

Evidence for heterogeneity. a An example of a HET WSI with markup on patches predicted as Basal-like and LumA. b Expression levels of key hormone receptors ESR1 and PGR in the three settings. Mann-Whitney U p values of being drawn from the same distribution are reported for each pair of settings. Inputs are BasalIMG, HET, and LumAIMG cohorts as defined by the WSI-based IMS system. c Kaplan-Meier curves for BasalIMG, HET, and LumAIMG cohorts show HET survival to be intermediate between the other two. Cox proportional hazard test is included

Back to article page