Brain Magnetic Resonance Research Articles

Amyloid PET as a marker of normal-appearing white matter early damage in multiple sclerosis: correlation with CSF β-amyloid levels and brain volumes.

The disease course of multiple sclerosis (MS) is unpredictable, and reliable prognostic biomarkers are needed. Positron emission tomography (PET) with β-amyloid tracers is a promising tool for evaluating white matter (WM) damage and repair. Our aim was to investigate amyloid uptake in damaged (DWM) and normal-appearing WM (NAWM) of MS patients, and to evaluate possible correlations between cerebrospinal fluid (CSF) β-amyloid1-42 (Aβ) levels, amyloid tracer uptake, and brain volumes. Twelve MS patients were recruited and divided according to their disease activity into active and non-active groups. All participants underwent neurological examination, neuropsychological testing, lumbar puncture, brain magnetic resonance (MRI) imaging, and 18F-florbetapir PET. Aβ levels were determined in CSF samples from all patients. MRI and PET images were co-registered, and mean standardized uptake values (SUV) were calculated for each patient in the NAWM and in the DWM. To calculate brain volumes, brain segmentation was performed using statistical parametric mapping software. Nonparametric statistical analyses for between-group comparisons and regression analyses were conducted. We found a lower SUV in DWM compared to NAWM (p< 0.001) in all patients. Decreased NAWM-SUV was observed in the active compared to non-active group (p< 0.05). Considering only active patients, NAWM volume correlated with NAWM-SUV (p = 0.01). Interestingly, CSF Aβ concentration was a predictor of both NAWM-SUV (r = 0.79; p = 0.01) and NAWM volume (r = 0.81, p = 0.01). The correlation between CSF Aβ levels and NAWM-SUV suggests that the predictive role of β-amyloid may be linked to early myelin damage and may reflect disease activity and clinical progression.

Multifocal Dysembryoplastic Neuroepithelial Tumor: A Report of a Rare Case

Introduction: Dysembryoplastic neuroepithelial tumor (DNET) is a benign tumor commonly observed in young people and mostly in the temporal lobe. This tumor usually presents with seizure and involves glial and neuronal cells. Case Presentation: The current report presents a case of a 25-year-old female, with intractable seizures including complex partials and secondary generalized seizure. The brain magnetic resonance (MR) showed multiple nonspecific hyperintense lesions in the bilateral temporal lobe. The patient underwent surgery, and selective temporal lobectomy was performed. Conclusions: Microscopic examination showed multifocal DNET, and the patient was seizure-free after surgery. Although DNET is very rare, the presentation of this case can help with its diagnosis and treatment.

Multivariate Approach for Alzheimer's Disease Detection Using Stationary Wavelet Entropy and Predator-Prey Particle Swarm Optimization.

The number of patients with Alzheimer's disease is increasing rapidly every year. Scholars often use computer vision and machine learning methods to develop an automatic diagnosis system. In this study, we developed a novel machine learning system that can make diagnoses automatically from brain magnetic resonance images. First, the brain imaging was processed, including skull stripping and spatial normalization. Second, one axial slice was selected from the volumetric image, and stationary wavelet entropy (SWE) was done to extract the texture features. Third, a single-hidden-layer neural network was used as the classifier. Finally, a predator-prey particle swarm optimization was proposed to train the weights and biases of the classifier. Our method used 4-level decomposition and yielded 13 SWE features. The classification yielded an overall accuracy of 92.73±1.03%, a sensitivity of 92.69±1.29%, and a specificity of 92.78±1.51%. The area under the curve is 0.95±0.02. Additionally, this method only cost 0.88 s to identify a subject in online stage, after its volumetric image is preprocessed. In terms of classification performance, our method performs better than 10 state-of-the-art approaches and the performance of human observers. Therefore, this proposed method is effective in the detection of Alzheimer's disease.

DEEP MR IMAGE SUPER-RESOLUTION USING STRUCTURAL PRIORS.

High resolution magnetic resonance (MR) images are desired for accurate diagnostics. In practice, image resolution is restricted by factors like hardware, cost and processing constraints. Recently, deep learning methods have been shown to produce compelling state of the art results for image superresolution. Paying particular attention to desired hi-resolution MR image structure, we propose a new regularized network that exploits image priors, namely a low-rank structure and a sharpness prior to enhance deep MR image superresolution. Our contributions are then incorporating these priors in an analytically tractable fashion in the learning of a convolutional neural network (CNN) that accomplishes the super-resolution task. This is particularly challenging for the low rank prior, since the rank is not a differentiable function of the image matrix (and hence the network parameters), an issue we address by pursuing differentiable approximations of the rank. Sharpness is emphasized by the variance of the Laplacian which we show can be implemented by a fixed feedback layer at the output of the network. Experiments performed on two publicly available MR brain image databases exhibit promising results particularly when training imagery is limited.

Biological Brain Age Prediction Using Cortical Thickness Data: A Large Scale Cohort Study.

Brain age estimation from anatomical features has been attracting more attention in recent years. This interest in brain age estimation is motivated by the importance of biological age prediction in health informatics, with an application to early prediction of neurocognitive disorders. It is well-known that normal brain aging follows a specific pattern, which enables researchers and practitioners to predict the age of a human's brain from its degeneration. In this paper, we model brain age predicted by cortical thickness data gathered from large cohort brain images. We collected 2,911 cognitively normal subjects (age 45–91 years) at a single medical center and acquired their brain magnetic resonance (MR) images. All images were acquired using the same scanner with the same protocol. We propose to first apply Sparse Group Lasso (SGL) for feature selection by utilizing the brain's anatomical grouping. Once the features are selected, a non-parametric non-linear regression using the Gaussian Process Regression (GPR) algorithm is applied to fit the final age prediction model. Experimental results demonstrate that the proposed method achieves the mean absolute error of 4.05 years, which is comparable with or superior to several recent methods. Our method can also be a critical tool for clinicians to differentiate patients with neurodegenerative brain disease by extracting a cortical thinning pattern associated with normal aging.

A New Optimized Thresholding Method Using Ant Colony Algorithm for MR Brain Image Segmentation.

Image segmentation is considered as one of the most fundamental tasks in image processing applications. Segmentation of magnetic resonance (MR) brain images is also an important pre-processing step, since many neural disorders are associated with brain's volume changes. As a result, brain image segmentation can be considered as an essential measure toward automated diagnosis or interpretation of regions of interest, which can help surgical planning, analyzing changes of brain's volume in different tissue types, and identifying neural disorders. In many neural disorders such as Alzheimer and epilepsy, determining the volume of different brain tissues (i.e., white matter, gray matter, and cerebrospinal fluids) has been proven to be effective in quantifying diseases. A traditional way for segmenting brain images involves the use of a medical expert's experience in manually determining the boundary of different regions of interest in brain images. It may seem that manual segmentation of MR brain images by an expert is the first and the best choice. However, this method is proved to be time-consuming and challenging. Hence, numerous MR brain image segmentation methods with different degrees of complexity and accuracy have been introduced recently. Our work proposes an optimized thresholding method for segmentation of MR brain images using biologically inspired ant colony algorithm. In this proposed algorithm, textural features are adopted as heuristic information. Besides, post-processing image enhancement based on homogeneity is also performed to achieve a better performance. The empirical results on axial T1-weighted MR brain images have demonstrated competitive accuracy to traditional meta-heuristic methods, K-means, and expectation maximization.

A Case of Adult-onset Pompe Disease with Cerebral Stroke and Left Ventricular Hypertrophy

BackgroundPompe disease is an autosomal recessive glycogen storage disorder caused by a deficiency of the lysosomal glycogen-hydrolyzing enzyme acid α-glucosidase. The adult-onset form, late-onset Pompe disease, has been characterized by glycogen accumulation, primarily in skeletal and smooth muscles, causing weakness of the proximal limb girdle and respiratory compromises. Case ReportA 59-year-old female was admitted to the hospital with acute cerebral stroke at the age of 57years. Following her admission, conventional conservative stroke management followed by cerebral arterial clipping was performed. However, weakness of lower extremities, predominantly in the right side, and evening headache were persisting. After obtaining a careful past history, she noticed that she had a history of recurrent respiratory tract infection and she did not like any physical exercise in school. She also complained of gait disturbance since 32years of age. She had also been suffering from systemic hypertension since 40years of age. She had mild respiratory and swallowing difficulties. Her brain Magnetic Resonance (MR) revealed multiple infractions and white matter degeneration with irregular basilar arterial walls. A computed tomography (CT) scan of lower extremities showed diffuse fibrosis of the proximal muscles predominantly on the right thigh. Cardiac echocardiogram showed left ventricular hypertrophy. Electron microscopy of blood cells including lymphocytes and platelets and skin fibroblasts showed marked granular inclusions in lysosomes, suggesting glycogen accumulation. Her measured acid α-glucosidase activity was very low, 1.3 pmol hour−1 punch−1, and we found a homozygous splice-site mutation c.546G>T in the GAA gene. ConclusionCerebral stoke as an initial finding for an adult-type Pompe disease is rare. Left ventricular hypertrophy is also rarely reported for adult onset of Pompe disease. This case will explore further ways to diagnose adult-onset Pompe disease.

Structural brain imaging in Alzheimer\u2019s disease and mild cognitive impairment: biomarker analysis and shared morphometry database

Magnetic resonance (MR) imaging is a powerful technique for non-invasive in-vivo imaging of the human brain. We employed a recently validated method for robust cross-sectional and longitudinal segmentation of MR brain images from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort. Specifically, we segmented 5074 MR brain images into 138 anatomical regions and extracted time-point specific structural volumes and volume change during follow-up intervals of 12 or 24 months. We assessed the extracted biomarkers by determining their power to predict diagnostic classification and by comparing atrophy rates to published meta-studies. The approach enables comprehensive analysis of structural changes within the whole brain. The discriminative power of individual biomarkers (volumes/atrophy rates) is on par with results published by other groups. We publish all quality-checked brain masks, structural segmentations, and extracted biomarkers along with this article. We further share the methodology for brain extraction (pincram) and segmentation (MALPEM, MALPEM4D) as open source projects with the community. The identified biomarkers hold great potential for deeper analysis, and the validated methodology can readily be applied to other imaging cohorts.

Automatic MR image segmentation using maximization of mutual information

Magnetic resonance (MR) brain image segmentation is an important task for the early detection of any deformation followed by the quantitative analysis for the prediction and stage defection of brain diseases. But segmentation of the MR brain image suffers from limited accuracy as captured images have non-uniform homogeneity over an organ, presence of noise, uneven and broken boundary etc. Due to the complex structure of the brain and varieties of the captured MR images, only a single feature based MR image segmentation cannot give sufficient accurate result. In the proposed method thresholds for segmenting the MR image are computed by maximizing the mutual information for the two features, compactness and homogeneity. The proposed algorithm is tested against the real T1 MR image to asses the accuracy. Further the output is validated and compared with the ground truth and other recently reported works.

Brain tumor segmentation on Multimodal MRI scans using EMAP Algorithm.

The utilization of digital images is becoming popular in multiple areas such as clinical applications. There are multiple diagnostic and machine vision-based applications, where image processing plays a vital role in analyzing, interpreting, and solving the problem. Digital image processing techniques are used to increase the quality of images for human interpretation and machine perception. Tumor segmentation in brain magnetic resonance (MRI) volumes is considered as a complex task because of tumor shape, location, and texture. Manual segmentation is a time-consuming task that can be inaccurate due to an increasing volume of MR scanning performed. The goal of this research is to propose an automated method that can identify the whole tumor in each slice in volumetric MRI brain images, and find out the sub-tumor (core tumor, enhancing and non-enhancing) regions. The proposed algorithm is fully automated to segment out both high-grade glioma (HGG) and low-grade glioma (LGG), using the information provided by a sequence of MRI volumes. The designed algorithm does not require any training database and estimates the tumor regions independently using image processing techniques based on expectation maximization and K-mean clustering. The method is evaluated on BRATS 2015 dataset of LGG and HGG MR volumes. The average DICE score achieved by using the proposed technique is 0.92 and is comparable to state-of-the-art techniques which rely on computationally expensive algorithms.

An efficient computerized decision support system for the analysis and 3D visualization of brain tumor

The quality of health services provided by medical centers varies widely, and there is often a large gap between the optimal standard of services when judged based on the locality of patients (rural or urban environments). This quality gap can have serious health consequences and major implications for patient’s timely and correct treatment. These deficiencies can manifest, for example, as a lack of quality services, misdiagnosis, medication errors, and unavailability of trained professionals. In medical imaging, MRI analysis assists radiologists and surgeons in developing patient treatment plans. Accurate segmentation of anomalous tissues and its correct 3D visualization plays an important role inappropriate treatment. In this context, we aim to develop an intelligent computer-aided diagnostic system focusing on human brain MRI analysis. We present brain tumor detection, segmentation, and its 3D visualization system, providing quality clinical services, regardless of geographical location, and level of expertise of medical specialists. In this research, brain magnetic resonance (MR) images are segmented using a semi-automatic and adaptive threshold selection method. After segmentation, the tumor is classified into malignant and benign based on a bag of words (BoW) driven robust support vector machine (SVM) classification model. The BoW feature extraction method is further amplified via speeded up robust features (SURF) incorporating its procedure of interest point selection. Finally, 3D visualization of the brain and tumor is achieved using volume marching cube algorithm which is used for rendering medical data. The effectiveness of the proposed system is verified over a dataset collected from 30 patients and achieved 99% accuracy. A subjective comparative analysis is also carried out between the proposed method and two state-of-the-art tools ITK-SNAP and 3D-Doctor. Experimental results indicate that the proposed system performed better than existing systems and assists radiologist determining the size, shape, and location of the tumor in the human brain.

Chaotic firefly algorithm-based fuzzy C-means algorithm for segmentation of brain tissues in magnetic resonance images

Image segmentation with clustering approach is widely used in biomedical application. Accurate brain Magnetic Resonance (MR) image segmentation is a challenging task due to the complex anatomical structure of brain tissues in addition to the existence of intensity inhomogeneity, partial volume effects and noise. In this study, a spatial modified bias corrected FCM algorithm is applied to brain MRI for the purpose of segmentation into White Matter (WM), Gray Matter (GM) and Cerebrospinal fluid (CSF) in MR images. So to overcome the uncertainty caused by the above effects, a modified Fuzzy C-Means (m-FCM) algorithm for MR brain image segmentation is presented in this paper. Also FCM suffers from initialization sensitivity, to overcome this we have used chaos theory based firefly algorithm. This paper presents a novel application of FCM clustering by using Firefly algorithm with a chaotic map to initialize the population of fireflies and tune the absorption coefficient (λ), for increasing the global search mobility. This algorithm is called chaotic firefly integrated Fuzzy C-means (C-FAFCM) algorithm, which embeds chaos map in the Firefly Algorithm. The proposed technique is applied to several simulated and real T1-weighted for normal magnetic resonance brain images, taken from IBSR and BrainWeb database. The algorithm is realized by incorporating the spatial neighborhood information into the standard FCM algorithm and modifying the membership weighting of each cluster by regularizing it by Total Variation (TV) denoising. The experimental results on both simulated and real brain MRI datasets demonstrate that our proposed method (C-FAFCM) has satisfactory outputs in comparison with some other state of the art, based on FCM and non FCM based algorithms.

Clinical study of children with cryofibrinogenemia: a retrospective study from a single center

BackgroundThis study aimed to evaluate the demographic, clinical features, laboratory data, pathology and other survey in pediatric patients with cryofibrinogenemia.MethodsA 12-year retrospective chart review identified eight pediatric patients at Mackay Memorial Hospital, Taipei, Taiwan.ResultsThe female-to-male ratio was 3:1. The mean age at symptom onset and of diagnosis was 10.3 ± 4.6 years and 12.3 ± 4 years, respectively. One child (12.5%) had primary cryofibrinogenemia. The common symptoms were purpura, arthralgia, and muscle weakness (100%). On laboratory examination, cryofibrinogen was positive in all patients. All patients had increased anti-thrombin III while 87.5% and 62.5% had abnormal protein S and protein C, respectively. All eight also complained of neurologic symptoms. One had vertebral artery narrowing, two showed increased T2-weighted signal intensity on the thalamus or white matter, and one had acute hemorrhagic encephalomyelitis on brain magnetic resonance imaging.ConclusionsThis study reports on the presentations of cryofibrinogenemia, which is rare in children. Most cases are associated with autoimmune disease and have severe and complex presentations. Central nervous system involvement is common.

New horizons for multiple sclerosis therapeutics: milestones in the development of ocrelizumab.

Multiple Sclerosis (MS) is an inflammatory and neurodegenerative disease of the central nervous system, and both T and B cells are involved in its pathogenesis. The vast majority of disease-modifying drugs used for MS act on the inflammatory component of the disease and are approved for use in relapsing-remitting (RR) patients. Ocrelizumab (OCR) is the only MS drug that has been approved by the US Food and Drug Administration (FDA) not only for patients with RRMS but also for patients with primary progressive (PP) MS. OCR is a humanized anti-CD20 monoclonal antibody that can deplete the targeted B cells through antibody-dependent cellular cytotoxicity. Treatment involves administration by intravenous infusion every 6 months. OCR can cause long-lasting B-cell depletion and change the pool of reconstituted B cells. Phase III clinical trials have confirmed the results of previous Phase II studies. In particular, OPERA I and II trials, which were performed in patients with RRMS, showed a reduction in the annualized relapse rate, the risk of disability progression, and the number of new/enlarging T2 lesions and enhancing lesions measured using brain magnetic resonance. The ORATORIO trial, performed in PP subjects, showed that OCR can reduce disability progression, improve performance on the timed 25-foot walk, and decrease the total volume of T2 lesions and the mean number of new or enlarging T2 lesions. The most frequent adverse events were the infusion-related reactions and infections. Infections were mostly nasopharyngitis, as well as upper respiratory and urinary tract infections. OCR gives no indication for severe or opportunistic infections. There is not a clear increased risk of malignancies. Nevertheless, it could not be excluded. Real-life registries will provide more information about the long-term safety, the risk of exposure during pregnancy, and the risk of rare adverse events. In this review, we analyze the evidence regarding the efficacy and the safety of OCR.

Fully Automated and Adaptive Intensity Normalization Using Statistical Features for Brain MR Images

Accuracy of the results obtained by automated processing of brain magnetic resonance images has vital importance for diagnosis and evaluation of a progressive disease during treatment. However, automated processing methods such as segmentation, registration and comparison of these images are challenging issues. Because intensity values do not only depend on the underlying tissue type. They can change due to scanner-related artifacts and noise, which usually occurs in magnetic resonance images. In addition to intensity variations, low contrast and partial volume effects increases the difficulty in automated methods with these images. Intensity normalization has a significant role to increase performance of automated image processing methods. Because it is applied as a pre-processing step and efficiency of the other steps in these methods is based on the results obtained from the pre-processing step. The goal of intensity normalization is to make uniform the mean and variance values in images. Different methods have been applied for this purpose in the literature and each method has been tested with different kind of images. In this work; 1) The state-of-art normalization methods applied for magnetic resonance images have been reviewed. 2) A fully automated and adaptive approach has been proposed for intensity normalization in brain magnetic resonance images. 3) Comparative performance evaluations of the results obtained by four different normalization approaches using the same images have been presented. Comparisons of all methods implemented in this work indicate a better performance of the proposed approach for brain magnetic resonance images.

Implementation of a Rapid, Protocol-based TIA Management Pathway.

IntroductionOur goal was to assess whether use of a standardized clinical protocol improves efficiency for patients who present to the emergency department (ED) with symptoms of transient ischemic attack (TIA).MethodsWe performed a structured, retrospective, cohort study at a large, urban, tertiary care academic center. In July 2012 this hospital implemented a standardized protocol for patients with suspected TIA. The protocol selected high-risk patients for admission and low/intermediate-risk patients to an ED observation unit for workup. Recommended workup included brain imaging, vascular imaging, cardiac monitoring, and observation. Patients were included if clinical providers determined the need for workup for TIA. We included consecutive patients presenting during a six-month period prior to protocol implementation, and those presenting between 6–12 months after implementation. Outcomes included ED length of stay (LOS), hospital LOS, use of neuroimaging, and 90-day risk of stroke or TIA.ResultsFrom 01/2012 to 06/2012, 130 patients were evaluated for TIA symptoms in the ED, and from 01/2013 to 06/2013, 150 patients. The final diagnosis was TIA or stroke in 45% before vs. 41% after (p=0.18). Following the intervention, the inpatient admission rate decreased from 62% to 24% (p<0.001), median ED LOS decreased by 1.2 hours (5.7 to 4.9 hours, p=0.027), and median total hospital LOS from 29.4 hours to 23.1 hours (p=0.019). The proportion of patients receiving head computed tomography (CT) went from 68% to 58% (p=0.087); brain magnetic resonance (MR) imaging from 83% to 88%, (p=0.44) neck CT angiography from 32% to 22% (p=0.039); and neck MR angiography from 61% to 72% (p=0.046). Ninety-day stroke or recurrent TIA among those with final diagnosis of TIA was 3% for both periods.ConclusionImplementation of a TIA protocol significantly reduced ED LOS and total hospital LOS.

Classification of tumor based on magnetic resonance (MR) brain images using wavelet energy feature and neuro-fuzzy model

The brain is the organ that coordinates all the activities that occur in our bodies. Small abnormalities in the brain will affect body activity. Tumor of the brain is a mass formed a result of cell growth not normal and unbridled in the brain. MRI is a non-invasive medical test that is useful for doctors in diagnosing and treating medical conditions. The process of classification of brain tumor can provide the right decision and correct treatment and right on the process of treatment of brain tumor. In this study, the classification process performed to determine the type of brain tumor disease, namely Alzheimer’s, Glioma, Carcinoma and normal, using energy coefficient and ANFIS. Process stages in the classification of images of MR brain are the extraction of a feature, reduction of a feature, and process of classification. The result of feature extraction is a vector approximation of each wavelet decomposition level. The feature reduction is a process of reducing the feature by using the energy coefficients of the vector approximation. The feature reduction result for energy coefficient of 100 per feature is 1 x 52 pixels. This vector will be the input on the classification using ANFIS with Fuzzy C-Means and FLVQ clustering process and LM back-propagation. Percentage of success rate of MR brain images recognition using ANFIS-FLVQ, ANFIS, and LM back-propagation was obtained at 100%.

Diffuse lesions in the limbic system with short‐term memory loss in a patient with multiple sclerosis

AbstractBackgroundThe role of B cells in the pathogenesis of multiple sclerosis (MS) is a subject of much research. We report a case of a MS patient who presented with subacute diffuse lesions in the limbic system and short‐term memory loss.Case presentationA male patient had experienced numbness in the left upper extremity at the age of 17 years, but this disappeared without treatment. At 19 years‐of‐age, he developed subacute short‐term memory loss followed by paralysis and sensory disturbance in the left upper and lower extremities. Magnetic resonance brain fluid‐attenuated/T2‐weighted imaging showed diffuse high‐intensity lesions in the medial parts of both temporal lobes, and spotty multiple lesions in the periventricular and subcortical white matter. Twenty days later, a new lesion was observed around the right putamen, and the temporal lesions had disappeared. A symptomatic cervical lesion with gadolinium enhancement was observed at C3. This lesion involved just one vertebral segment. Cerebrospinal fluid analysis showed no pleocytosis or elevated protein levels, but did show oligoclonal immunoglobulin G bands. No antibodies against aquaporin‐4 or myelin oligodendrocyte glycoprotein were detected in the serum, and anti‐N‐methyl‐D‐aspartate receptor was not detected in the cerebrospinal fluid. We diagnosed the patient with MS despite the atypical diffuse lesions in the limbic system and the short‐term memory loss. Corticosteroid medication effectively treated the symptoms in his extremities and his impaired memory.ConclusionsThe present case shows that cortical lesions can occur in MS patients and might be induced by B cell‐mediated mechanisms.

Neurodevelopmental outcomes in newborns with neonatal seizures caused by rotavirus-associated leukoencephalopathy

PurposeRotavirus infection has recently been reported to be associated with seizures accompanied by leukoencephalopathy in newborns. We aimed to determine long-term outcomes and prognostic factors in newborns with neonatal seizures caused by rotavirus-associated leukoencephalopathy. MethodsWe retrospectively reviewed the records and brain magnetic resonance (MR) images of 32 patients who fulfilled the following criteria: (1) neonatal seizures, (2) distinctive symmetric cerebral white matter lesions on diffusion-weighted MR images (DWI), (3) rotavirus infection, (4) absence of a specific etiology of seizures, except for the aforementioned DWI lesions, and (5) Korean Bayley Scales of Infant Development II (K-BSID-II) assessment after 12 months of age. ResultsThe mean age at seizure onset was 4.7 ± 0.8 days. The median age of the patients at the time of K-BSID-II assessment was 22 months. Fourteen patients (43.8%) showed normal or accelerated performance in the mental and motor scales, while 18 patients (56.2%) had delayed performance in the mental and/or motor scales. Seven patients (21.9%) had significantly delayed performances on the mental and/or motor scales. The percentage of volume of diffusion-restricted lesions based on total brain volume was significantly negatively correlated with the mental developmental index (MDI) score (r = −0.507, p = .003), but not with the psychomotor developmental index (PDI) score (r = −0.324, p = .071). ConclusionsRotavirus-associated leukoencephalopathy in newborns around 5 days of age can cause adverse neurodevelopmental outcomes with a wide range of severity. The extent of white matter lesion on initial DWI can predict neurocognitive outcome.

Brain MR Image Restoration Using an Automatic Trilateral Filter With GPU-Based Acceleration.

Noise reduction in brain magnetic resonance (MR) images has been a challenging and demanding task. This study develops a new trilateral filter that aims to achieve robust and efficient image restoration. Extended from the bilateral filter, the proposed algorithm contains one additional intensity similarity funct-ion, which compensates for the unique characteristics of noise in brain MR images. An entropy function adaptive to intensity variations is introduced to regulate the contributions of the weighting components. To hasten the computation, parallel computing based on the graphics processing unit (GPU) strategy is explored with emphasis on memory allocations and thread distributions. To automate the filtration, image texture feature analysis associated with machine learning is investigated. Among the 98 candidate features, the sequential forward floating selection scheme is employed to acquire the optimal texture features for regularization. Subsequently, a two-stage classifier that consists of support vector machines and artificial neural networks is established to predict the filter parameters for automation. A speedup gain of 757 was reached to process an entire MR image volume of 256 × 256 × 256 pixels, which completed within 0.5s. Automatic restoration results revealed high accuracy with an ensemble average relative error of 0.53 ± 0.85% in terms of the peak signal-to-noise ratio. This self-regulating trilateral filter outperformed many state-of-the-art noise reduction methods both qualitatively and quantitatively. We believe that this new image restoration algorithm is of potential in many brain MR image processing applications that require expedition and automation.