Advances in automated fetal brain MRI segmentation and biometry: Insights from the FeTA 2024 challenge

Zalevskyi V, Sanchez T, Kaandorp M, Roulet M, Fajardo-Rojas D, Li L, Hutter J, Li HB, Barkovich MJ, Ji H, Wilhelmi L, Dändliker A, Steger C, Koob M, Gomez Y, Jakovčić A, Klaić M, Adžić A, Marković P, Grabarić G, Rados M, Aviles Verdera J, Kasprian G, Dovjak G, Gaubert-Rachmühl R, Aschwanden M, Zeng Q, Karimi D, Peruzzo D, Ciceri T, Longari G, Hamadache RE, Bouzid A, Lladó X, Chiarella S, Martí-Juan G, González Ballester MÁ, Castellaro M, Pinamonti M, Visani V, Cremese R, Sam K, Gaudfernau F, Ahir P, Parikh M, Zenk M, Baumgartner M, Maier-Hein K, Tianhong L, Hong Y, Longfei Z, Preloznik D, Špiclin Ž, Won Choi J, Li M, Fu J, Wang G, Jiang J, Tong L, Du B, Gondova A, You S, Im K, Qayyum A, Mazher M, Niederer SA, Jakab A, Licandro R, Payette K, Bach Cuadra M (2026)

Publication Type: Journal article

Publication year: 2026

Journal

Book Volume: 109

Article Number: 103941

DOI: 10.1016/j.media.2026.103941

Abstract

Accurate fetal brain tissue segmentation and biometric measurement are essential for monitoring neurodevelopment and detecting abnormalities in utero. The Fetal Tissue Annotation (FeTA) Challenges have established robust multi-center benchmarks for evaluating state-of-the-art segmentation methods. This paper presents the results of the 2024 challenge edition, which introduced three key innovations. First, we introduced a topology-aware metric based on the Euler characteristic difference (ED) to overcome the performance plateau observed with traditional metrics like Dice or Hausdorff distance (HD), as the performance of the best models in segmentation surpassed the inter-rater variability. While the best teams reached similar scores in Dice (0.81-0.82) and HD95 (2.1–2.3 mm), ED provided greater discriminative power: the winning method achieved an ED of 20.9, representing roughly a 50% improvement over the second- and third-ranked teams despite comparable Dice scores. Second, we introduced a new 0.55T low-field MRI test set, which, when paired with high-quality super-resolution reconstruction, achieved the highest segmentation performance across all test cohorts (Dice=0.86, HD95=1.69, ED=6.26). This provides the first quantitative evidence that low-cost, low-field MRI can match or surpass high-field systems in automated fetal brain segmentation. Third, the new biometry estimation task exposed a clear performance gap: although the best model reached a mean average percentage error (MAPE) of 7.72%, most submissions failed to outperform a simple gestational-age-based linear regression model (MAPE=9.56%), and all remained above inter-rater variability with a MAPE of 5.38%. Finally, by analyzing the top-performing models from FeTA 2024 alongside those from previous challenge editions, we identify ensembles of 3D nnU-Net trained on both real and synthetic data with both image- and anatomy-level augmentations as the most effective approaches for fetal brain segmentation. Our quantitative analysis reveals that acquisition site, super-resolution strategy, and image quality are the primary sources of domain shift, informing recommendations to enhance the robustness and generalizability of automated fetal brain analysis methods.

Authors with CRIS profile

Involved external institutions

Universitäts-Kinderspital Zürich Switzerland (CH) Universitat de Girona Spain (ES) University of Zagreb / Sveučilište u Zagrebu Croatia (HR) Universitat Pompeu Fabra (UPF) Spain (ES) Université Sorbonne Paris Cité France (FR) King’s College London United Kingdom (GB) Tel Aviv University Israel (IL) Scientific Institute for Rehabilitation Medicine Eugenio Medea (IRCCS) Italy (IT) Canon Medical Systems (China) Co., Ltd. China (CN) Universitätsklinikum Heidelberg Germany (DE) University of Zurich / Universität Zürich (UZH) Switzerland (CH) Harvard University United States (USA) (US) Lalbhai Dalpatbhai College of Engineering India (IN) Lausanne University Hospital / Centre hospitalier universitaire vaudois (CHUV) Switzerland (CH) University of Electronic Science and Technology of China (UESTC) / 电子科技大学 China (CN) Wuhan University China (CN) University College London (UCL) United Kingdom (GB) University of Bologna / Università di Bologna Italy (IT) Medizinische Universität Wien Austria (AT) Institut national de Recherche en Informatique et en Automatique (INRIA) France (FR) UCSF Benioff Children’s Hospital United States (USA) (US) Seoul National University (SNU) / 서울대학교 Korea, Republic of (KR) University of Ljubljana (UL) / Univerza v Ljubljani Slovenia (SI) Massachusetts General Hospital United States (USA) (US) Imperial College London / The Imperial College of Science, Technology and Medicine United Kingdom (GB) University of Padua / Universita degli Studi di Padova Italy (IT) University of Milano-Bicocca / Università degli Studi di Milano-Bicocca, UNIMIB Italy (IT)

How to cite

APA:

Zalevskyi, V., Sanchez, T., Kaandorp, M., Roulet, M., Fajardo-Rojas, D., Li, L.,... Bach Cuadra, M. (2026). Advances in automated fetal brain MRI segmentation and biometry: Insights from the FeTA 2024 challenge. Medical Image Analysis, 109. https://doi.org/10.1016/j.media.2026.103941

MLA:

Zalevskyi, Vladyslav, et al. "Advances in automated fetal brain MRI segmentation and biometry: Insights from the FeTA 2024 challenge." Medical Image Analysis 109 (2026).

BibTeX: Download