Domain generalization in fetal brain MRI segmentation \\with multi-reconstruction augmentation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: de Dumast, Priscille, Cuadra, Meritxell Bach
Natura: Preprint
Pubblicazione: 2022
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909365715337216
author de Dumast, Priscille
Cuadra, Meritxell Bach
author_facet de Dumast, Priscille
Cuadra, Meritxell Bach
contents Quantitative analysis of in utero human brain development is crucial for abnormal characterization. Magnetic resonance image (MRI) segmentation is therefore an asset for quantitative analysis. However, the development of automated segmentation methods is hampered by the scarce availability of fetal brain MRI annotated datasets and the limited variability within these cohorts. In this context, we propose to leverage the power of fetal brain MRI super-resolution (SR) reconstruction methods to generate multiple reconstructions of a single subject with different parameters, thus as an efficient tuning-free data augmentation strategy. Overall, the latter significantly improves the generalization of segmentation methods over SR pipelines.
format Preprint
id arxiv_https___arxiv_org_abs_2211_14282
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Domain generalization in fetal brain MRI segmentation \\with multi-reconstruction augmentation
de Dumast, Priscille
Cuadra, Meritxell Bach
Image and Video Processing
Computer Vision and Pattern Recognition
Quantitative analysis of in utero human brain development is crucial for abnormal characterization. Magnetic resonance image (MRI) segmentation is therefore an asset for quantitative analysis. However, the development of automated segmentation methods is hampered by the scarce availability of fetal brain MRI annotated datasets and the limited variability within these cohorts. In this context, we propose to leverage the power of fetal brain MRI super-resolution (SR) reconstruction methods to generate multiple reconstructions of a single subject with different parameters, thus as an efficient tuning-free data augmentation strategy. Overall, the latter significantly improves the generalization of segmentation methods over SR pipelines.
title Domain generalization in fetal brain MRI segmentation \\with multi-reconstruction augmentation
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2211.14282