Related Experiment Video
Updated: Sep 22, 2025
06:56
Human Fetal Blood Flow Quantification with Magnetic Resonance Imaging and Motion Compensation
Published on: January 7, 2021
2.5K
A Fetal Brain magnetic resonance Acquisition Numerical phantom (FaBiAN)
Hélène Lajous1,2, Christopher W Roy3, Tom Hilbert3,4,5
1Department of Radiology, Lausanne University Hospital (CHUV) and University of Lausanne (UNIL), Lausanne, Switzerland. helene.lajous@unil.ch.
Scientific Reports
|May 23, 2022
Summary
FaBiAN, an open-source fetal brain magnetic resonance imaging phantom, simulates realistic fetal brain development. This tool aids in validating advanced imaging techniques and deep learning for improved fetal neuroimaging analysis.
Area of Science:
- Neuroimaging
- Medical Physics
- Developmental Biology
Background:
- Accurate characterization of in utero human brain maturation is crucial for understanding neurodevelopmental trajectories and long-term health outcomes.
- Magnetic resonance imaging (MRI) is vital for studying fetal brain development, but limited high-quality data hinders advanced technique validation.
- Numerical phantoms offer a controlled environment with ground truth data to overcome limitations of clinical fetal MRI acquisition.
Purpose of the Study:
- To introduce FaBiAN, an open-source numerical phantom for simulating fetal brain MRI.
- To provide a realistic simulation of T2-weighted fast spin echo sequences, incorporating stochastic fetal movements.
- To demonstrate the utility of FaBiAN in evaluating super-resolution algorithms and supporting deep learning methods for fetal neuroimaging.
Main Methods:
- Development of FaBiAN, an open-source fetal brain MRI numerical phantom.
- Simulation of T2-weighted fast spin echo sequences with realistic fetal movements.
- Evaluation of a super-resolution algorithm using simulated motion-corrupted low-resolution MRI data against a synthetic high-resolution reference.
Main Results:
- FaBiAN generates realistic fetal brain MRI data comparable to clinical acquisitions throughout development.
- The phantom successfully aids in evaluating the robustness and optimizing the accuracy of super-resolution algorithms.
- Simulated data from FaBiAN can augment clinical datasets for deep learning-based fetal brain tissue segmentation.
Conclusions:
- FaBiAN is a valuable, flexible, and open-source tool for advancing fetal brain MRI research.
- The phantom facilitates the development and validation of sophisticated image processing techniques, including super-resolution and deep learning.
- FaBiAN contributes to overcoming data scarcity challenges in fetal neuroimaging, promoting more accurate analysis of brain maturation.
