Comparative Analysis of Deep Learning Architectures for Breast Region Segmentation with a Novel Breast Boundary Proposal

TL;DR

This study compares seven deep learning models for breast region segmentation in DCE-MRI, introducing a novel breast boundary method to improve accuracy and reduce computational costs, with UNet++ and FCNResNet50 showing top performance.

Contribution

The paper presents a new breast boundary proposal method and a comprehensive comparison of deep learning architectures for breast segmentation in MRI.

Findings

UNet++ achieved the highest Dice score.

UNet demonstrated strong validation and generalizability.

FCNResNet50 offered a good balance of performance and environmental impact.

Abstract

Purpose: Segmentation of the breast region in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is essential for the automatic measurement of breast density and the quantitative analysis of imaging findings. This study aims to compare various deep learning methods to enhance whole breast segmentation and reduce computational costs as well as environmental effect for future research. Methods: We collected fifty-nine DCE-MRI scans from Stavanger University Hospital and, after preprocessing, analyzed fifty-eight scans. The preprocessing steps involved standardizing imaging protocols and resampling slices to ensure consistent volume across all patients. Using our novel approach, we defined new breast boundaries and generated corresponding segmentation masks. We evaluated seven deep learning models for segmentation namely UNet, UNet++, DenseNet, FCNResNet50, FCNResNet101, DeepLabv3ResNet50, and DeepLabv3ResNet101. To ensure robust model validation, we employed 10-fold cross-validation, dividing the dataset into ten subsets, training on nine, and validating on the remaining one, rotating this process to use all subsets for validation. Results: The models demonstrated significant potential across multiple metrics. UNet++ achieved the highest performance in Dice score, while UNet excelled in validation and generalizability. FCNResNet50, notable for its lower carbon footprint and reasonable inference time, emerged as a robust model following UNet++. In boundary detection, both UNet and UNet++ outperformed other models, with DeepLabv3ResNet also delivering competitive results.