Brain age predicted using graph convolutional neural network explains neurodevelopmental trajectory in preterm neonates.

OBJECTIVES: Dramatic brain morphological changes occur throughout the third trimester of gestation. In this study, we investigated whether the predicted brain age (PBA) derived from graph convolutional network (GCN) that accounts for cortical morphometrics in third trimester is associated with postnatal abnormalities and neurodevelopmental outcome. METHODS: In total, 577 T1 MRI scans of preterm neonates from two different datasets were analyzed; the NEOCIVET pipeline generated cortical surfaces and morphological features, which were then fed to the GCN to predict brain age. The brain age index (BAI; PBA minus chronological age) was used to determine the relationships among preterm birth (i.e., birthweight and birth age), perinatal brain injuries, postnatal events/clinical conditions, BAI at postnatal scan, and neurodevelopmental scores at 30 months. RESULTS: Brain morphology and GCN-based age prediction of preterm neonates without brain lesions (mean absolute error [MAE]: 0.96 weeks) outperformed conventional machine learning methods using no topological information. Structural equation models (SEM) showed that BAI mediated the influence of preterm birth and postnatal clinical factors, but not perinatal brain injuries, on neurodevelopmental outcome at 30 months of age. CONCLUSIONS: Brain morphology may be clinically meaningful in measuring brain age, as it relates to postnatal factors, and predicting neurodevelopmental outcome. CLINICAL RELEVANCE STATEMENT: Understanding the neurodevelopmental trajectory of preterm neonates through the prediction of brain age using a graph convolutional neural network may allow for earlier detection of potential developmental abnormalities and improved interventions, consequently enhancing the prognosis and quality of life in this vulnerable population. KEY POINTS: •Brain age in preterm neonates predicted using a graph convolutional network with brain morphological changes mediates the pre-scan risk factors and post-scan neurodevelopmental outcomes. •Predicted brain age oriented from conventional deep learning approaches, which indicates the neurodevelopmental status in neonates, shows a lack of sensitivity to perinatal risk factors and predicting neurodevelopmental outcomes. •The new brain age index based on brain morphology and graph convolutional network enhances the accuracy and clinical interpretation of predicted brain age for neonates.

Neonatal brain MRI and short-term outcomes after acute provoked seizures.

OBJECTIVE: We investigated how diagnosis and injury location on neonatal brain MRI following onset of acute provoked seizures was associated with short term outcome. STUDY DESIGN: A multicenter cohort of neonates with acute provoked seizures enrolled in the Neonatal Seizure Registry. MRIs were centrally evaluated by a neuroradiologist for location of injury and radiologic diagnosis. Clinical outcomes were determined by chart review. Multivariate logistic regression was used to examine the association between MRI findings and outcomes. RESULTS: Among 236 newborns with MRI at median age 4 days (IQR 3-8), 91% had abnormal MRI. Radiologic diagnoses of intracranial hemorrhage (OR 3.2 [1.6-6.5], p < 0.001) and hypoxic-ischemic encephalopathy (OR 2.7 [1.4-5.4], p < 0.003) were associated with high seizure burden. Radiologic signs of intracranial infection were associated with abnormal neurologic examination at discharge (OR 3.9 [1.3-11.6], p < 0.01). CONCLUSION: Findings on initial MRI can help with expectant counseling on short-term outcomes following acute provoked neonatal seizures.

Cyto/myeloarchitecture of cortical gray matter and superficial white matter in early neurodevelopment: multimodal MRI study in preterm neonates.

The cerebral cortex undergoes rapid microstructural changes throughout the third trimester. Recently, there has been growing interest on imaging features that represent cyto/myeloarchitecture underlying intracortical myelination, cortical gray matter (GM), and its adjacent superficial whitematter (sWM). Using 92 magnetic resonance imaging scans from 78 preterm neonates, the current study used combined T1-weighted/T2-weighted (T1w/T2w) intensity ratio and diffusion tensor imaging (DTI) measurements, including fractional anisotropy (FA) and mean diffusivity (MD), to characterize the developing cyto/myeloarchitectural architecture. DTI metrics showed a linear trajectory: FA decreased in GM but increased in sWM with time; and MD decreased in both GM and sWM. Conversely, T1w/T2w measurements showed a distinctive parabolic trajectory, revealing additional cyto/myeloarchitectural signature inferred. Furthermore, the spatiotemporal courses were regionally heterogeneous: central, ventral, and temporal regions of GM and sWM exhibited faster T1w/T2w changes; anterior sWM areas exhibited faster FA increases; and central and cingulate areas in GM and sWM exhibited faster MD decreases. These results may explain cyto/myeloarchitectural processes, including dendritic arborization, synaptogenesis, glial proliferation, and radial glial cell organization and apoptosis. Finally, T1w/T2w values were significantly associated with 1-year language and cognitive outcome scores, while MD significantly decreased with intraventricular hemorrhage.

A proposed magnetic resonance imaging grading system for the spectrum of central neonatal parasagittal hypoxic-ischaemic brain injury.

AIM: To describe the spectrum of parasagittal injury on MRI studies performed on children following severe perinatal term hypoxia-ischaemia, using a novel MRI grading system, and propose a new central pattern correlated with neuropathologic features. METHODS: MR scans of 297 patients with perinatal term hypoxia-ischaemia were evaluated for typical patterns of brain injury. A total of 83 patients that demonstrated the central/basal ganglia-thalamus and perirolandic pattern of injury were categorised according to the degree of severity. The perirolandic injury was graded by the degree of interhemispheric widening, paracentral lobule involvement and perirolandic cortex destruction leading to a tiered categorisation. Of these 83 patients, 19 had the most severe subtype of injury. A detailed analysis of the clinical data of a subset of 11 of these 19 patients was conducted. RESULTS: We demonstrated the mild subtype in 21/83(25%), the moderate subtype in 22/83(27%) and the severe subtype in 21/83(25%). A fourth pattern was identified in 19/83(23%) patients with a diamond-shaped expansion of the interhemispheric fissure, concomitant thalamic, putaminal, hippocampal and other smaller substrate involvement indicative of the most destructive subtype. CONCLUSIONS: We propose a new MR grading system of injury at the parasagittal perirolandic region related to severe, sustained central perinatal term hypoxia-ischaemia. We also introduce a previously undescribed pattern of injury, the most severe form of this spectrum, seen especially after prolongation of the second stage of labour. This constellation of high metabolic substrate, targeted tissue destruction is consistently demonstrated by MRI, termed the massive paramedian injury pattern.

Robust Cortical Thickness Morphometry of Neonatal Brain and Systematic Evaluation Using Multi-Site MRI Datasets.

The human brain grows the most dramatically during the perinatal and early post-natal periods, during which pre-term birth or perinatal injury that may alter brain structure and lead to developmental anomalies. Thus, characterizing cortical thickness of developing brains remains an important goal. However, this task is often complicated by inaccurate cortical surface extraction due to small-size brains. Here, we propose a novel complex framework for the reconstruction of neonatal WM and pial surfaces, accounting for large partial volumes due to small-size brains. The proposed approach relies only on T1-weighted images unlike previous T2-weighted image-based approaches while only T1-weighted images are sometimes available under the different clinical/research setting. Deep neural networks are first introduced to the neonatal magnetic resonance imaging (MRI) pipeline to address the mis-segmentation of brain tissues. Furthermore, this pipeline enhances cortical boundary delineation using combined models of the cerebrospinal fluid (CSF)/GM boundary detection with edge gradient information and a new skeletonization of sulcal folding where no CSF voxels are seen due to the limited resolution. We also proposed a systematic evaluation using three independent datasets comprising 736 pre-term and 97 term neonates. Qualitative assessment for reconstructed cortical surfaces shows that 86.9% are rated as accurate across the three site datasets. In addition, our landmark-based evaluation shows that the mean displacement of the cortical surfaces from the true boundaries was less than a voxel size (0.532 ± 0.035 mm). Evaluating the proposed pipeline (namely NEOCIVET 2.0) shows the robustness and reproducibility across different sites and different age-groups. The mean cortical thickness measured positively correlated with post-menstrual age (PMA) at scan (p < 0.0001); Cingulate cortical areas grew the most rapidly whereas the inferior temporal cortex grew the least rapidly. The range of the cortical thickness measured was biologically congruent (1.3 mm at 28 weeks of PMA to 1.8 mm at term equivalent). Cortical thickness measured on T1 MRI using NEOCIVET 2.0 was compared with that on T2 using the established dHCP pipeline. It was difficult to conclude that either T1 or T2 imaging is more ideal to construct cortical surfaces. NEOCIVET 2.0 has been open to the public through CBRAIN (https://mcin-cnim.ca/technology/cbrain/), a web-based platform for processing brain imaging data.

MR Imaging of Normal Brain Development.

Normal brain development is best evaluated by MR imaging in the fetal and pediatric patient. As the developing brain grows, myelinates, and sulcates rapidly, understanding the normal appearance of the brain throughout development is critical. The fetal brain can be evaluated by MR imaging after 16 weeks gestational age, both morphologically and biometrically. Sulcation of the fetal brain lags behind premature neonates of equivalent gestational age. Sensory axons generally myelinate before motor axons with central to peripheral and dorsal to ventral myelination gradients. By 2 years of age, the brain has a near adult appearance by conventional anatomic MR imaging.

Maternal or neonatal infection: association with neonatal encephalopathy outcomes.

This corrects the article DOI: 10.1038/pr.2014.47.

Quantitative surface analysis of combined MRI and PET enhances detection of focal cortical dysplasias.

OBJECTIVE: Focal cortical dysplasias (FCDs) often cause pharmacoresistant epilepsy, and surgical resection can lead to seizure-freedom. Magnetic resonance imaging (MRI) and positron emission tomography (PET) play complementary roles in FCD identification/localization; nevertheless, many FCDs are small or subtle, and difficult to find on routine radiological inspection. We aimed to automatically detect subtle or visually-unidentifiable FCDs by building a classifier based on an optimized cortical surface sampling of combined MRI and PET features. METHODS: Cortical surfaces of 28 patients with histopathologically-proven FCDs were extracted. Morphology and intensity-based features characterizing FCD lesions were calculated vertex-wise on each cortical surface, and fed to a 2-step (Support Vector Machine and patch-based) classifier. Classifier performance was assessed compared to manual lesion labels. RESULTS: Our classifier using combined feature selections from MRI and PET outperformed both quantitative MRI and multimodal visual analysis in FCD detection (93% vs 82% vs 68%). No false positives were identified in the controls, whereas 3.4% of the vertices outside FCD lesions were also classified to be lesional ("extralesional clusters"). Patients with type I or IIa FCDs displayed a higher prevalence of extralesional clusters at an intermediate distance to the FCD lesions compared to type IIb FCDs (p < 0.05). The former had a correspondingly lower chance of positive surgical outcome (71% vs 91%). CONCLUSIONS: Machine learning with multimodal feature sampling can improve FCD detection. The spread of extralesional clusters characterize different FCD subtypes, and may represent structurally or functionally abnormal tissue on a microscopic scale, with implications for surgical outcomes.

De novo mutations in KIF1A cause progressive encephalopathy and brain atrophy.

OBJECTIVE: To determine the cause and course of a novel syndrome with progressive encephalopathy and brain atrophy in children. METHODS: Clinical whole-exome sequencing was performed for global developmental delay and intellectual disability; some patients also had spastic paraparesis and evidence of clinical regression. Six patients were identified with de novo missense mutations in the kinesin gene KIF1A. The predicted functional disruption of these mutations was assessed in silico to compare the calculated conformational flexibility and estimated efficiency of ATP binding to kinesin motor domains of wild-type (WT) versus mutant alleles. Additionally, an in vitro microtubule gliding assay was performed to assess the effects of de novo dominant, inherited recessive, and polymorphic variants on KIF1A motor function. RESULTS: All six subjects had severe developmental delay, hypotonia, and varying degrees of hyperreflexia and spastic paraparesis. Microcephaly, cortical visual impairment, optic neuropathy, peripheral neuropathy, ataxia, epilepsy, and movement disorders were also observed. All six patients had a degenerative neurologic course with progressive cerebral and cerebellar atrophy seen on sequential magnetic resonance imaging scans. Computational modeling of mutant protein structures when compared to WT kinesin showed substantial differences in conformational flexibility and ATP-binding efficiency. The de novo KIF1A mutants were nonmotile in the microtubule gliding assay. INTERPRETATION: De novo mutations in KIF1A cause a degenerative neurologic syndrome with brain atrophy. Computational and in vitro assays differentiate the severity of dominant de novo heterozygous versus inherited recessive KIF1A mutations. The profound effect de novo mutations have on axonal transport is likely related to the cause of progressive neurologic impairment in these patients.

Update on neuroimaging phenotypes of mid-hindbrain malformations.

PURPOSE: Neuroimaging techniques including structural magnetic resonance imaging (MRI) and functional positron emission tomography (PET) are useful in categorizing various midbrain-hindbrain (MHB) malformations, both in allowing diagnosis and in helping to understand the developmental processes that were disturbed. Brain imaging phenotypes of numerous malformations are characteristic features that help in guiding the genetic testing in case of direct neuroimaging-genotype correlation or, at least, to differentiate among MHB malformations entities. The present review aims to provide the reader with an update of the use of neuroimaging applications in the fine analysis of MHB malformations, using a comprehensive, recently proposed developmental and genetic classification. METHODS: We have performed an extensive systematic review of the literature, from the embryology main steps of MHB development through the malformations entities, with regard to their molecular and genetic basis, conventional MRI features, and other neuroimaging characteristics. RESULTS: We discuss disorders in which imaging features are distinctive and how these features reflect the structural and functional impairment of the brain. CONCLUSION: Recognition of specific MRI phenotypes, including advanced imaging features, is useful to recognize the MHB malformation entities, to suggest genetic investigations, and, eventually, to monitor the disease outcome after supportive therapies.

Maternal or neonatal infection: association with neonatal encephalopathy outcomes.

BACKGROUND: Perinatal infection may potentiate brain injury among children born preterm. The objective of this study was to examine whether maternal and/or neonatal infection are associated with adverse outcomes among term neonates with encephalopathy. METHODS: This study is a cohort study of 258 term newborns with encephalopathy whose clinical records were examined for signs of maternal infection (chorioamnionitis) and infant infection (sepsis). Multivariate regression was used to assess associations between infection, pattern, and severity of injury on neonatal magnetic resonance imaging, as well as neurodevelopment at 30 mo (neuromotor examination, or Bayley Scales of Infant Development, second edition mental development index <70 or Bayley Scales of Infant Development, third edition cognitive score <85). RESULTS: Chorioamnionitis was associated with lower risk of moderate-severe brain injury (adjusted odds ratio: 0.3; 95% confidence interval: 0.1-0.7; P = 0.004) and adverse cognitive outcome in children when compared with no chorioamnionitis. Children with signs of neonatal sepsis were more likely to exhibit watershed predominant injury than those without (P = 0.007). CONCLUSION: Among neonates with encephalopathy, chorioamnionitis was associated with a lower risk of brain injury and adverse outcomes, whereas signs of neonatal sepsis carried an elevated risk. The etiology of encephalopathy and timing of infection and its associated inflammatory response may influence whether infection potentiates or mitigates injury in term newborns.

Fast 3D (1)H MRSI of the corticospinal tract in pediatric brain.

PURPOSE: To develop a (1)H magnetic resonance spectroscopic imaging (MRSI) sequence that can be used to image infants/children at 3T and by combining it with diffusion tensor imaging (DTI) tractography, extract relevant metabolic information corresponding to the corticospinal tract (CST). MATERIALS AND METHODS: A fast 3D MRSI sequence was developed for pediatric neuroimaging at 3T using spiral k-space readout and dual band RF pulses (32 x 32 x 8 cm field of view [FOV], 1 cc iso-resolution, TR/TE = 1500/130, 6:24 minute scan). Using DTI tractography to identify the motor tracts, spectra were extracted from the CSTs and quantified. Initial data from infants/children with suspected motor delay (n = 5) and age-matched controls (n = 3) were collected and N-acetylaspartate (NAA) ratios were quantified. RESULTS: The average signal-to-noise ratio of the NAA peak from the studies was approximately 22. Metabolite profiles were successfully acquired from the CST by using DTI tractography. Decreased NAA ratios in those with motor delay compared to controls of approximately 10% at the CST were observed. CONCLUSION: A fast and robust 3D MRSI technique targeted for pediatric neuroimaging has been developed. By combining with DTI tractography, metabolic information from the CSTs can be retrieved and estimated. By combining DTI and 3D MRSI, spectral information from various tracts can be obtained and processed.

Malformations of cortical development.

Malformations of cortical development are an important cause of developmental delay and epilepsy. Proper identification of these malformations can greatly help in accurately counseling affected families and, in some cases, in the treatment of the epilepsy. Modem neuroimaging is an important tool in the diagnosis of these malformations.