Search

Background: Fetal ventriculomegaly is the the most common intracranial abnormality detected antenatally. When ventriculomegaly is mild and the only, isolated, abnormality detected (isolated mild ventriculomegaly (IMVM)) the prognosis is generally considered to be good. We aim to determine if there are features on in utero MRI (iuMRI) that can identify fetuses with IMVM who have lower risks of abnormal neurodevelopment outcome.\ud \ud \ud \ud Methods: We studied cases recruited into the MRI to enhance the diagnosis of fetal developmental brain abnormalities in utero (MERIDIAN) study, specifically those with: confirmed IMVM, 3D volume imaging of the fetal brain and neurodevelopmental outcomes at 3 years. We explored the influence of sex of the fetus, laterality of the ventriculomegaly and intracranial compartmental volumes in relation to neurodevelopmental outcome.\ud \ud \ud \ud Findings: Forty-two fetuses met the criteria (33 male and 9 female). There was no obvious correlation between fetal sex and the risk of poor neurodevelopmental outcome. Unilateral IMVM was present in 23 fetuses and bilateral IMVM in 19 fetuses. All fetuses with unilateral IMVM had normal neurodevelopmental outcomes, while only 12/19 with bilateral IMVM had normal neurodevelopmental outcomes. There was no obvious correlation between measure of intracranial volumes and risk of abnormal developmental outcomes.\ud \ud \ud \ud Interpretation: The most important finding is the very high chance of a good neurodevelopmental outcome observed in fetuses with unilateral IMVM, which is a potentially important finding for antenatal counselling. There does not appear to be a link between the volume of the ventricular system or brain volume and the risk of poor neurodevelopmental outcome.

In this work, we develop the Single-Input Multi-Output U-Net (SIMOU-Net), a hybrid network for foetal brain segmentation inspired by the original U-Net fused with the holistically nested edge detection (HED) network. The SIMOU-Net is similar to the original U-Net but it has a deeper architecture and takes account of the features extracted from each side output. It acts similar to an ensemble neural network, however, instead of averaging the outputs from several independently trained models, which is computationally expensive, our approach combines outputs from a single network to reduce the variance of predications and generalization errors. Experimental results using 200 normal foetal brains consisting of over 11,500 2D images produced Dice and Jaccard coefficients of 94.2 ± 5.9% and 88.7 ± 6.9%, respectively. We further tested the proposed network on 54 abnormal cases (over 3500 images) and achieved Dice and Jaccard coefficients of 91.2 ± 6.8% and 85.7 ± 6.6%, respectively.

Purpose\ud \ud The aim of this study is to provide normative data about the appearances and dimensions of the cavum septi pellucidi and vergae (CSPV) on in utero MR (iuMR) imaging in second and third trimester foetuses.\ud \ud \ud \ud Methods\ud \ud Two hundred normal foetuses (from a low-risk pregnancy, with normal ante-natal USS findings and no intracranial abnormality of iuMR) had iuMR imaging between 18 and 37 gestational weeks (gw). The anatomical features on those studies were compared with published atlases of post-mortem foetal brains. The length, width and volume of the CSPV were measured in all foetuses.\ud \ud \ud \ud Results\ud \ud The anatomy of the CSPV and its relationship with the corpus callosum and the fornices on iuMR imaging was comparable with post-mortem data at all gestational ages studied. The length of the CSPV increased throughout pregnancy, whereas the width and volume of CSPV reached a maximum between 29 and 31 gw and then showed a reduction later in pregnancy.\ud \ud \ud \ud Conclusion\ud \ud The iuMR imaging features of the CSPV and its close anatomical relations closely correspond to post-mortem data. The CSPV was patent in all cases but we have shown that closure commences in the midpart of the third trimester and advances in a posterior to anterior direction.

Purpose\ud To refine methods that assess structural brain abnormalities and calculate intracranial volumes in fetuses with congenital heart diseases (CHD) using in utero MR (iuMR) imaging. Our secondary objective was to assess the prevalence of brain abnormalities in this high-risk cohort and compare the brain volumes with normative values.\ud \ud Methods\ud We performed iuMR on 16 pregnant women carrying a fetus with CHD and gestational age ≥ 28-week gestation and no brain abnormality on ultrasonography. All cases had fetal echocardiography by a pediatric cardiologist. Structural brain abnormalities on iuMR were recorded. Intracranial volumes were made from 3D FIESTA acquisitions following manual segmentation and the use of 3D Slicer software and were compared with normal fetuses. Z scores were calculated, and regression analyses were performed to look for differences between the normal and CHD fetuses.\ud \ud Results\ud Successful 2D and 3D volume imaging was obtained in all 16 cases within a 30-min scan. Despite normal ultrasonography, 5/16 fetuses (31%) had structural brain abnormalities detected by iuMR (3 with ventriculomegaly, 2 with vermian hypoplasia). Brain volume, extra-axial volume, and total intracranial volume were statistically significantly reduced, while ventricular volumes were increased in the CHD cohort.\ud \ud Conclusion\ud We have shown that it is possible to perform detailed 2D and 3D studies using iuMR that allow thorough investigation of all intracranial compartments in fetuses with CHD in a clinically appropriate scan time. Those fetuses have a high risk of structural brain abnormalities and smaller brain volumes even when brain ultrasonography is normal.

In this article we provide an overview of fetal brain development, describe the range of more common fetal neuropathology, and discuss the relevance of in utero MR (iuMR). Although ultrasonography remains the mainstay of fetal brain imaging, iuMR imaging is both feasible and safe, but presents several challenges. We discuss those challenges, the techniques employed to overcome them, and new approaches that may extend the clinical applicability of fetal iuMR.

Objectives In utero Magnetic Resonance (iuMR) imaging to diagnose fetal brain abnormalities has been established and is supported by meta-analyses of retrospective and prospective studies. In this paper we describe and classify the iuMR errors made in the largest diagnostic accuracy study to date (MERIDIAN). We also correlate the error rates and types with the prior experience of the reporting radiologists in order to inform how to provide a national programme with the best diagnostic accuracy achievable. Methods The MERIDIAN cohort of 570 fetuses formed the basis of this study and included 40 cases with a confirmed diagnostic error, compared with the Outcome Reference Diagnosis. Analysis included the potential clinical effect of the error and classification of error type through an Expert Neuroradiological Panel re-reporting the study. Assessments were made regarding radiologists experience prior to MERIDIAN. Results The overall confirmed error rate for iuMR was 7·0% and it was considered that there would have been an adverse effect on prognostic information in 22/40 cases if the iuMR had informed counselling. The experienced central reporter made statistically significant fewer errors than the less experienced non-central reporters (3·8% v 11·0%) and the central reporter made fewer clinically significant errors. Furthermore, the type of cognitive errors differed between central and non-central reporters. Conclusions Although iuMR imaging improves the diagnostic accuracy of detecting fetal brain abnormalities there remains a substantial error rate, which can have major clinical significance. We have shown that error rates are lower for more experienced reporting radiologists with fewer potential deleterious clinical implications. We discuss the implications of these findings in terms of providing a uniform national service.

PURPOSE: In this paper, I will report the range of appearances of schizencephaly in children and fetuses by reviewing a 10-year experience from a single centre and detail classification systems for the different forms of schizencephaly. This will lead to re-assessment of possible aetiological and mechanistic causes of schizencephaly. METHODS: All cases of pediatric and fetal schizencephaly were located on the local database between 2007 and 2016 inclusive. The studies were reviewed for the presence, location and type of schizencephaly, as well as the state of the (cavum) septum pellucidum, the location of the fornices and the presence of other brain abnormalities. RESULTS: The review included 21 children and 11 fetuses with schizencephaly. Schizencephaly (type 1) was found in 9% of children but no fetuses, schizencephaly (type 2) was present in 67% of the pediatric cases and in 45% of fetuses, whilst schizencephaly (type 3) was present in approximately 24% of children and 55% of fetuses. Other brain abnormalities were found in 67% of children and 55% of fetuses. CONCLUSION: I have proposed a new system for classifying schizencephaly that takes into account all definitions of the abnormality in the literature. Using that approach, I have described the appearances and associations of pediatric and fetal cases of schizencephaly from a single centre. Review of the current literature appears to favour an acquired destructive aetiology for most cases of schizencephaly, and I have proposed a mechanism to explain the cortical formation abnormalities found consistently in and around areas of schizencephaly.

OBJECTIVE: To assess the diagnostic and clinical contribution of in utero magnetic resonance (iuMR) imaging in fetuses diagnosed with abnormalities of the posterior fossa as the only intracranial abnormality recognised on antenatal ultrasonography (USS). METHODS: We report a sub-group analysis of fetuses with abnormalities of the posterior fossa diagnosed on antenatal USS (with or without ventriculomegaly) from the MERIDIAN cohort who had iuMR imaging within 2 weeks of USS and outcome reference data were available. The diagnostic accuracy of USS and iuMR are reported as well as indicators of diagnostic confidence and effects on prognosis and clinical management. RESULTS: Abnormalities confined to the posterior fossa according to USS were found in 81 fetuses (67 with parenchymal and 14 with CSF-containing lesions). The overall diagnostic accuracy for detecting an isolated posterior fossa abnormality was 65% for USS and 88% for iuMR (difference = 22%, 95% CI: 14.0 to 30.5%, p

Purpose To evaluate the feasibility of directly imaging perfusion of human brain tissue by using magnetic resonance (MR) imaging with inhaled hyperpolarized xenon 129 ((129)Xe). Materials and Methods In vivo imaging with (129)Xe was performed in three healthy participants. The combination of a high-yield spin-exchange optical pumping (129)Xe polarizer, custom-built radiofrequency coils, and an optimized gradient-echo MR imaging protocol was used to achieve signal sensitivity sufficient to directly image hyperpolarized (129)Xe dissolved in the human brain. Conventional T1-weighted proton (hydrogen 1 [(1)H]) images and perfusion images by using arterial spin labeling were obtained for comparison. Results Images of (129)Xe uptake were obtained with a signal-to-noise ratio of 31 ± 9 and demonstrated structural similarities to the gray matter distribution on conventional T1-weighted (1)H images and to perfusion images from arterial spin labeling. Conclusion Hyperpolarized (129)Xe MR imaging is an injection-free means of imaging the perfusion of cerebral tissue. The proposed method images the uptake of inhaled xenon gas to the extravascular brain tissue compartment across the intact blood-brain barrier. This level of sensitivity is not readily available with contemporary MR imaging methods.

Objective: To assess the contribution of in utero magnetic resonance (iuMR) imaging in fetuses diagnosed with either agenesis of the corpus callosum or hypogenesis of the corpus callosum (grouped as failed commissuration) on antenatal ultrasonography (USS) from the MERIDIAN cohort.\ud Methods: We report a sub-group analysis of fetuses with failed commissuration diagnosed on USS (with or without ventriculomegaly) from the MERIDIAN study who had iuMR imaging within 2 weeks of USS and outcome reference data were available. The diagnostic accuracy of USS and iuMR are reported as well as indicators of diagnostic confidence and effects on prognosis/clinical management.\ud Results: 79 fetuses with failed commissuration are reported (55 with agenesis and 24 with hypogenesis as the USS diagnoses). The diagnostic accuracy for detecting ‘failed commissuration’ as a group label was 34.2% for USS and 94.9% for iuMR (difference = 60.7%, 95% confidence interval 47.6% to 73.9%, p

Objective: Interpretation of magnetic resonance (MR) imaging of the fetal brain in utero is primarily undertaken using 2D images to provide anatomical information about structural abnormalities. It is now possible to obtain 3D image acquisitions that allow measurement of fetal brain volumes that are potentially useful clinically. The aim of our current work is to provide reference values of total brain volumes obtained from a cohort of low risk fetuses with no abnormalities on ante-natal ultrasonography and in utero MR imaging.\ud Method: Images from volume MR acquisitions of 132 fetuses were used to extract brain volumes by manual segmentation. Reproducibility and reliability were assessed by analysis of the results of two subgroups who had repeated measurements made by the primary and a secondary observer.\ud Results: Intra-observer and inter-observer agreement was high with no statistically significant differences between and within observers (p = 0.476 and p = 0.427, respectively). The results of the brain volume assessments are presented graphically with mean and 95% prediction limits alongside estimates of normal growth rates.\ud Conclusion: We have shown that fetal brain volumes can be reliably extracted from in utero MR (iuMR) imaging 3D datasets with a high degree of reproducibility. The resultant data could potentially be used as a reference tool in the clinical setting.

BACKGROUND: In-utero MRI (iuMRI) has shown promise as an adjunct to ultrasound but the comparative diagnostic performance has been poorly defined. We aimed to assess whether the diagnostic accuracy and confidence of the prenatal diagnosis of fetal brain abnormalities is improved with iuMRI and assess the clinical impact and patient acceptability of iuMRI. METHODS: We did a multicentre, prospective, cohort study in the UK, at 16 fetal medicine centres, of pregnant women aged 16 years or older whose fetus had a brain abnormality detected by ultrasound at a gestational age of 18 weeks or more, had no contraindications to iuMRI, and consented to enter the study. Women carrying a fetus suspected of having a brain anomaly on ultrasound had iuMRI done within 14 days of ultrasound. The findings were reviewed by two independent panels and used to estimate diagnostic accuracy and confidence by comparison with outcome diagnoses. Changes in diagnosis, prognosis, and clinical management brought about by iuMRI and patient acceptability were assessed. FINDINGS: Participants were recruited between July 29, 2011, and Aug 31, 2014. The cohort was subdivided by gestation into the 18 weeks to less than 24 weeks fetus cohort (n=369) and into the 24 weeks or older fetus cohort (n=201). Diagnostic accuracy was improved by 23% (95% CI 18-27) in the 18 weeks to less than 24 weeks group and 29% (23-36) in the 24 weeks and older group (p