T1-weighted MRI Datasets Research Articles

Strength of spatial correlation between gray matter connectivity and patterns of proto-oncogene and neural network construction gene expression is associated with diffuse glioma survival.

Like other forms of neuropathology, gliomas appear to spread along neural pathways. Accordingly, our group and others have previously shown that brain network connectivity is highly predictive of glioma survival. In this study, we aimed to examine the molecular mechanisms of this relationship via imaging transcriptomics. We retrospectively obtained presurgical, T1-weighted MRI datasets from 669 adult patients, newly diagnosed with diffuse glioma. We measured brain connectivity using gray matter networks and coregistered these data with a transcriptomic brain atlas to determine the spatial co-localization between brain connectivity and expression patterns for 14 proto-oncogenes and 3 neural network construction genes. We found that all 17 genes were significantly co-localized with brain connectivity (p < 0.03, corrected). The strength of co-localization was highly predictive of overall survival in a cross-validated Cox Proportional Hazards model (mean area under the curve, AUC = 0.68 +/- 0.01) and significantly (p < 0.001) more so for a random forest survival model (mean AUC = 0.97 +/- 0.06). Bayesian network analysis demonstrated direct and indirect causal relationships among gene-brain co-localizations and survival. Gene ontology analysis showed that metabolic processes were overexpressed when spatial co-localization between brain connectivity and gene transcription was highest (p < 0.001). Drug-gene interaction analysis identified 84 potential candidate therapies based on our findings. Our findings provide novel insights regarding how gene-brain connectivity interactions may affect glioma survival.

Normative body mass-adjusted reference ranges of magnetic resonance imaging signs commonly used in diagnosing idiopathic intracranial hypertension in a healthy standard population

Patients with chronic daily headaches (CDH) are often a diagnostic challenge and frequently undergo neuroimaging. One common underlying cause of CDH is idiopathic intracranial hypertension (IIH). However, certain neuroimaging abnormalities that suggest IIH, such as optic nerve sheath diameters (ONSD), pituitary gland height, and venous sinus diameter, require interpretation due to the absence of established normative values. Notably, intracranial pressure is known to varies with age, sex and weight, further complicating the determination of objectively abnormal findings within a specific patient group. This study aims to assist clinical neuroradiologists in differentiating neuroimaging results in CDH by providing weight-adjusted normative values for imaging characteristics of IIH. In addition to age and BMI we here assessed 1924 population-based T1-weighted MRI datasets of healthy participants for relevant MRI aspects of IIH. Association to BMI was analyzed using linear/logistic regression controlled for age and stratified for sex. ONSD was 4.3 mm [2.8; 5.9]/4.6 mm [3.6; 5.7] and diameter of transverse sinus was 4.67 mm [1.6; 6.5]/4.45 mm [3.0; 7.9]. Height of pituitary gland was 5.1 mm [2.2;8.1]/4.6 mm [1.9;7.1] for female and male respectively. Values generally varied with BMI with regression slopes spanning 0.0001 to 0.05 and were therefor presented as normative values stratified by BMI. Protrusion of ocular papilla, empty sella and transverse sinus occlusion were rare in total. Our data show an association between BMI and commonly used MRI features for diagnosing IIH. We provide categorized normative BMI values for ONSD, pituitary gland height, and transverse sinus diameter. This distinction helps objectively identify potential IIH indicators compared to regular population norms, enhancing diagnostic accuracy for suspected IIH patients. Notably, optic nerve head protrusion, empty sella, and transverse sinus occlusion are rare in healthy individuals, solidifying their importance as imaging markers regardless of BMI.

The electrophysiology and structural difference between humans with distinct risky preference: a study based on EEG and MRI.

Risky decision-making is affected by past feedback, especially after encountering the beneficial loss in the past decision-making round, yet little is known about the mechanism accounting for the distinctive decision-making that different individuals may make under the past loss context. We extracted decision functional medial frontal negative (MFN) and the cortical thickness (CT) from multi-modality electroencephalography (EEG) and T1-weighted structural MRI (sMRI) datasets to assess the individual risky decision under the past loss context. First, concerning the MFN, the low-risk group (LRG) exhibits larger MFN amplitude and longer reaction time than the high-risk group (HRG) when making risky decisions under the loss context. Subsequently, the sMRI analysis reveals a greater CT in the left anterior insula (AI) for HRG compared with LRG, and a greater CT in AI is associated with a high level of impulsivity, driving individuals to make risky choices under the past loss context. Furthermore, for all participants, the corresponding risky decision behavior could be exactly predicted as a correlation coefficient of 0.523 was acquired, and the classification by combing the MFN amplitude and the CT of the left AI also achieves an accuracy of 90.48% to differentiate the two groups. This study may offer new insight into understanding the mechanism that accounts for the inter-individual variability of risky decisions under the loss context and denotes new indices for the prediction of the risky participants.

Explainable classification of Parkinson’s disease using deep learning trained on a large multi-center database of T1-weighted MRI datasets

IntroductionParkinson’s disease (PD) is a severe neurodegenerative disease that affects millions of people. Early diagnosis is important to facilitate prompt interventions to slow down disease progression. However, accurate PD diagnosis can be challenging, especially in the early disease stages. The aim of this work was to develop and evaluate a robust explainable deep learning model for PD classification trained from one of the largest collections of T1-weighted magnetic resonance imaging datasets. Materials and MethodsA total of 2,041 T1-weighted MRI datasets from 13 different studies were collected, including 1,024 datasets from PD patients and 1,017 datasets from age- and sex-matched healthy controls (HC). The datasets were skull stripped, resampled to isotropic resolution, bias field corrected, and non-linearly registered to the MNI PD25 atlas. The Jacobian maps derived from the deformation fields together with basic clinical parameters were used to train a state-of-the-art convolutional neural network (CNN) to classify PD and HC subjects. Saliency maps were generated to display the brain regions contributing the most to the classification task as a means of explainable artificial intelligence. ResultsThe CNN model was trained using an 85%/5%/10% train/validation/test split stratified by diagnosis, sex, and study. The model achieved an accuracy of 79.3%, precision of 80.2%, specificity of 81.3%, sensitivity of 77.7%, and AUC-ROC of 0.87 on the test set while performing similarly on an independent test set. Saliency maps computed for the test set data highlighted frontotemporal regions, the orbital-frontal cortex, and multiple deep gray matter structures as most important. ConclusionThe developed CNN model, trained on a large heterogenous database, was able to differentiate PD patients from HC subjects with high accuracy with clinically feasible classification explanations. Future research should aim to investigate the combination of multiple imaging modalities with deep learning and on validating these results in a prospective trial as a clinical decision support system.

Fairness-related performance and explainability effects in deep learning models for brain image analysis.

Purpose: Explainability and fairness are two key factors for the effective and ethical clinical implementation of deep learning-based machine learning models in healthcare settings. However, there has been limited work on investigating how unfair performance manifests in explainable artificial intelligence (XAI) methods, and how XAI can be used to investigate potential reasons for unfairness. Thus, the aim of this work was to analyze the effects of previously established sociodemographic-related confounders on classifier performance and explainability methods. Approach: A convolutional neural network (CNN) was trained to predict biological sex from T1-weighted brain MRI datasets of 4547 9- to 10-year-old adolescents from the Adolescent Brain Cognitive Development study. Performance disparities of the trained CNN between White and Black subjects were analyzed and saliency maps were generated for each subgroup at the intersection of sex and race. Results: The classification model demonstrated a significant difference in the percentage of correctly classified White male ( ) and Black male ( ) children. Conversely, slightly higher performance was found for Black female ( ) compared with White female ( ) children. Saliency maps showed subgroup-specific differences, corresponding to brain regions previously associated with pubertal development. In line with this finding, average pubertal development scores of subjects used in this study were significantly different between Black and White females ( ) and males ( ). Conclusions: We demonstrate that a CNN with significantly different sex classification performance between Black and White adolescents can identify different important brain regions when comparing subgroup saliency maps. Importance scores vary substantially between subgroups within brain structures associated with pubertal development, a race-associated confounder for predicting sex. We illustrate that unfair models can produce different XAI results between subgroups and that these results may explain potential reasons for biased performance.

Augmented networks for faster brain metastases detection in T1-weighted contrast-enhanced 3D MRI.

Early detection of brain metastases (BM) is one of the determining factors for the successful treatment of patients with cancer; however, the accurate detection of small BM lesions (< 15mm) remains a challenging task. We previously described a framework for the detection of small BM in single-sequence gadolinium-enhanced T1-weighted 3D MRI datasets. It combined classical image processing (IP) with a dedicated convolutional neural network, taking approximately 30s to process each dataset due to computation-intensive IP stages. To overcome the speed limitation, this study aims to reformulate the framework via an augmented pair of CNNs (eliminating the IP) to reduce the processing times while preserving the BM detection performance. Our previous implementation of the BM detection algorithm utilized Laplacian of Gaussians (LoG) for the candidate selection portion of the solution. In this study, we introduce a novel BM candidate detection CNN (cdCNN) to replace this classical IP stage. The network is formulated to have (1) a similar receptive field as the LoG method, and (2) a bias for the detection of BM lesion loci. The proposed CNN is later augmented with a classification CNN to perform the BM detection task. The cdCNN achieved 97.4% BM detection sensitivity when producing 60K candidates per 3D MRI dataset, while the LoG achieved 96.5% detection sensitivity with 73K candidates. The augmented BM detection framework generated on average 9.20 false-positive BM detections per patient for 90% sensitivity, which is comparable with our previous results. However, it processes each 3D data in 1.9s, presenting a 93.5% reduction in the computation time.

Detection of diabetes from whole-body MRI using deep learning.

Obesity is one of the main drivers of type 2 diabetes, but it is not uniformly associated with the disease. The location of fat accumulation is critical for metabolic health. Specific patterns of body fat distribution, such as visceral fat, are closely related to insulin resistance. There might be further, hitherto unknown, features of body fat distribution that could additionally contribute to the disease. We used machine learning with dense convolutional neural networks to detect diabetes-related variables from 2371 T1-weighted whole-body MRI data sets. MRI was performed in participants undergoing metabolic screening with oral glucose tolerance tests. Models were trained for sex, age, BMI, insulin sensitivity, HbA1c, and prediabetes or incident diabetes. The results were compared with those of conventional models. The area under the receiver operating characteristic curve was 87% for the type 2 diabetes discrimination and 68% for prediabetes, both superior to conventional models. Mean absolute regression errors were comparable to those of conventional models. Heatmaps showed that lower visceral abdominal regions were critical in diabetes classification. Subphenotyping revealed a group with high future diabetes and microalbuminuria risk.Our results show that diabetes is detectable from whole-body MRI without additional data. Our technique of heatmap visualization identifies plausible anatomical regions and highlights the leading role of fat accumulation in the lower abdomen in diabetes pathogenesis.

Uncertainty estimation and explainability in deep learning-based age estimation of the human brain: Results from the German National Cohort MRI study.

Brain ageing is a complex neurobiological process associated with morphological changes that can be assessed on MRI scans. Recently, Deep learning (DL)-based approaches have been proposed for the prediction of chronological brain age from MR images yielding high accuracy. These approaches, however, usually do not address quantification of uncertainty and, therefore, intrinsic physiological variability. Considering uncertainty is essential for the interpretation of the difference between predicted and chronological age. In addition, DL-based models lack in explainability compared to classical approaches like voxel-based morphometry. In this study, we aim to address both, modeling uncertainty and providing visual explanations to explore physiological patterns in brain ageing. T1-weighted brain MRI datasets of 10691 participants of the German National Cohort Study, drawn from the general population, were included in this study (chronological age from 20 to 72 years). A regression model based on a 3D Convolutional Neural Network taking into account aleatoric noise was implemented for global as well as regional brain age estimation. We observed high overall accuracy of global brain age estimation with a mean absolute error of 3.2 ± 2.5 years and mean uncertainty of 2.9 ± 0.6 years. Regional brain age estimation revealed higher estimation accuracy and lower uncertainty in central compared to peripheral brain regions. Visual explanations illustrating the importance of brain sub-regions were generated using Grad-CAM: the derived saliency maps showed a high relevance of the lateral and third ventricles, the insular lobe as well as parts of the basal ganglia and the internal capsule.

Utility of Multi-Modal MRI for Differentiating of Parkinson's Disease and Progressive Supranuclear Palsy Using Machine Learning.

Background: Patients with Parkinson's disease (PD) and progressive supranuclear palsy Richardson's syndrome (PSP-RS) often show overlapping clinical features, leading to misdiagnoses. The objective of this study was to investigate the feasibility and utility of using multi-modal MRI datasets for an automatic differentiation of PD patients, PSP-RS patients, and healthy control (HC) subjects.Material and Methods: T1-weighted, T2-weighted, and diffusion-tensor (DTI) MRI datasets from 45 PD patients, 20 PSP-RS patients, and 38 HC subjects were available for this study. Using an atlas-based approach, regional values of brain morphology (T1-weighted), brain iron metabolism (T2-weighted), and microstructural integrity (DTI) were measured and employed for feature selection and subsequent classification using combinations of various established machine learning methods.Results: The optimal machine learning model using regional morphology features only achieved a classification accuracy of 65% (67/103 correct classifications) differentiating PD patients, PSP-RS patients, and HC subjects. The optimal machine learning model using only quantitative T2 values performed slightly better and achieved an accuracy of 75.7% (78/103). The optimal classifier using DTI features alone performed considerably better with 95.1% accuracy (98/103). The optimal multi-modal classifier using all features also achieved an accuracy of 95.1% but required more features and achieved a slightly lower F1-score compared to the optimal model using DTI features alone.Conclusion: Machine learning models using multi-modal MRI perform significantly better than uni-modal machine learning models using morphological parameters based on T1-weighted MRI datasets alone or brain iron metabolism markers based on T2-weighted MRI datasets alone. However, machine learnig models using regional brain microstructural integrity metrics computed from DTI datasets perform similar to the optimal multi-modal machine learning model. Thus, given the results from this study cohort, it appears that morphology and brain iron metabolism markers may not provide additional value for classification compared to using DTI metrics alone.

Unsupervised MRI Homogenization: Application to Pediatric Anterior Visual Pathway Segmentation.

Deep learning strategies have become ubiquitous optimization tools for medical image analysis. With the appropriate amount of data, these approaches outperform classic methodologies in a variety of image processing tasks. However, rare diseases and pediatric imaging often lack extensive data. Specially, MRI are uncommon because they require sedation in young children. Moreover, the lack of standardization in MRI protocols introduces a strong variability between different datasets. In this paper, we present a general deep learning architecture for MRI homogenization that also provides the segmentation map of an anatomical region of interest. Homogenization is achieved using an unsupervised architecture based on variational autoencoder with cycle generative adversarial networks, which learns a common space (i.e. a representation of the optimal imaging protocol) using an unpaired image-to-image translation network. The segmentation is simultaneously generated by a supervised learning strategy. We evaluated our method segmenting the challenging anterior visual pathway using three brain T1-weighted MRI datasets (variable protocols and vendors). Our method significantly outperformed a non-homogenized multi-protocol U-Net.

Does Meningioma Volume Correlate With Clinical Disease Manifestation Irrespective of Histopathologic Tumor Grade?

The aim of the study was to investigate the association between meningioma volume and the occurrence of clinic-radiologic signs of tumor aggressiveness. For volumetric approximation, the authors evaluated the method of semiautomatic image segmentation at hand of high-resolution MRI-image sequences. ITK-SNAP was utilized for semiautomatic image segmentation of 58 gadolinium-contrast enhanced T1-weighted thin-slice MRI datasets for volumetric analysis. Furthermore, multimodal imaging datasets (including T2, FLAIR, T1) were evaluated for radiological biomarkers of aggressiveness and growth potential. Thereby generated data was checked for association with retrospectively collected data points. Location (P = 0.001), clinical disease manifestation (P = 0.033), peritumoral edema (P = 0.038), tumor intrinsic cystic degeneration (P = 0.007), three-dimensional complexity (P = 0.022), and the presence of meningioma mass effect (P = 0.001) were statistically associated with higher tumor volumes. There was no association between higher tumor volumes and histopathological tumor grade. The size of a meningioma does not seem to reliably predict tumor grade. Growth potential seems to be influenced by tumor location. Higher tumor volumes were significantly associated with the occurrence of clinical symptoms.

A statistical atlas of cerebral arteries generated using multi-center MRA datasets from healthy subjects

Magnetic resonance angiography (MRA) can capture the variation of cerebral arteries with high spatial resolution. These measurements include valuable information about the morphology, geometry, and density of brain arteries, which may be useful to identify risk factors for cerebrovascular and neurological diseases at an early time point. However, this requires knowledge about the distribution and morphology of vessels in healthy subjects. The statistical arterial brain atlas described in this work is a free and public neuroimaging resource that can be used to identify vascular morphological changes. The atlas was generated based on 544 freely available multi-center MRA and T1-weighted MRI datasets. The arteries were automatically segmented in each MRA dataset and used for vessel radius quantification. The binary segmentation and vessel size information were non-linearly registered to the MNI brain atlas using the T1-weighted MRI datasets to construct atlases of artery occurrence probability, mean artery radius, and artery radius standard deviation. This public neuroimaging resource improves the understanding of the distribution and size of arteries in the healthy human brain.

Hypothalamic atrophy is related to body mass index and age at onset in amyotrophic lateral sclerosis

ObjectiveOur objective was to study the hypothalamic volume in a cohort of patients with amyotrophic lateral sclerosis (ALS) including symptomatic and presymptomatic ALS mutation carriers.MethodsHigh-resolution three-dimensional T1-weighted MRI datasets from...

Growth Dynamics of Intracranial Tumors in Patients with Neurofibromatosis Type 2

Patients with neurofibromatosis type 2 (NF2) are prone to develop multiple intracranial neoplasms, such as schwannomas of the cranial nerves and meningiomas. The aim of our study was to investigate 1) the median growth rate per year, 2) the growth-free interval, and 3) the growth patterns of NF2-associated intracranial tumors. All available magnetic resonance (MR) images of patients from the regional neurofibromatosis center were collected. The depicted tumors' volumes were then calculated if the tumors met the following inclusion criteria: contrast enhanced T1-weighted MRI datasets had to be available from at least two investigations and tumors had to be measurable on at least two slices. One-hundred and eighty-eight tumors from 52 patients (20 male, 32 female) met the inclusion criteria for volumetric analysis. Overall, the median follow-up time was 76.5 months per patient (range 13-199 months). After 5 years, the median tumor size was 196% ± 338% for vestibular schwannomas (VS), 204% ± 702% meningiomas (M), 128% ± 64.9% for non-vestibular schwannomas (NVS) and 139% ± 270% for pre-operated tumors of the cerebellopontine angle (TX), respectively. The median time to 20% tumor progression was 21 months for VS, NVS and TX, and 17 months for M. Overall, saltatory growth was the most common growth pattern (46.9%). Most NF2-associated tumors display a saltatory growth pattern. Meningiomas and untreated vestibular schwannomas grow rapidly and overall at similar rates. Tumors of the CPA that have been operated on and non-vestibular schwannomas show less relative growth per year.

Gender-specific growth dynamics of neurofibromatosis type-2-related tumors of the central nervous system.

To date, few studies have been published about the growth dynamics of tumors associated with neurofibromatosis type-2 (NF2), none of which evaluated gender-specific differences. Our aim was to compare radiographic data of female and male patients with NF2. MR images of 40 patients (20 female, 20 male) from the regional NF2 referral center were included in this analysis. Tumor sizes were determined by semi-automated volumetric measurement. Intracranial tumors were measured on post-contrast T1-weighted MRI datasets and volumes of intramedullary spinal tumors were determined from sagittal T2-weighted MRI datasets. The median follow-up time was 91months (range, 16-199months) per patient. Intracranial tumors: On average, female patients had 13.4 neoplasms, while male patients had 6.75 (p = 0.042). The overall median time to tumor progression of ≥20% was 20months for females and 18months for males. Tumors of the cerebellopontine angle (CPA) that had undergone previous surgery had shorter progression-free intervals in females than in males (16 and 24months, respectively; p = 0.012). The median 1-year growth rate was 17.5 ± 44.6% in females compared to 12.5 ± 44.9% in males (p = 0.625). Intramedullary spinal tumors: On average, females had 2.05 tumors and males had 1.75 tumors (p = 0.721). Median time to tumor progression was 21months in females and 44months in males (p = 0.204). After 2years, the median growth rate was 24.4 ± 56.8% in female and 13.5 ± 40.4% in male patients (p = 0.813). The radiographic data in this study suggest that female patients are affected by a greater number of tumors than male patients and that post-surgery tumors of the CPA grow faster in females than in males.

Selective increase in posterior corpus callosum thickness between the age of 4 and 11 years

Establishing an efficient functional and structural connectivity between the two cerebral hemispheres is an important developmental task during childhood, and alterations in this development have accordingly been linked to a series of neurodevelopmental and pediatric disorders. The corpus callosum, the major white-matter structure connecting the hemispheres, has been shown to increase in size throughout the three first decades of life. However, behavioral studies indicate that adult-like performance levels of functional hemispheric interaction are already reached during middle and late childhood. Thus, here we specifically examine the structural development of the corpus callosum during the functionally relevant time period by for the first time (a) selectively addressing prospective childhood development and (b) analyzing a sample in which also younger children are well represented. Corpus callosum anatomy was assessed from 732 T1-weighted MRI datasets acquired from 428 children (213 boys, 215 girls) aged of 4.1 and 10.9years, of which 304 were scanned at two time points. Regional callosal thickness was determined from an outline-based segmentation of the mid-sagittal cross-sectional surface area. Linear-mixed model analyses revealed a significant increase in thickness with age (effect size: up to 15% explained variance) equivalent to a growth in callosal thickness of up to 0.19mm per year in the posterior corpus callosum. The age effect was found to be stronger in posterior segments (i.e., splenium) than in other callosal subregions. Also, the age effect was found to be comparable between boys and girls, and was detected irrespective of whether developmental or individual differences in overall brain size where accounted for or not. Our results demonstrate a selective increase in posterior corpus-callosum thickness during middle and late childhood. Since axons crossing the midline in the splenium mainly connect occipital and parietal cortices, the accentuated posterior growth might reflect the onset of a posterior-to-anterior moving maturation wave in cortical development known to take place in the same time period.

Evaluating vestibular schwannoma size and volume on magnetic resonance imaging: An inter- and intra-rater agreement study

ObjectivesIn the management of patients with vestibular schwannoma it is essential to reliably assess tumor size. In respect to volumetric and linear measurements of these tumors we evaluated a) the inter-rater reliability, b) the intra-rater variability, c) the concordance of volume measurements derived from axial versus those from coronal MRI datasets, and d) the correlation of one-dimensional and volumetric measurements. Patients and methodsWe selected gadolinium-enhanced T1-weighted MRI datasets from 20 patients who had both axial and coronal datasets available with the same slice thickness in each of these orientations. Tumor volumes and diameters were independently determined by two investigators. ResultsThe inter-rater reliability was determined based on the intra-class correlation coefficient, which was 0.998 for volumetric measurements and 0.950 for diameters. The relative smallest detectable difference between both raters was 21.2% for volumetric and 21.2% for linear measurements. Regarding the intra-rater variability we found a relative smallest detectable difference of 17.5% (rater 1) and 24.3% (rater 2) for volumetric measurements. The correlation between measurements on axial and those on coronal datasets was ρ=0.999. In order to find a function that reliably predicts tumor volume from diameter, we fitted a series of equations based on linear and polynomial regression, with the highest regression coefficient being r2=0.79. ConclusionThe longitudinal use of semi-automated volumetric measurements has the potential to accurately inform vestibular schwannoma disease management. We have quantified the reliability of this technique. A strict MRI protocol for follow-up investigations should be adhered to in order to minimize measuring errors.

Lenticulostriate arterial distribution pathology may underlie pediatric anoxic brain injury in drowning.

Drowning is a leading cause of neurological morbidity and mortality in young children. Anoxic brain injury (ABI) can result from nonfatal drowning and typically entails substantial neurological impairment. The neuropathology of drowning-induced pediatric ABI is not well established. Specifically, quantitative characterization of the spatial extent and tissue distribution of anoxic damage in pediatric nonfatal drowning has not previously been reported but could clarify the underlying pathophysiological processes and inform clinical management. To this end, we used voxel-based morphometric (VBM) analyses to quantify the extent and spatial distribution of consistent, between-subject alterations in gray and white matter volume. Whole-brain, high-resolution T1-weighted MRI datasets were acquired in 11 children with chronic ABI and 11 age- and gender-matched neurotypical controls (4–12 years). Group-wise VBM analyses demonstrated predominantly central subcortical pathology in the ABI group in both gray matter (bilateral basal ganglia nuclei) and white matter (bilateral external and posterior internal capsules) (P < 0.001); minimal damage was found outside of these deep subcortical regions. These highly spatially convergent gray and white matter findings reflect the vascular distribution of perforating lenticulostriate arteries, an end-arterial watershed zone, and suggest that vascular distribution may be a more important determinant of tissue loss than oxygen metabolic rate in pediatric ABI. Further, these results inform future directions for diagnostic and therapeutic modalities.

The relationship between brain morphology and polysomnography in healthy good sleepers.

BackgroundNormal sleep continuity and architecture show remarkable inter-individual variability. Previous studies suggest that brain morphology may explain inter-individual differences in sleep variables.MethodThirty-eight healthy subjects spent two consecutive nights at the sleep laboratory with polysomnographic monitoring. Furthermore, high-resolution T1-weighted MRI datasets were acquired in all participants. EEG sleep recordings were analyzed using standard sleep staging criteria and power spectral analysis. Using the FreeSurfer software for automated segmentation, 174 variables were determined representing the volume and thickness of cortical segments and the volume of subcortical brain areas. Regression analyses were performed to examine the relationship with polysomnographic and spectral EEG power variables.ResultsThe analysis did not provide any support for the a-priori formulated hypotheses of an association between brain morphology and polysomnographic variables. Exploratory analyses revealed that the thickness of the left caudal anterior cingulate cortex was positively associated with EEG beta2 power (24–32 Hz) during REM sleep. The volume of the left postcentral gyrus was positively associated with periodic leg movements during sleep (PLMS).ConclusionsThe function of the anterior cingulate cortex as well as EEG beta power during REM sleep have been related to dreaming and sleep-related memory consolidation, which may explain the observed correlation. Increased volumes of the postcentral gyrus may be the result of increased sensory input associated with PLMS. However, due to the exploratory nature of the corresponding analyses, these results have to be replicated before drawing firm conclusions.

A comparison of three brain atlases for MCI prediction

BackgroundAlthough previous voxel-based studies using features extracted by atlas-based parcellation produced relatively poor performances on the prediction of Alzheimer's disease (AD) in subjects with mild cognitive impairment (MCI), classification performance usually depends on features extracted from the original images by atlas-based parcellation. To establish whether classification performance differs depending on the choice of atlases, support vector machine (SVM)-based classification using different brain atlases was performed. New methodSeventy-seven three-dimensional T1-weighted MRI data sets of subjects with amnestic MCI, including 39 subjects who developed AD (MCI-C) within three years and 38 who did not (MCI-NC), were used for voxel-based morphometry (VBM) analyses and analyzed using SVM-based pattern recognition methods combined with a feature selection method based on the SVM recursive feature elimination (RFE) method. Three brain atlases were used for the feature selections: the Automated Anatomical Labeling (AAL) Atlas, Brodmann's Areas (BA), and the LONI Probabilistic Brain Atlas (LPBA40). ResultsThe VBM analysis showed a significant cluster of gray matter density reduction, located at the left hippocampal region, in MCI-C compared to MCI-NC. The SVM analyses with the SVM-RFE algorithm revealed that the best classification performance was achieved by LPBA40 with 37 selected features, giving an accuracy of 77.9%. The overall performance in LPBA40 was better than that of AAL and BA regardless of the number of selected features. ConclusionsThese results suggest that feature selection is crucial to improve the classification performance in atlas-based analysis and that the choice of atlases is also important.