Back to Projects

AI-enabled Imaging

Exploring early fetal brain development: a deep learning approach

Project ID: 2021_041

1st Supervisor: Maria Deprez, King’s College London
Second Supervisor and Clinical Champion: Alexander Hammers, King’s College London
Additional supervisor: J-Donald Tournier, King’s College London

Aim of the PhD Project:

In this project we will develop deep learning image analysis tools to enable understanding of early brain development as depicted by multi-modal MRI. We will develop methods to delineate transient and emerging brain structures and characterise their microstructure in average healthy brain development as well as on individual level. We will extract relevant biomarkers of normal and abnormal early brain development, including anatomy known to be involved in epilepsy.

Project Description / Background:

During the second half of pregnancy the brain undergoes rapid development. Important changes include neuronal migration, formation of white matter tracts, the onset of myelination, cortical folding and elaboration of cortical microstructure (Fig.1). Understanding precise timing and variation of these developmental processes is essential for deeper understanding of normal and atypical brain development [1,2]. Building models of maturing fetal brain requires image analysis tools such as registration and segmentation, but existing standard approaches often fail on brain MRI of young foetuses, due to maturational changes and presence of transient structures that disappear by birth. Though some tailored image analysis tools for early fetal brain MRI have been proposed [3], they generally focus on a single modality, simple diffusion acquisitions and have shown limited success with delineation of developing white matter tracts. We have recently acquired and reconstructed high quality advanced fetal diffusion MRI [4] (Fig. 1) as a part of the Developing Human Connectome project that together with structural T1 and T2 weighted scans offer new opportunities for detailed multi-modal characterisation of early fetal brain development.

The aim of this project is to fill this gap by developing novel artificial intelligence methods for image analysis of rapid early fetal brain development as observed by multi-modal MRI to

  • Perform segmentation of anatomical brain structures during early fetal development, including transient anatomy and developing white matter tracts, and evaluate their location, shape and microstructure to estimate stages of development
  • Derive quantitative indices of structural development for brain regions like hippocampus (rotation/folding indices) and Sylvian fissure (opercularisation indices), known to be involved in Epilepsy
  • accurately correct intensity artefacts and evaluate local image quality in individual fetal structural and diffusion MRI scans to allow accurate modelling of early brain development and individual comparisons

The novel tools for analysis and staging of the early brain development will form basis for deeper understanding of normal human brain development and its deviations in various conditions, such as Autism, risk of preterm birth, Down syndrome and epilepsy.

The project would suit a candidate from a quantitative discipline, interested in image analysis, machine learning, deep learning as well as early brain development.

Maturing fetal brain as shown by structural MRI (top) and diffusion MRI (bottom)

Figure 1: Maturing fetal brain as shown by structural MRI (top) and diffusion MRI (bottom).


  1. Jakab, A: Developmental Pathoconnectomics and Advanced Fetal MRI, Topics in MRI, 2019
  2. Biegon, A.,Hoffmann, C. Quantitative magnetic resonance imaging of the fetal brain in utero: Methods and applications. World journal of radiology, 6(8), 2014.
  3. Studholme, C. Mapping the developing human brain in utero using quantitative MR imaging techniques. Seminars in perinatology 39(2), 2015.
  4. Deprez, M., et al. Higher Order Spherical Harmonics Reconstruction of Fetal Diffusion MRI With Intensity Correction. IEEE Transactions on Medical Imaging 39(4), 2020.

Back to Projects