Distribution-Aware Multi-Label FixMatch for Semi-Supervised Learning on CheXpert.

verfasst von: Sontje Ihler, Felix Kuhnke, Timo Kuhlgatz, Thomas Seel
Abstract: <!DOCTYPE html>
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<title>Semi-supervised Learning for Multi-label Classification</title>
<meta charset="UTF-8" />
</head>
<body>

Semi-supervised learning (SSL) has achieved remarkable success for multiclass classification in recent years,
yielding a promising solution for medical image classification where labeled data is scarce but unlabeled images are accessible.
In the context of multi-label problems, however, SSL is still under-explored.
In this work, we adapt FixMatch to the multi-label scenario, specifically focusing on CheXpert,
a multi-label chest X-ray classification dataset which is imbalanced and only partially labeled.


Leveraging distribution alignment, our proposed method, ML-FixMatch+DA, achieves solid performance gains in SSL tasks
(AUC: +2.6%) and in a missing label scenario (AUC: +1.9%).
In contrast to previous work, we achieve a performance gain on CheXpert using FixMatch.
We show that in contrast to multiclass FixMatch, where distribution alignment is optional,
it is essential for multi-label FixMatch to handle class imbalance and generate reliable (positive and negative) pseudo-labels.


Our pseudo-label selection is based on a single threshold for all classes and handles imbalance with no prior knowledge on label distributions.
Our adaptation keeps the simplicity of the original multiclass FixMatch with no added hyperparameters (even for imbalanced data)
and demonstrates the feasibility of simple SSL for multi-label problems, filling a crucial gap in the literature.

</body>
</html>
Organisationseinheit(en): Institut für Mechatronische Systeme
Medizintechnik & Bildverarbeitung
Institut für Informationsverarbeitung
Typ: Aufsatz in Konferenzband
Seiten: 2295-2304
Anzahl der Seiten: 10
Publikationsdatum: 01.06.2024
Publikationsstatus: Elektronisch veröffentlicht (E-Pub)
Peer-reviewed: Ja

BibTeX

@inproceedings{a80b16e0a99d48d6891b50ebdabcd4be,
title = "Distribution-Aware Multi-Label FixMatch for Semi-Supervised Learning on CheXpert.",
abstract = "<!DOCTYPE html>Semi-supervised Learning for Multi-label ClassificationSemi-supervised learning (SSL) has achieved remarkable success for multiclass classification in recent years,yielding a promising solution for medical image classification where labeled data is scarce but unlabeled images are accessible.In the context of multi-label problems, however, SSL is still under-explored.In this work, we adapt FixMatch to the multi-label scenario, specifically focusing on CheXpert,a multi-label chest X-ray classification dataset which is imbalanced and only partially labeled.Leveraging distribution alignment, our proposed method, ML-FixMatch+DA, achieves solid performance gains in SSL tasks(AUC: +2.6%) and in a missing label scenario (AUC: +1.9%).In contrast to previous work, we achieve a performance gain on CheXpert using FixMatch.We show that in contrast to multiclass FixMatch, where distribution alignment is optional,it is essential for multi-label FixMatch to handle class imbalance and generate reliable (positive and negative) pseudo-labels.Our pseudo-label selection is based on a single threshold for all classes and handles imbalance with no prior knowledge on label distributions.Our adaptation keeps the simplicity of the original multiclass FixMatch with no added hyperparameters (even for imbalanced data)and demonstrates the feasibility of simple SSL for multi-label problems, filling a crucial gap in the literature.",
keywords = "semi supervised learning, CheXpert, multi-label classification",
author = "Sontje Ihler and Felix Kuhnke and Timo Kuhlgatz and Thomas Seel",
year = "2024",
month = jun,
day = "1",
language = "English",
pages = "2295--2304",
booktitle = "Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops",
}