Brain Volumes at Term-Equivalent Age in Preterm Infants: Imaging Biomarkers for Neurodevelopmental Outcome through Early School Age.

OBJECTIVE: To evaluate the relationship between brain volumes at term and neurodevelopmental outcome through early school age in preterm infants. STUDY DESIGN: One hundred twelve preterm infants (born mean gestational age 28.6 ± 1.7 weeks) were studied prospectively with magnetic resonance imaging (imaged at mean 41.6 ± 1.0 weeks). T2- and T1-weighted images were automatically segmented, and volumes of 6 tissue types were related to neurodevelopmental outcome assessed using the Bayley Scales of Infant and Toddler Development, Third Edition (cognitive, fine, and gross motor scores) at 24 months corrected age (n = 112), Griffiths Mental Development Scales (developmental quotient) at age 3.5 years (n = 98), Movement Assessment Battery for Children, Second Edition (n = 85), and Wechsler Preschool and Primary Scale of Intelligence, Third Edition at age 5.5 years (n = 44). Corrections were made for intracranial volume, maternal education, and severe brain lesions. RESULTS: Ventricular volumes were negatively related to neurodevelopmental outcome at age 24 months and 3.5 years, as well as processing speed at age 5.5 years. Unmyelinated white matter (UWM) volume was positively associated with motor outcome at 24 months and with processing speed at age 5.5 years. Cortical gray matter (CGM) volume demonstrated a negative association with motor performance and cognition at 24 months and with developmental quotient at age 3.5 years. Cerebellar volume was positively related to cognition at these time points. Adjustment for brain lesions attenuated the relations between cerebellar and CGM volumes and cognition. CONCLUSIONS: Brain volumes of ventricles, UWM, CGM, and cerebellum may serve as biomarkers for neurodevelopmental outcome in preterm infants. The relationship between larger CGM volumes and adverse neurodevelopment may reflect disturbances in neuronal and/or axonal migration at the UWM-CGM boundary and warrants further investigation.

Unmyelinated white matter loss in the preterm brain is associated with early increased levels of end-tidal carbon monoxide.

OBJECTIVE: Increased levels of end-tidal carbon monoxide (ETCOc) in preterm infants during the first day of life are associated with oxidative stress, inflammatory processes and adverse neurodevelopmental outcome at 2 years of age. Therefore, we hypothesized that early ETCOc levels may also be associated with impaired growth of unmyelinated cerebral white matter. METHODS: From a cohort of 156 extremely and very preterm infants in which ETCOc was determined within 24 h after birth, in 36 infants 3D-MRI was performed at term-equivalent age to assess cerebral tissue volumes of important brain regions. RESULTS: Linear regression analysis between cerebral ventricular volume, unmyelinated white matter/total brain volume-, and cortical grey matter/total brain volume-ratio and ETCOc showed a positive, negative and positive correlation, respectively. Multivariable analyses showed that solely ETCOc was positively related to cerebral ventricular volume and cortical grey matter/total brain volume ratio, and that solely ETCOc was inversely related to the unmyelinated white matter/total brain volume ratio, suggesting that increased levels of ETCOc, associated with oxidative stress and inflammation, were related with impaired growth of unmyelinated white matter. CONCLUSION: Increased values of ETCOc, measured within the first 24 hours of life may be indicative of oxidative stress and inflammation in the immediate perinatal period, resulting in impaired growth of the vulnerable unmyelinated white matter of the preterm brain.

Automatic segmentation of eight tissue classes in neonatal brain MRI.

PURPOSE: Volumetric measurements of neonatal brain tissues may be used as a biomarker for later neurodevelopmental outcome. We propose an automatic method for probabilistic brain segmentation in neonatal MRIs. MATERIALS AND METHODS: In an IRB-approved study axial T1- and T2-weighted MR images were acquired at term-equivalent age for a preterm cohort of 108 neonates. A method for automatic probabilistic segmentation of the images into eight cerebral tissue classes was developed: cortical and central grey matter, unmyelinated and myelinated white matter, cerebrospinal fluid in the ventricles and in the extra cerebral space, brainstem and cerebellum. Segmentation is based on supervised pixel classification using intensity values and spatial positions of the image voxels. The method was trained and evaluated using leave-one-out experiments on seven images, for which an expert had set a reference standard manually. Subsequently, the method was applied to the remaining 101 scans, and the resulting segmentations were evaluated visually by three experts. Finally, volumes of the eight segmented tissue classes were determined for each patient. RESULTS: The Dice similarity coefficients of the segmented tissue classes, except myelinated white matter, ranged from 0.75 to 0.92. Myelinated white matter was difficult to segment and the achieved Dice coefficient was 0.47. Visual analysis of the results demonstrated accurate segmentations of the eight tissue classes. The probabilistic segmentation method produced volumes that compared favorably with the reference standard. CONCLUSION: The proposed method provides accurate segmentation of neonatal brain MR images into all given tissue classes, except myelinated white matter. This is the one of the first methods that distinguishes cerebrospinal fluid in the ventricles from cerebrospinal fluid in the extracerebral space. This method might be helpful in predicting neurodevelopmental outcome and useful for evaluating neuroprotective clinical trials in neonates.

Cerebellar volume and proton magnetic resonance spectroscopy at term, and neurodevelopment at 2 years of age in preterm infants.

AIM: To assess the relation between cerebellar volume and spectroscopy at term equivalent age, and neurodevelopment at 24 months corrected age in preterm infants. METHODS: Magnetic resonance imaging of the brain was performed around term equivalent age in 112 preterm infants (mean gestational age 28wks 3d [SD 1wk 5d]; birthweight 1129g [SD 324g]). Cerebellar volume (60 males, 52 females), and proton magnetic resonance spectroscopy ((1) H-MRS) of the cerebellum in a subgroup of 58 infants were assessed in relation to cognitive, fine motor, and gross motor scores on the Bayley Scales of Infant and Toddler Development-III. Different neonatal variables and maternal education were regarded possible confounders. RESULTS: Cerebellar volume was significantly associated with postmenstrual age at time of magnetic resonance imaging. Cerebellar volume corrected for postmenstrual age was significantly and positively associated with cognition. Cognitive scores related significantly with N-acetylaspartate/choline (NAA/Cho) ratio obtained from cerebellar (1) H-MRS in 53 infants. Correction for neonatal and maternal variables did not change these results. Cerebellar variables were not related to motor performance. INTERPRETATION: In preterm infants, both cerebellar volume and cerebellar NAA/Cho ratio at term equivalent age were positively associated with cognition; however, no relation was found with motor outcome at 2 years of age. These findings support the importance of the cerebellum in cognitive development in preterm infants.

Construction of periventricular white matter hyperintensity maps by spatial normalization of the lateral ventricles.

Subcortical and periventricular white matter hyperintensities (WMHs) may have different associations with cognition and pathophysiology. The aim of the present study is to develop an automated method for construction of periventricular WMH maps that enables the analysis of between-group differences in WMH location and characteristics in the periventricular region without the requirement of prior boundary definition. To avoid influence of WMHs on spatial normalization, a reference image of the lateral ventricles was constructed based on images of 24 subjects. Construction was not biased to a single subject. WMHs were segmented by k-nearest neighbor-based classification of magnetic resonance inversion recovery and fluid attenuated inversion recovery images. Cerebrospinal fluid segmentations of individual subjects were nonrigidly mapped to the reference image of the lateral ventricles. The subject's WMHs were transformed to the reference space accordingly. Spatial normalization accuracy was validated using measures of overlap and of displacement relative to the boundary of the lateral ventricles. After spatial normalization, the boundaries of the lateral ventricles closely matched the reference image and in an area of approximately 1 cm around the lateral ventricles the relative displacement was less than 1 mm. To illustrate the method, it was applied to 61 patients with Type 2 diabetes and 26 control subjects, whereupon periventricular WMH maps were constructed and compared. The proposed method is particularly suited to analyze WMH distribution differences at the level of the lateral ventricles between large groups of patients.

Probabilistic brain tissue segmentation in neonatal magnetic resonance imaging.

A fully automated method has been developed for segmentation of four different structures in the neonatal brain: white matter (WM), central gray matter (CEGM), cortical gray matter (COGM), and cerebrospinal fluid (CSF). The segmentation algorithm is based on information from T2-weighted (T2-w) and inversion recovery (IR) scans. The method uses a K nearest neighbor (KNN) classification technique with features derived from spatial information and voxel intensities. Probabilistic segmentations of each tissue type were generated. By applying thresholds on these probability maps, binary segmentations were obtained. These final segmentations were evaluated by comparison with a gold standard. The sensitivity, specificity, and Dice similarity index (SI) were calculated for quantitative validation of the results. High sensitivity and specificity with respect to the gold standard were reached: sensitivity >0.82 and specificity >0.9 for all tissue types. Tissue volumes were calculated from the binary and probabilistic segmentations. The probabilistic segmentation volumes of all tissue types accurately estimated the gold standard volumes. The KNN approach offers valuable ways for neonatal brain segmentation. The probabilistic outcomes provide a useful tool for accurate volume measurements. The described method is based on routine diagnostic magnetic resonance imaging (MRI) and is suitable for large population studies.

Probabilistic segmentation of brain tissue in MR imaging.

A new method has been developed for probabilistic segmentation of five different types of brain structures: white matter, gray matter, cerebro-spinal fluid without ventricles, ventricles and white matter lesion in cranial MR imaging. The algorithm is based on information from T1-weighted (T1-w), inversion recovery (IR), proton density-weighted (PD), T2-weighted (T2-w) and fluid attenuation inversion recovery (FLAIR) scans. It uses the K-Nearest Neighbor classification technique that builds a feature space from spatial information and voxel intensities. The technique generates for each tissue type an image representing the probability per voxel being part of it. By application of thresholds on these probability maps, binary segmentations can be obtained. A similarity index (SI) and a probabilistic SI (PSI) were calculated for quantitative evaluation of the results. The influence of each image type on the performance was investigated by alternately leaving out one of the five scan types. This procedure showed that the incorporation of the T1-w, PD or T2-w did not significantly improve the segmentation results. Further investigation indicated that the combination of IR and FLAIR was optimal for segmentation of the five brain tissue types. Evaluation with respect to the gold standard showed that the SI-values for all tissues exceeded 0.8 and all PSI-values exceeded 0.7, implying an excellent agreement.

Automatic segmentation of different-sized white matter lesions by voxel probability estimation.

A new method for fully automated segmentation of white matter lesions (WMLs) on cranial MR imaging is presented. The algorithm uses five types of regular MRI-scans. It is based on a K-Nearest Neighbor (KNN) classification technique, which builds a feature space from voxel intensity features and spatial information. The technique generates images representing the probability per voxel being part of a WML. By application of thresholds on these probability maps binary segmentations can be produced. ROC-curves show that the segmentations achieve a high sensitivity and specificity. Three similarity measures, the similarity index (SI), the overlap fraction (OF) and the extra fraction (EF), are calculated for evaluation of the results and determination of the optimal threshold on the probability map. Investigation of the relation between the total lesion volume and the similarity measures shows that the method performs well for lesions larger than 2 cc. The maximum SI per patient is correlated to the total WML volume. No significant relation between the lesion volume and the optimal threshold has been observed. The probabilistic equivalents of the SI, OF en EF (PSI, POF and PEF) allow direct evaluation of the probability maps, which provides a strong tool for comparison of different classification results. A significant correlation between the lesion volume and the PSI and the PEF has been noticed. This method for automated WML segmentation is applicable to lesions of different sizes and shapes, and reaches an accuracy that is comparable to existing methods for multiple sclerosis lesion segmentation. Furthermore, it is suitable for detection of WMLs in large and longitudinal population studies.

Probabilistic segmentation of white matter lesions in MR imaging.

A new method has been developed for fully automated segmentation of white matter lesions (WMLs) in cranial MR imaging. The algorithm uses information from T1-weighted (T1-w), inversion recovery (IR), proton density-weighted (PD), T2-weighted (T2-w) and fluid attenuation inversion recovery (FLAIR) scans. It is based on the K-Nearest Neighbor (KNN) classification technique that builds a feature space from voxel intensities and spatial information. The technique generates images representing the probability per voxel being part of a WML. By application of thresholds on these probability maps, binary segmentations can be obtained. ROC curves show that the segmentations achieve both high sensitivity and specificity. A similarity index (SI), overlap fraction (OF) and extra fraction (EF) are calculated for additional quantitative analysis of the result. The SI is also used for determination of the optimal probability threshold for generation of the binary segmentation. Using probabilistic equivalents of the SI, OF and EF, the probability maps can be evaluated directly, providing a powerful tool for comparison of different classification results. This method for automated WML segmentation reaches an accuracy that is comparable to methods for multiple sclerosis (MS) lesion segmentation and is suitable for detection of WMLs in large and longitudinal population studies.