Introduction: Cerebellar injury is increasingly recognized as a relevant complication of premature birth. However, the prevalence of reduced cerebellar growth and its consequences for neurodevelopmental outcome in preterm infants without overt brain injury remain to be defined in detail. The aim of this study was to assess the transcerebellar diameter (TCD) at term-equivalent age (TEA) in very preterm infants without brain injury and to evaluate whether TCD is related to neurodevelopmental outcome in this population. Methods: Very preterm infants underwent magnetic resonance imaging at TEA. Infants with any grade of supra- or infratentorial brain injury were excluded. TCD was measured and categorized using existing cut-off values as normal TCD and mild or severe TCD reduction. Psychomotor Developmental index (PDI) and Mental Developmental index (MDI) were assessed using Bayley Scales of Infant Development II and III at a corrected age of 2 years. Results: A total of 166 infants with a mean gestational age of 29.9 ± 1.8 weeks and a mean birth weight of 1,317 ± 393 g were included. Mean TCD of girls was significantly lower compared to the mean TCD of boys (p = 0.004). TCD reduction was present in 8 infants (4.8%). Infants with a mild TCD reduction achieved lower mean MDI than infants with normal TCD (p = 0.021). Discussion: We found that reduced TCD was associated with a 17% lower mean MDI at a corrected age of 2 years. Thus, TCD at TEA may be used as an imaging marker for adverse cognitive outcome in the apparently low-risk group of preterm infants without brain injury.
Introduction: Very preterm infants are at risk for adverse neurodevelopmental outcome. To better identify children without brain injury at risk for developmental sequelae, we assessed predictive values of supratentorial brain metrics in relation to outcome. Methods: Very preterm infants underwent magnetic resonance imaging (MRI) at term-equivalent age. Infants with any grade of supra- or infratentorial brain injury according to Kidokoro et al. [Pediatrics 2014;134:e444–53] were excluded. Supratentorial brain metrics (biparietal width, extracerebral space, interhemispheric distance) were measured and categorised using existing cut-off values. The Psychomotor Developmental Index (PDI) and Mental Developmental Index (MDI) were assessed using the Bayley Scales of Infant Development, second and third edition, at 2 years of age. Developmental delay was defined as a score <85. Positive and negative predictive values for developmental delay were calculated. Results: A total of 237 very preterm infants were enrolled. Of all infants, 59 (21.2%) showed developmental delay. Infants with z-scores less than –0.5 for biparietal width had significantly lower PDI (p = 0.039) and MDI (p = 0.042) than infants with normal z-scores. Enlargement of extracerebral spaces was also related to lower PDI (p = 0.047) and MDI (p = 0.036). Negative predictive value was highest when all brain metrics were within the normal range (PDI <85: 96.6%, MDI <85: 90.0%). Combining the biparietal width and the interhemispheric distance showed highest positive predictive values for developmental delay (MDI or PDI <85: 58.3%). Discussion: Supratentorial brain metrics are predictive for neurodevelopmental outcome in infants with ostensibly normal MRI. A combination of supratentorial brain metrics is most meaningful for identifying infants at risk for long-term sequelae.
Introduction: Perinatal asphyxia is a leading cause of neonatal death. Up to one-third of asphyxiated neonates suffer from hypoxic-ischaemic encephalopathy (HIE) with substantial long-term morbidity. Currently available diagnostic and prognostic tools bear limitations, and additional reliable biomarkers are needed for all stages of clinical management. A novel tool in neuroscientific research is micro-ribonucleic acid (miRNA) profiling. The aim of the present study was to determine miRNA expression profiles of healthy and asphyxiated neonates with and without HIE and to assess their potential as diagnostic and prognostic biomarkers. Methods: We prospectively enrolled 49 neonates with a gestational age of ≥36 weeks, 15 of which fulfilled the diagnostic criteria of perinatal asphyxia and 34 served as healthy controls. Dried blood spots were collected from umbilical cord blood (UCB) and from venous blood upon admission to neonatal intensive care unit (NICU) and at 48 h of life. Samples were analysed by means of FirePlex™ technology (Abcam, Cambridge, MA, USA). Results: In the UCB, miRNA expression levels of hsa-mir-124-3p, hsa-mir-1285-5p, and hsa-mir-331-5p were significantly lower in asphyxiated neonates compared to healthy controls. Asphyxiated neonates requiring therapeutic hypothermia had significantly increased expression of hsa-miR-30e-5p and significantly decreased expression of hsa-miR-142-3p, hsa-miR-338-3p, hsa-miR-34b-3p, hsa-miR-497-5p, and hsa-miR-98-5p at the time of admission to the NICU. At 48 h, infants suffering from moderate/severe HIE with a poor long-term neurodevelopmental outcome showed a significant increase in hsa-mir-145-5p. Discussion/Conclusion: MiRNA profiling shows promise as a biomarker for perinatal asphyxia, hypothermia-requiring HIE, and poor neurodevelopmental outcome. Confirmatory studies are called for.
24 August 2022 in Artificial Intelligence in Medicine
Segmentation of specific brain tissue from MRI volumes is of great significance for brain disease diagnosis, progression assessment, and monitoring of neurological conditions. Manual segmentation is time-consuming, laborious, and subjective, which significantly amplifies the need for automated processes. Over the last decades, the active development in the field of deep learning, especially convolutional neural networks (CNNs), and the associated performance improvements have increased the demand for the application of CNN-based methods to provide consistent measurements and quantitative analyses. In this paper, we present an efficient deep learning approach for the segmentation of brain tissue. More specifically, we address the problem of segmentation of the posterior limb of the internal capsule (PLIC) in preterm neonates. To this end, we propose a CNN-based pipeline comprised of slice-selection modules and a multi-view segmentation model, which exploits the 3D information contained in the MRI volumes to improve segmentation performance. One special feature of the proposed method is its ability to identify one desired slice out of the whole image volume, which is relevant for pediatricians in terms of prognosis. To increase computational efficiency, we apply a strategy that automatically reduces the information contained in the MRI volumes to its relevant parts. Finally, we conduct an expert rating alongside standard evaluation metrics, such as dice score, to evaluate the performance of the proposed framework. We demonstrate the benefit of the multi-view technique by comparing it with its single-view counterparts, which reveals that the proposed method strikes a good balance between exploiting the available image information and reducing the required computing power compared to 3D segmentation networks. Standard evaluation metrics as, well as expert-based assessment, confirm the good performance of the proposed framework, with the latter being more relevant in terms of clinical applicability. We demonstrate that the proposed deep learning pipeline can compete with the experts in terms of accuracy. To prove the generalisability of the proposed method, we additionally assess our deep learning pipeline to data from the Developing Human Connectome Project (dHCP).
Introduction: After the release of the Bayley Scales of Infant and Toddler Development, third edition (Bayley-III), US norms, an overestimation of outcome was observed. But, the conformity between the Bayley Scales of Infant Development, second edition (BSID-II), and the Bayley-III German norms is unknown. This retrospective analysis aimed to compare outcomes of very preterm infants tested with BSID-II and Bayley-III German norms. Methods: Infants born from November 2007 to July 2018 were included. Exclusion criteria were death or missing outcome. Infants underwent testing with either BSID-II until December 2013 or Bayley-III from January 2014 onward, at 12 and/or 24 months. BSID-II Mental Developmental Index (MDI) was compared to Bayley-III cognitive score and a combined Bayley-Score (CB-III) consisting of the cognitive and language composite score. BSID-II Psychomotor Developmental Index (PDI) was compared to Bayley-III motor composite score. Abnormal outcomes were defined as scores <85 (delay) or <70 (impairment). Results: 649 infants were included. At 12 months, the Bayley-III cohort achieved higher scores in all domains compared to the BSID-II cohort (all p<0.05), with lower rates of motor delay in the Bayley-III cohort (p<0.001). At 24 months, only Bayley-III motor composite scores were higher than the BSID-II PDI (p<0.001). Rates of cognitive impairment were higher in the Bayley-III cohort (p = 0.013). Interpretation: Our findings indicate that the Bayley-III German norms effectively identify children needing interventions, particularly at 24 months corrected age. This supports both clinical application and scientific comparability with the BSID-II.