Experience with Single Domain Generalization in Real World Medical Imaging Deployments

TL;DR

This paper explores the challenges of single domain generalization in real-world medical imaging, demonstrating the limitations of current methods and proposing an expert knowledge integrated deep learning approach that improves performance in specific applications.

Contribution

The paper introduces DL+EKE, a novel deep learning technique incorporating expert knowledge, which outperforms existing SDG methods in medical imaging applications.

Findings

State-of-the-art SDG methods fail to generalize across domains in medical imaging.

DL+EKE outperforms existing SDG techniques in diabetic retinopathy.

Real-world deployment reveals issues faced by SDG methods in practice.

Abstract

A desirable property of any deployed artificial intelligence is generalization across domains, i.e. data generation distribution under a specific acquisition condition. In medical imagining applications the most coveted property for effective deployment is Single Domain Generalization (SDG), which addresses the challenge of training a model on a single domain to ensure it generalizes well to unseen target domains. In multi-center studies, differences in scanners and imaging protocols introduce domain shifts that exacerbate variability in rare class characteristics. This paper presents our experience on SDG in real life deployment for two exemplary medical imaging case studies on seizure onset zone detection using fMRI data, and stress electrocardiogram based coronary artery detection. Utilizing the commonly used application of diabetic retinopathy, we first demonstrate that state-of-the-art SDG techniques fail to achieve generalized performance across data domains. We then develop a generic expert knowledge integrated deep learning technique DL+EKE and instantiate it for the DR application and show that DL+EKE outperforms SOTA SDG methods on DR. We then deploy instances of DL+EKE technique on the two real world examples of stress ECG and resting state (rs)-fMRI and discuss issues faced with SDG techniques.