Externally Validated Multi-Task Learning via Consistency Regularization Using Differentiable BI-RADS Features for Breast Ultrasound Tumor Segmentation

TL;DR

This paper introduces a novel consistency regularization method using differentiable BI-RADS features to enhance multi-task learning for breast ultrasound tumor segmentation, significantly improving generalization across multiple external datasets.

Contribution

It proposes a new regularization approach that mitigates task interference in multi-task learning for breast ultrasound segmentation, validated through extensive external dataset testing.

Findings

Significant improvement in Dice coefficient across datasets

State-of-the-art performance on UDIAT dataset

Statistically significant generalization enhancement (p<0.001)

Abstract

Multi-task learning can suffer from destructive task interference, where jointly trained models underperform single-task baselines and limit generalization. To improve generalization performance in breast ultrasound-based tumor segmentation via multi-task learning, we propose a novel consistency regularization approach that mitigates destructive interference between segmentation and classification. The consistency regularization approach is composed of differentiable BI-RADS-inspired morphological features. We validated this approach by training all models on the BrEaST dataset (Poland) and evaluating them on three external datasets: UDIAT (Spain), BUSI (Egypt), and BUS-UCLM (Spain). Our comprehensive analysis demonstrates statistically significant (p<0.001) improvements in generalization for segmentation task of the proposed multi-task approach vs. the baseline one: UDIAT, BUSI, BUS-UCLM (Dice coefficient=0.81 vs 0.59, 0.66 vs 0.56, 0.69 vs 0.49, resp.). The proposed approach also achieves state-of-the-art segmentation performance under rigorous external validation on the UDIAT dataset.