Metadata-Aligned 3D MRI Representations for Contrast Understanding and Quality Control

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Avci, Mehmet Yigit, Borges, Pedro, Fernandez, Virginia, Wright, Paul, Yigitsoy, Mehmet, Ourselin, Sebastien, Cardoso, Jorge
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918181816238080
author Avci, Mehmet Yigit
Borges, Pedro
Fernandez, Virginia
Wright, Paul
Yigitsoy, Mehmet
Ourselin, Sebastien
Cardoso, Jorge
author_facet Avci, Mehmet Yigit
Borges, Pedro
Fernandez, Virginia
Wright, Paul
Yigitsoy, Mehmet
Ourselin, Sebastien
Cardoso, Jorge
contents Magnetic Resonance Imaging suffers from substantial data heterogeneity and the absence of standardized contrast labels across scanners, protocols, and institutions, which severely limits large-scale automated analysis. A unified representation of MRI contrast would enable a wide range of downstream utilities, from automatic sequence recognition to harmonization and quality control, without relying on manual annotations. To this end, we introduce MR-CLIP, a metadata-guided framework that learns MRI contrast representations by aligning volumetric images with their DICOM acquisition parameters. The resulting embeddings shows distinct clusters of MRI sequences and outperform supervised 3D baselines under data scarcity in few-shot sequence classification. Moreover, MR-CLIP enables unsupervised data quality control by identifying corrupted or inconsistent metadata through image-metadata embedding distances. By transforming routinely available acquisition metadata into a supervisory signal, MR-CLIP provides a scalable foundation for label-efficient MRI analysis across diverse clinical datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2511_00681
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Metadata-Aligned 3D MRI Representations for Contrast Understanding and Quality Control
Avci, Mehmet Yigit
Borges, Pedro
Fernandez, Virginia
Wright, Paul
Yigitsoy, Mehmet
Ourselin, Sebastien
Cardoso, Jorge
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Magnetic Resonance Imaging suffers from substantial data heterogeneity and the absence of standardized contrast labels across scanners, protocols, and institutions, which severely limits large-scale automated analysis. A unified representation of MRI contrast would enable a wide range of downstream utilities, from automatic sequence recognition to harmonization and quality control, without relying on manual annotations. To this end, we introduce MR-CLIP, a metadata-guided framework that learns MRI contrast representations by aligning volumetric images with their DICOM acquisition parameters. The resulting embeddings shows distinct clusters of MRI sequences and outperform supervised 3D baselines under data scarcity in few-shot sequence classification. Moreover, MR-CLIP enables unsupervised data quality control by identifying corrupted or inconsistent metadata through image-metadata embedding distances. By transforming routinely available acquisition metadata into a supervisory signal, MR-CLIP provides a scalable foundation for label-efficient MRI analysis across diverse clinical datasets.
title Metadata-Aligned 3D MRI Representations for Contrast Understanding and Quality Control
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2511.00681