Skip to main content
Fig. 1 | Breast Cancer Research

Fig. 1

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

Fig. 1

Proposed WSI-based IMS classifier and heterogeneity detection system. WSIs are broken into multiscale 400px × 400px patches and converted to descriptive tensors using the Inception v3 neural net architecture. A subset of cancer-enriched patches is selected to summarize WSI tumor content. Each patch is assigned a subtype in a 4-way classifier (Basal-like, HER2-enriched, Luminal A, and Luminal B). WSI-based subtype classifications can be made by employing a voting mechanism upon the patch-based results. Heterogeneity analysis is further performed on WSIs displaying significant concurrent Basal-like and Luminal A image-based predictions

Back to article page

Breast Cancer Research

ISSN: 1465-542X

Contact us