Automatic Measurement of Fetal Brain Development from Magnetic Resonance Imaging: New Reference Data

Abstract

Background: Accurate fetal brain volume estimation is of paramount importance in evaluating fetal development. The aim of this study was to develop an automatic method for fetal brain segmentation from magnetic resonance imaging (MRI) data, and to create for the first time a normal volumetric growth chart based on a large cohort. Subjects and Methods: A semi-automatic segmentation method based on Seeded Region Growing algorithm was developed and applied to MRI data of 199 typically developed fetuses between 18 and 37 weeks' gestation. The accuracy of the algorithm was tested against a sub-cohort of ground truth manual segmentations. A quadratic regression analysis was used to create normal growth charts. The sensitivity of the method to identify developmental disorders was demonstrated on 9 fetuses with intrauterine growth restriction (IUGR). Results: The developed method showed high correlation with manual segmentation (r² = 0.9183, p < 0.001) as well as mean volume and volume overlap differences of 4.77 and 18.13%, respectively. New reference data on 199 normal fetuses were created, and all 9 IUGR fetuses were at or below the third percentile of the normal growth chart. Discussion: The proposed method is fast, accurate, reproducible, user independent, applicable with retrospective data, and is suggested for use in routine clinical practice.

Normative volume measurements of the fetal intra-cranial compartments using 3D volume in utero MR imaging

“…Additionally, fetuses with a family history of abnormalities were also excluded. This is in contrast to other studies whose normative data is derived from fetuses who were either referred for MR imaging due to siblings with abnormalities, had suspected brain abnormality on USS that were subsequently excluded on MR imaging or have a normal brain examination but an abnormality affecting another anatomical area [ 6 , 13 , 30 ]. These cannot be considered a truly normal population, although one study [ 6 ] carried out postnatal assessments of the children studied to confirm the data reported was drawn from a normal population.…”

“…A further limitation of our method is the time required for manual segmentation (between 2 and 6 h) which restricts its application routinely in clinical practice. Whilst there has been a great deal of effort to develop automated segmentation methods by several groups [ 13 , 22 – 26 , 28 , 33 ] user input is required to increase precision and there has yet to be a proven clinical utility.…”

“…These three volumes summated constitute the total intracranial volume (TICV). It is now possible to acquire data in individual fetuses using ultrafast three-dimensional (3D) MR imaging [ 9 – 13 ] that permits the volume of these compartments to be measured after post-processing. This development could be important for accurate diagnosis because different types of fetal neuropathology are expected to affect the compartments in different ways.…”

Apparent diffusion coefficient of different areas of brain in foetuses with intrauterine growth restriction

“…Foetal brain magnetic resonance imaging (MRI) is used to provide distinctive information about the structural and functional status of foetuses [12]. Restricted brain growth, delayed brain gyration, or focal ischaemic/ haemorrhagic lesions could be investigated by conventional MRI; however, identifying diffuse ischaemic changes or earlier and subtle injuries remains challenging [4].…”

Automated Brain Masking of Fetal Functional MRI

“…They are increasingly utilized in image processing applications in both medical and non-medical settings [33,34]. In the context of fetal brain segmentation, prior work has investigated the application of CNN's to segment fetal structural T2-weighted volumes [35,36,37,38,16,39,40,41]. These models, however, were developed to segment the fetal brain from anatomical images, and do not translate to functional time series data.…”