Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
167 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Improving cross-domain brain tissue segmentation in fetal MRI with synthetic data (2403.15103v1)

Published 22 Mar 2024 in eess.IV, cs.CV, and cs.LG

Abstract: Segmentation of fetal brain tissue from magnetic resonance imaging (MRI) plays a crucial role in the study of in utero neurodevelopment. However, automated tools face substantial domain shift challenges as they must be robust to highly heterogeneous clinical data, often limited in numbers and lacking annotations. Indeed, high variability of the fetal brain morphology, MRI acquisition parameters, and superresolution reconstruction (SR) algorithms adversely affect the model's performance when evaluated out-of-domain. In this work, we introduce FetalSynthSeg, a domain randomization method to segment fetal brain MRI, inspired by SynthSeg. Our results show that models trained solely on synthetic data outperform models trained on real data in out-ofdomain settings, validated on a 120-subject cross-domain dataset. Furthermore, we extend our evaluation to 40 subjects acquired using lowfield (0.55T) MRI and reconstructed with novel SR models, showcasing robustness across different magnetic field strengths and SR algorithms. Leveraging a generative synthetic approach, we tackle the domain shift problem in fetal brain MRI and offer compelling prospects for applications in fields with limited and highly heterogeneous data.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (35)
  1. O. M. Benkarim et al., “Toward the automatic quantification of in utero brain development in 3d structural MRI: A review,” Human Brain Mapping, vol. 38, p. 2772–2787, Feb. 2017.
  2. A. Jakab et al., “Emerging magnetic resonance imaging techniques in open spina bifida in utero,” European Radiology Experimental, vol. 5, no. 1, 2021.
  3. J. Aviles Verdera et al., “Reliability and feasibility of low-field-strength fetal MRI at 0.55 t during pregnancy,” Radiology, vol. 309, no. 1, 2023.
  4. M. C. Cortes-Albornoz et al., “MR insights into fetal brain development: what is normal and what is not,” Pediatric Radiology, pp. 1–11, 2024.
  5. J. Wu et al., “Age-specific structural fetal brain atlases construction and cortical development quantification for chinese population,” NeuroImage, vol. 241, 2021.
  6. C. M. Pfeifer et al., “MRI depiction of fetal brain abnormalities,” Acta Radiologica Open, vol. 8, Dec. 2019.
  7. F. Vahedifard et al., “Automatic ventriculomegaly detection in fetal brain MRI: A step-by-step deep learning model for novel 2d-3d linear measurements,” Diagnostics, vol. 13, no. 14, 2023.
  8. H. Guan and M. Liu, “Domain adaptation for medical image analysis: A survey,” IEEE Transactions on Biomedical Engineering, no. 3, 2022.
  9. R. Lin et al., “Cross-age and cross-site domain shift impacts on deep learning-based white matter fiber estimation in newborn and baby brains,” arXiv preprint arXiv:2312.14773, 2023.
  10. T. Sanchez et al., “FetMRQC: Automated quality control for fetal brain MRI,” in Perinatal, Preterm and Paediatric Image Analysis, (Cham), pp. 3–16, Springer Nature Switzerland, 2023.
  11. J. Xu et al., “NeSVoR: Implicit neural representation for slice-to-volume reconstruction in MRI,” IEEE Transactions on Medical Imaging, 2023.
  12. K. Payette et al., “Multi-center fetal brain tissue annotation (FeTA) challenge 2022 results,” arXiv preprint arXiv:2402.09463, 2024.
  13. P. de Dumast et al., “Synthetic magnetic resonance images for domain adaptation: Application to fetal brain tissue segmentation,” in 2022 IEEE 19th International Symposium on Biomedical Imaging (ISBI), IEEE, 2022.
  14. Z. Xu et al., “Asc: Appearance and structure consistency for unsupervised domain adaptation in fetal brain mri segmentation,” in Medical Image Computing and Computer Assisted Intervention – MICCAI, 2023.
  15. N. Karani et al., “A lifelong learning approach to brain mr segmentation across scanners and protocols,” in MICCAI, 2018.
  16. D. Li et al., “Learning to generalize: meta-learning for domain generalization,” in Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence and Thirtieth Innovative Applications of Artificial Intelligence Conference and Eighth AAAI Symposium on Educational Advances in Artificial Intelligence, AAAI Press, 2018.
  17. K. Zhou et al., “Domain generalization with mixstyle,” in International Conference on Learning Representations, 2021.
  18. F. Hu et al., “Image harmonization: A review of statistical and deep learning methods for removing batch effects and evaluation metrics for effective harmonization,” NeuroImage, vol. 274, 2023.
  19. D. Li et al., “Episodic training for domain generalization,” in 2019 IEEE/CVF International Conference on Computer Vision (ICCV), (Los Alamitos, CA, USA), pp. 1446–1455, IEEE Computer Society, 2019.
  20. Q. Liu et al., “Shape-aware meta-learning for generalizing prostate MRI segmentation to unseen domains,” in MICCAI, 2020.
  21. J. P. Marques et al., “Low-field MRI: An MR physics perspective,” Journal of magnetic resonance imaging, vol. 49, no. 6, pp. 1528–1542, 2019.
  22. C. Ouyang et al., “Causality-inspired single-source domain generalization for medical image segmentation,” IEEE Transactions on Medical Imaging, vol. 42, no. 4, p. 1095–1106, 2023.
  23. H. Li et al., “Frequency-mixed single-source domain generalization for medical image segmentation,” in Medical Image Computing and Computer Assisted Intervention – MICCAI 2023, Springer Nature Switzerland, 2023.
  24. B. Billot et al., “Synthseg: domain randomisation for segmentation of brain scans of any contrast and resolution,” arXiv:2107.09559, 2021.
  25. B. Billot et al., “Robust machine learning segmentation for large-scale analysis of heterogeneous clinical brain MRI datasets,” Proceedings of the National Academy of Sciences, vol. 120, no. 9, 2023.
  26. J. Tremblay et al., “Training deep networks with synthetic data: Bridging the reality gap by domain randomization,” in Proceedings of the IEEE conference on computer vision and pattern recognition workshops, 2018.
  27. K. Payette et al., “An automatic multi-tissue human fetal brain segmentation benchmark using the fetal tissue annotation dataset,” Scientific Data, vol. 8, no. 1, 2021.
  28. K. Payette and A. Jakab, “Fetal tissue annotation dataset feta,” 2021.
  29. S. Tourbier et al., “Medical-image-analysis-laboratory/mialsuperresol- utiontoolkit: MIAL super-resolution toolkit v2.0.1.” Zenodo, 2020.
  30. M. Kuklisova-Murgasova et al., “Reconstruction of fetal brain MRI with intensity matching and complete outlier removal,” Medical image analysis, vol. 16, no. 8, pp. 1550–1564, 2012.
  31. A. U. Uus et al., “Automated 3d reconstruction of the fetal thorax in the standard atlas space from motion-corrupted MRI stacks for 21–36 weeks ga range,” Medical image analysis, vol. 80, p. 102484, 2022.
  32. A. P. Dempster et al., “Maximum likelihood from incomplete data via the em algorithm,” Journal of the Royal Statistical Society: Series B (Methodological), vol. 39, no. 1, pp. 1–22, 1977.
  33. H. Lajous et al., “A fetal brain magnetic resonance acquisition numerical phantom (FaBiAN),” Scientific Reports, vol. 12, May 2022.
  34. R. Valabregue et al., “Comprehensive analysis of synthetic learning applied to neonatal brain MRI segmentation,” arXiv:2309.05306, 2023.
  35. A. Paszke et al., “PyTorch: An imperative style, high-performance deep learning library,” 2019.
Citations (1)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now