EUVIMEDEUVIMED entdecken
Crossref

Contrastive Registration for Unsupervised Medical Image Segmentation

Lihao Liu, Angelica I Aviles-Rivero, Carola-Bibiane Schonlieb

IEEE Transactions on Neural Networks and Learning Systems · 2025-01 · 36 (1) · 147-159

DOI: 10.1109/tnnls.2023.3332003 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Neural Networks and Learning Systems, 2025-01. Verfasst von Lihao Liu und 2 weiteren Autor:innen. Hauptthemen laut MEDLINE: Neural Networks, Computer, Unsupervised Machine Learning, Image Processing, Computer-Assisted, Diagnostic Imaging. Die Arbeit wurde bisher 48-mal zitiert (Crossref) und nennt 72 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Medical image segmentation is an important task in medical imaging, as it serves as the first step for clinical diagnosis and treatment planning. While major success has been reported using deep learning supervised techniques, they assume a large and well-representative labeled set. This is a strong assumption in the medical domain where annotations are expensive, time-consuming, and inherent to human bias. To address this problem, unsupervised segmentation techniques have been proposed in the literature. Yet, none of the existing unsupervised segmentation techniques reach accuracies that come even near to the state-of-the-art of supervised segmentation methods. In this work, we present a novel optimization model framed in a new convolutional neural network (CNN)-based contrastive registration architecture for unsupervised medical image segmentation called CLMorph. The core idea of our approach is to exploit image-level registration and feature-level contrastive learning, to perform registration-based segmentation. First, we propose an architecture to capture the image-to-image transformation mapping via registration for unsupervised medical image segmentation. Second, we embed a contrastive learning mechanism in the registration architecture to enhance the discriminative capacity of the network at the feature level. We show that our proposed CLMorph technique mitigates the major drawbacks of existing unsupervised techniques. We demonstrate, through numerical and visual experiments, that our technique substantially outperforms the current state-of-the-art unsupervised segmentation methods on two major medical image datasets.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Neural Networks, ComputerUnsupervised Machine LearningImage Processing, Computer-AssistedDiagnostic ImagingHumansAlgorithmsDeep LearningDatabases, Factual

Zitieren

APA

Liu, L., Aviles-Rivero, A. I., & Schonlieb, C. B. (2025). Contrastive Registration for Unsupervised Medical Image Segmentation. IEEE Transactions on Neural Networks and Learning Systems, 36(1), 147-159. https://doi.org/10.1109/tnnls.2023.3332003

Vancouver

Liu L, Aviles-Rivero AI, Schonlieb CB. Contrastive Registration for Unsupervised Medical Image Segmentation. IEEE Transactions on Neural Networks and Learning Systems. 2025;36(1):147-159. doi:10.1109/tnnls.2023.3332003

BibTeX
@article{Liu2025,
  author = {Liu, Lihao and Aviles-Rivero, Angelica I and Schonlieb, Carola-Bibiane},
  title = {Contrastive Registration for Unsupervised Medical Image Segmentation},
  journal = {IEEE Transactions on Neural Networks and Learning Systems},
  year = {2025},
  volume = {36},
  number = {1},
  pages = {147--159},
  doi = {10.1109/tnnls.2023.3332003}
}
Kostenlos starten →