Deep Learning for Identifying Metastatic Breast Cancer

TLDR

The ISBI grand challenge evaluated computational systems for automated detection of metastatic breast cancer in whole‑slide images of sentinel lymph node biopsies. Our deep‑learning system achieved an AUC of 0.925 for whole‑slide classification and 0.7051 for tumor localization, outperformed a pathologist’s AUC of 0.966 and localization score of 0.733, and when combined with the pathologist’s diagnoses raised the AUC to 0.995, demonstrating a substantial (~85 %) reduction in human error and the potential of deep learning to enhance pathological accuracy.

Abstract

The International Symposium on Biomedical Imaging (ISBI) held a grand challenge to evaluate computational systems for the automated detection of metastatic breast cancer in whole slide images of sentinel lymph node biopsies. Our team won both competitions in the grand challenge, obtaining an area under the receiver operating curve (AUC) of 0.925 for the task of whole slide image classification and a score of 0.7051 for the tumor localization task. A pathologist independently reviewed the same images, obtaining a whole slide image classification AUC of 0.966 and a tumor localization score of 0.733. Combining our deep learning system's predictions with the human pathologist's diagnoses increased the pathologist's AUC to 0.995, representing an approximately 85 percent reduction in human error rate. These results demonstrate the power of using deep learning to produce significant improvements in the accuracy of pathological diagnoses.