DOMINO++: Domain-aware Loss Regularization for Deep Learning Generalizability

TL;DR

DOMINO++ introduces a dynamic, domain-aware loss regularization method that enhances deep learning models' ability to generalize to out-of-distribution data, especially in medical imaging applications.

Contribution

It develops a dual-guidance, adaptive regularization technique that improves OOD generalization over previous fixed-rate methods like DOMINO.

Findings

DOMINO++ outperforms DOMINO and baseline models on OOD MRI data.

Enhanced model calibration leads to better feature-class association.

Improved robustness to synthetic noise and rotations in test data.

Abstract

Out-of-distribution (OOD) generalization poses a serious challenge for modern deep learning (DL). OOD data consists of test data that is significantly different from the model's training data. DL models that perform well on in-domain test data could struggle on OOD data. Overcoming this discrepancy is essential to the reliable deployment of DL. Proper model calibration decreases the number of spurious connections that are made between model features and class outputs. Hence, calibrated DL can improve OOD generalization by only learning features that are truly indicative of the respective classes. Previous work proposed domain-aware model calibration (DOMINO) to improve DL calibration, but it lacks designs for model generalizability to OOD data. In this work, we propose DOMINO++, a dual-guidance and dynamic domain-aware loss regularization focused on OOD generalizability. DOMINO++ integrates expert-guided and data-guided knowledge in its regularization. Unlike DOMINO which imposed a fixed scaling and regularization rate, DOMINO++ designs a dynamic scaling factor and an adaptive regularization rate. Comprehensive evaluations compare DOMINO++ with DOMINO and the baseline model for head tissue segmentation from magnetic resonance images (MRIs) on OOD data. The OOD data consists of synthetic noisy and rotated datasets, as well as real data using a different MRI scanner from a separate site. DOMINO++'s superior performance demonstrates its potential to improve the trustworthy deployment of DL on real clinical data.