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.

To investigate whether hyperpolarised xenon-129 MRI (HXeMRI) enables regional and physiological resolution of diffusing capacity limitations in chronic obstructive pulmonary disease (COPD), we evaluated 34 COPD subjects and 11 healthy volunteers. We report significant correlations between airflow abnormality quantified by HXeMRI and per cent predicted forced expiratory volume in 1 s; HXeMRI gas transfer capacity to red blood cells and carbon monoxide diffusion capacity (%DLCO); and HXeMRI gas transfer capacity to interstitium and per cent emphysema quantified by multidetector chest CT. We further demonstrate the capability of HXeMRI to distinguish varying pathology underlying COPD in subjects with low %DLCO and minimal emphysema.

Idiopathic pulmonary fibrosis, a pattern of interstitial lung disease, is often clinically unpredictable in its progression. This paper presents hyperpolarized Xenon-129 chemical shift imaging as a noninvasive, nonradioactive method of probing lung physiology as well as anatomy to monitor subtle changes in subjects with IPF. Twenty subjects, nine healthy and eleven IPF, underwent HP Xe-129 ventilation MRI and 3D-SBCSI. Spirometry was performed on all subjects before imaging, and DLCO and hematocrit were measured in IPF subjects after imaging. Images were post-processed in MATLAB and segmented using ANTs. IPF subjects exhibited, on average, higher Tissue/Gas ratios and lower RBC/Gas ratios compared with healthy subjects, and quantitative maps were more heterogeneous in IPF subjects. The higher ratios are likely due to fibrosis and thickening of the pulmonary interstitium. T2* relaxation was longer in IPF subjects and corresponded with hematocrit scores, although the mechanism is not well understood. A lower chemical shift in the red blood cell spectroscopic peak correlated well with a higher Tissue/RBC ratio and may be explained by reduced blood oxygenation. Tissue/RBC also correlated well, spatially, with areas of fibrosis in HRCT images. These results may help us understand the underlying mechanism behind gas exchange impairment and disease progression.