Multi-Scale Multi-Target Domain Adaptation for Angle Closure Classification

TL;DR

This paper introduces M2DAN, a multi-scale adversarial network that enables angle closure classification across multiple unlabeled domains by extracting domain-invariant features from AS-OCT images.

Contribution

The paper proposes a novel multi-scale, multi-target domain adversarial network for angle closure classification, capable of handling multiple unlabeled target domains.

Findings

Effective cross-domain angle closure classification demonstrated

Outperforms existing domain adaptation methods

Robust to variations across different imaging devices

Abstract

Deep learning (DL) has made significant progress in angle closure classification with anterior segment optical coherence tomography (AS-OCT) images. These AS-OCT images are often acquired by different imaging devices/conditions, which results in a vast change of underlying data distributions (called "data domains"). Moreover, due to practical labeling difficulties, some domains (e.g., devices) may not have any data labels. As a result, deep models trained on one specific domain (e.g., a specific device) are difficult to adapt to and thus may perform poorly on other domains (e.g., other devices). To address this issue, we present a multi-target domain adaptation paradigm to transfer a model trained on one labeled source domain to multiple unlabeled target domains. Specifically, we propose a novel Multi-scale Multi-target Domain Adversarial Network (M2DAN) for angle closure classification. M2DAN conducts multi-domain adversarial learning for extracting domain-invariant features and develops a multi-scale module for capturing local and global information of AS-OCT images. Based on these domain-invariant features at different scales, the deep model trained on the source domain is able to classify angle closure on multiple target domains even without any annotations in these domains. Extensive experiments on a real-world AS-OCT dataset demonstrate the effectiveness of the proposed method.