Topology-Aware Learning for Semi-supervised Cross Domain Retinal Artery/Vein Classification

Hanlin Liu, Jianyang Xie, YONGHUAI LIU, Huaying Hao, Lijun Guo, Huazhu Fu, Jiong Zhang, Yitian Zhao

Research output: Chapter in Book/Report/Conference proceedingConference proceeding (ISBN)peer-review

Abstract

Pixel-level Artery/Vein (A/V) classification of retinal blood
vessels is important in diagnosing and understanding a wide spectrum of
diseases. Deep learning-based methods have achieved promising results,
but manual image annotation is a labor-intensive and time-consuming
process and annotations between different datasets cannot be fully utilized.
The model performance might be significantly decreased when
training on one dataset but testing on another. Some unsupervised domain
adaptation (UDA) methods can alleviate this problem. However,
annotating a few target samples is usually very manageable and worthwhile
especially if it improves the adaptation performance substantially.
Consequently, we propose a novel semi-supervised retinal A/V classification
method where a few labeled target samples and the labeled publiclyaccessible
source dataset are available. We first employ the teacherstudent
framework to achieve semi-supervised learning. Then, a new
regional mixing method is proposed to reduce the domain gap at the
region level, which takes into account the topological features of retinal
vasculature. Furthermore, we combine spatial transformation with
regional mixing to generate additional synthetic images, thus improving
the generalization ability of the model.We evaluate the proposed method
on a publicly accessible (DRIVE-AV) dataset and a private dataset, and
the results show that the proposed method achieves state-of-the-art performance
for A/V classification.
Original languageEnglish
Title of host publicationComputer graphics International -CGI 2022
Publisherspringer
Publication statusAccepted/In press - 13 Jul 2022
EventComputer Graphics International - Full Virtual
Duration: 12 Sep 202216 Sep 2022

Publication series

NameLecture Notes in Computer Science
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

ConferenceComputer Graphics International
Abbreviated titleCGI
Period12/09/2216/09/22

Keywords

  • A/V classification
  • domain-shift
  • semi-supervision

Fingerprint

Dive into the research topics of 'Topology-Aware Learning for Semi-supervised Cross Domain Retinal Artery/Vein Classification'. Together they form a unique fingerprint.

Cite this