Automated 3D Fetal Brain Segmentation Using an Optimized Deep Learning Approach.

BACKGROUND AND PURPOSE: MR imaging provides critical information about fetal brain growth and development. Currently, morphologic analysis primarily relies on manual segmentation, which is time-intensive and has limited repeatability. This work aimed to develop a deep learning-based automatic fetal brain segmentation method that provides improved accuracy and robustness compared with atlas-based methods. MATERIALS AND METHODS: A total of 106 fetal MR imaging studies were acquired prospectively from fetuses between 23 and 39 weeks of gestation. We trained a deep learning model on the MR imaging scans of 65 healthy fetuses and compared its performance with a 4D atlas-based segmentation method using the Wilcoxon signed-rank test. The trained model was also evaluated on data from 41 fetuses diagnosed with congenital heart disease. RESULTS: The proposed method showed high consistency with the manual segmentation, with an average Dice score of 0.897. It also demonstrated significantly improved performance (P < .001) based on the Dice score and 95% Hausdorff distance in all brain regions compared with the atlas-based method. The performance of the proposed method was consistent across gestational ages. The segmentations of the brains of fetuses with high-risk congenital heart disease were also highly consistent with the manual segmentation, though the Dice score was 7% lower than that of healthy fetuses. CONCLUSIONS: The proposed deep learning method provides an efficient and reliable approach for fetal brain segmentation, which outperformed segmentation based on a 4D atlas and has been used in clinical and research settings.

Regional Differences in Gamma-Aminobutyric Acid and Glutamate Concentrations in the Healthy Newborn Brain.

BACKGROUND AND PURPOSE: Gamma-aminobutyric acid and glutamate system disruptions may underlie neonatal brain injury. However, in vivo investigations are challenged by the need for special 1H-MR spectroscopy sequences for the reliable measurement of the neurotransmitters in this population. We used J-edited 1H-MR spectroscopy (Mescher-Garwood point-resolved spectroscopy) to quantify regional in vivo gamma-aminobutyric acid and glutamate concentrations during the early postnatal period in healthy neonates. MATERIALS AND METHODS: We prospectively enrolled healthy neonates and acquired Mescher-Garwood point-resolved spectroscopy spectra on a 3T MR imaging scanner from voxels located in the cerebellum, the right basal ganglia, and the right frontal lobe. CSF-corrected metabolite concentrations were compared for regional variations and cross-sectional temporal trends with advancing age. RESULTS: Fifty-eight neonates with acceptable spectra acquired at postmenstrual age of 39.1 (SD, 1.3) weeks were included for analysis. Gamma-aminobutyric acid (+ macromolecule) (2.56 [SD, 0.1]) i.u., glutamate (3.80 [SD, 0.2]), Cho, and mIns concentrations were highest in the cerebellum, whereas NAA (6.72 [SD, 0.2]), NAA/Cho, Cr/Cho, and Glx/Cho were highest in the basal ganglia. Frontal gamma-aminobutyric acid (1.63 [SD, 0.1]), Glx (4.33 [SD, 0.3]), Cr (3.64 [SD, 0.2]), and Cho concentrations were the lowest among the ROIs. Glx, NAA, and Cr demonstrated a significant adjusted increase with postmenstrual age (ß = 0.2-0.35), whereas gamma-aminobutyric acid and Cho did not. CONCLUSIONS: We report normative regional variations and temporal trends of in vivo gamma-aminobutyric acid and glutamate concentrations reflecting the functional and maturational status of 3 distinct brain regions of the neonate. These measures will serve as important normative values to allow early detection of subtle neurometabolic alterations in high-risk neonates.