An MRI Atlas of the Human Fetal Brain: Reference and Segmentation Tools for Fetal Brain MRI Analysis

Mahdi Bagheri, Clemente Velasco-Annis, Jian Wang, Razieh Faghihpirayesh, Shadab Khan, Camilo Calixto, Camilo Jaimes, Lana Vasung, Abdelhakim Ouaalam, Onur Afacan, Simon K. Warfield, Caitlin K. Rollins, Ali Gholipour · 2025

Read An MRI Atlas of the Human Fetal Brain: Reference and Segmentation Tools for Fetal Brain MRI Analysis by Mahdi Bagheri, Clemente Velasco-Annis, Jian Wang, Razieh Faghihpirayesh, Shadab Khan, Camilo Calixto, Camilo Jaimes, Lana Vasung, Abdelhakim Ouaalam, Onur Afacan, Simon K. Warfield, Caitlin K. Rollins, Ali Gholipour (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-15034.txt.

Introduction

Mahdi Bagheri and co-authors published a paper in 2025 presenting a new atlas of the fetal brain from MRI scans. Knowing how the brain grows before birth helps doctors spot typical and atypical growth. The atlas covers weeks twenty-one to thirty-seven of pregnancy. It was built from scans of fetuses with typical brain growth.

The atlas adds fine tissue maps, short-lived white matter parts, and a split into many named regions. It offers more detail than the group's earlier atlas from 2017. The team also released the scan data used to build it, plus a learning-based tool that marks tissues in new scans on its own. Both are shared for wider research use.

Catalog record

CREATORMahdi Bagheri, Clemente Velasco-Annis, Jian Wang, Razieh Faghihpirayesh, Shadab Khan, Camilo Calixto, Camilo Jaimes, Lana Vasung, Abdelhakim Ouaalam, Onur Afacan, Simon K. Warfield, Caitlin K. Rollins, Ali Gholipour
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE47 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-15034

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

An MRI Atlas of the Human Fetal Brain: Reference and Segmentation Tools for Fetal Brain MRI Analysis Mahdi Bagheri, Clemente Velasco-Annis, Jian Wang, Razieh Faghihpirayesh, Shadab Khan, Camilo Calixto, Camilo Jaimes, Lana Vasung, Abdelhakim Ouaalam, Onur Afacan, Simon K. Warfield, Caitlin K. Rollins, Ali Gholipour Abstract Accurate characterization of in-utero brain development is essential for understanding typical and atypical neurodevelopment. Building upon previous efforts to construct spatiotemporal fetal brain MRI atlases, we present the CRL-2025 fetal brain atlas, which is a spatiotemporal (4D) atlas of the developing fetal brain between 21 and 37 gestational weeks. This atlas is constructed from carefully processed MRI scans of 160 fetuses with typically-developing brains using a diffeomorphic deformable registration framework integrated with kernel regression on age. CRL-2025 uniquely includes detailed tissue segmentations, transient white matter compartments, and parcellation into 126 anatomical regions. This atlas offers significantly enhanced anatomical details over the CRL-2017 atlas, and is released along with the CRL diffusion MRI atlas with its newly created tissue segmentation and labels as well as deep learning-based multiclass segmentation models for fine-grained fetal brain MRI segmentation. The CRL-2025 atlas and its associated tools provide a robust and scalable platform for fetal brain MRI segmentation, groupwise analysis, and early neurodevelopmental research, and these materials are publicly released to support the broader research community. Background and Summary Human cognition and behavior are largely rooted in prenatal and early postnatal brain development, a critical and vulnerable stage of maturation [ 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 ] . During this time, the brain undergoes a dynamic series of neurodevelopmental processes, including neuronal proliferation, neuronal migration, axonal development and retraction (or reorganization), synaptogenesis and apoptosis, the onset of myelination, and the formation of transient developmental zones within the telencephalic wall [ 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 ] . These compartments provide the scaffolding that lays the foundation for the permanent cortical architecture seen in the adult brain. As a result, accurately characterizing early brain development in utero is essential to understand typical neurodevelopment and identify deviations from normal development. Fetal brain Magnetic Resonance Imaging (MRI), including quantitative MRI analysis enables the study of these processes in vivo , but requires the incorporation of accurate brain atlases as templates for spatial normalization and as references defining normative spatiotemporal developmental trajectories. Consequently, significant efforts have been made to build early brain development atlases [ 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 ] using MRI, a modality that provides unmatched images of brain anatomy throughout the human lifespan including, in specific, the fetal period [ 27 , 28 ] . Constructing atlases from fetal MRI presents distinct challenges compared to adult atlas generation, primarily due to two factors. First, there are technical challenges in scanning fetuses with MRI. MRI is very sensitive to motion, and fetuses move near continuously during MRI acquisitions [ 29 ] . Intermittent fetal and maternal respiratory movements prohibit precise spatiotemporal encoding that is necessary for real, high-resolution 3D MRI of the fetal brain anatomy. This challenge is exacerbated by the low level of MRI signal that is received from MRI body coils from the small fetal anatomy. Despite these challenges, the landscape of fetal MRI has changed dramatically in the past decade by the significant advances in retrospective slice-to-volume reconstruction (SVR) techniques [ 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 40 , 41 ] and the subsequent development of fetal brain MRI atlases and processing tools [ 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 ] . Second, the brain structure and function change rapidly and dramatically during the fetal period of development. As a result, single static or 3D atlases are insufficient to capture these dynamic processes. Consequently, atlases covering this period should be spatiotemporal (i.e., dynamic or 4D) to accurately reflect the continuous and age-dependent nature of fetal brain development. The construction of digital spatiotemporal MRI atlases of early brain development is relatively new: Kuklisova-Murgasova et al. [ 42 ] developed a 4D probabilistic atlas of early brain growth from in-vivo MRI of 142 preterm infants in the 29 to 44 weeks post-menstrual age. They used pairwise affine registration of anatomy with kernel regression in age for atlas construction. Serag et al. [ 20 ] used a non-rigid registration approach based on Bspline free-form deformations (FFD) [ 43 ] and showed a marked improvement over the use of affine registration in atlas construction. Makropoulos et al. [ 44 ] used a similar approach to construct a probabilistic spatiotemporal atlas of the neonatal brain from 420 segmented MRIs of neonates (including preterm neonates) scanned between 27 to 45 weeks post-menstrual age. To improve the FFD-based 4D atlas construction framework, Schuh et al. [ 45 ] developed diffeomorphic registration based on the Log-Euclidean mean of inverse consistent FFD transformations. For the fetal period of brain development, a recent review [ 46 ] found 18 atlases presented in the literature, of which 12 are publicly available. Most of these atlases characterize the anatomy (structure) of the fetal brain based on 3D-reconstructed T2-weighted (T2w) MRI, but a few atlases also characterize the microstructure of the fetal brain based on diffusion-weighted MRI (dMRI) or the cortical surface based on cortical surface meshes reconstructed from anatomical MRI. The first fetal MRI atlas was developed by Habas et al. [ 19 ] , based on 20 normal-appearing fetal brains scanned between 21–24 gestational weeks (GAs), offering age-specific T2w MRI templates and tissue probability maps of key brain structures such as cortical gray matter (cGM), white matter (WM), germinal matrix, and lateral ventricles (LV), using manual segmentation followed by groupwise registration and polynomial modeling of structural changes. Serag et al. [ 20 ] used non-rigid FFD-based registration along with adaptive kernel regression on age to build a spatiotemporal atlas of fetal brain anatomy in a much broader age range of 23 to 37 gestational weeks. This atlas was built by using 3D-reconstructed [ 34 ] T2w brain MRI images of 80 typically-developing fetuses. Dittrich et al. [ 21 ] created an atlas from 32 fetuses between 20–30 GA using a semi-supervised method to reduce manual annotation by segmenting from partially labeled data. Gholipour et al. [ 22 ] built a spatiotemporal atlas of the fetal brain anatomy from super-resolution reconstructed [ 32 ] fetal brain MRIs of 81 healthy fetuses, covering 21-38 GA. This atlas, referred to as the CRL (Computational Radiology Lab) Fetal Brain Atlas, was built using kernel regression on age integrated into an ANTS [ 47 ] group-wise symmetric diffeomorphic deformable registration framework. The CRL atlas uniquely offers detailed labels and tissue segmentations, and has been used as the standard space in several fetal SVR tools such as NiftyMIC [ 38 ] , MIALSRTK [ 37 ] , and NeSVoR [ 41 ] , and was used as a reference for comparison to the fetal brain ultrasound atlas developed by Namburete et al. [ 48 ] . Urru et

Read the complete text