Deep Multi-Magnification Networks for Multi-Class Breast Cancer Image Segmentation

TL;DR

This paper introduces a deep multi-magnification neural network for automated multi-class breast cancer tissue segmentation in digitized pathology slides, improving accuracy over existing methods.

Contribution

The proposed architecture with multi-encoder, multi-decoder, and multi-concatenation techniques advances multi-magnification image segmentation for breast cancer pathology.

Findings

Achieves the highest mean intersection-over-union among tested models.

Outperforms other single and multi-magnification architectures.

Facilitates pathologists' assessments of breast cancer.

Abstract

Pathologic analysis of surgical excision specimens for breast carcinoma is important to evaluate the completeness of surgical excision and has implications for future treatment. This analysis is performed manually by pathologists reviewing histologic slides prepared from formalin-fixed tissue. In this paper, we present Deep Multi-Magnification Network trained by partial annotation for automated multi-class tissue segmentation by a set of patches from multiple magnifications in digitized whole slide images. Our proposed architecture with multi-encoder, multi-decoder, and multi-concatenation outperforms other single and multi-magnification-based architectures by achieving the highest mean intersection-over-union, and can be used to facilitate pathologists' assessments of breast cancer.