Gray Matter Segmentation Research Articles

Predicting the brain age of children with cerebral palsy using a two-dimensional convolutional neural networks prediction model without gray and white matter segmentation.

Abnormal brain development is common in children with cerebral palsy (CP), but there are no recent reports on the actual brain age of children with CP. Our objective is to use the brain age prediction model to explore the law of brain development in children with CP. A two-dimensional convolutional neural networks brain age prediction model was designed without segmenting the white and gray matter. Training and testing brain age prediction model using magnetic resonance images of healthy people in a public database. The brain age of children with CP aged 5-27 years old was predicted. The training dataset mean absolute error (MAE) = 1.85, r = 0.99; test dataset MAE = 3.98, r = 0.95. The brain age gap estimation (BrainAGE) of the 5- to 27-year-old patients with CP was generally higher than that of healthy peers (p < 0.0001). The BrainAGE of male patients with CP was higher than that of female patients (p < 0.05). The BrainAGE of patients with bilateral spastic CP was higher than those with unilateral spastic CP (p < 0.05). A two-dimensional convolutional neural networks brain age prediction model allows for brain age prediction using routine hospital T1-weighted head MRI without segmenting the white and gray matter of the brain. At the same time, these findings suggest that brain aging occurs in patients with CP after brain damage. Female patients with CP are more likely to return to their original brain development trajectory than male patients after brain injury. In patients with spastic CP, brain aging is more serious in those with bilateral cerebral hemisphere injury than in those with unilateral cerebral hemisphere injury.

Applications of strategically acquired gradient echo imaging to neurodegenerative diseases

Background: One major thrust in radiology today is image standardization with a focus on rapid, quantitative, multi-contrast data collection and processing.&#x0D; Methods: Strategically acquired gradient echo (STAGE) imaging1-4 is one such method that uses multiple flip angles and multiple echo times. It can provide 8 qualitative and 7 quantitative images as well as transmit field B1 transmit field and B1 receive field maps in 4-6 minutes or less on a 3T magnetic resonance (MR) scanner. STAGE provides qualitative images in the form of proton density-weighted images, T1 weighted images and T2* weighted images. STAGE provides quantitative data in the form of proton spin density (PSD), T1, T2* and susceptibility maps as well as segmentation of white matter, gray matter and cerebrospinal fluid via simulated double inversion recovery (sDIR) images. STAGE has been tested using the NIST phantom and yields intrasubject errors of only 1-2% and intrasubject variation of 2 to 5%. Contrast-to-noise ratio (CNR) measurements show that the T1WE images are comparable to the conventional T1W MP-RAGE images. Today these quantitative measures are providing new biomarkers for imaging a variety of neurodegenerative diseases (Figure 1).&#x0D; Results: During the last few years, we have focused on measuring iron content and neuromelanin (NM) in the substantia nigra (SN) for comparing idiopathic Parkinson’s disease (PD) with healthy controls and patients with other movement disorders. We have found that the volume of NM, the iron content of the SN, the volume of the SN and the N1 sign all together can provide an area under the curve of 95% in distinguishing PD from healthy controls.5 We have developed a template of the midbrain to allow for automatic detection and quantification of these properties. We use tSWI to enhance the N1 sign visibility. We have also used STAGE to study multiple sclerosis (MS) lesions. QSM can be used to map changes in white matter susceptibility and potentially correlated with demyelination. We provide a composite image using tSWI combined with fluid-attenuated inversion recovery (FLAIR) to highlight those lesions that are purely inflammatory from inflammatory demyelinating lesions. Recently we have begun to study the use of absolute water content as a measure of lesion atrophy. Higher water content means a higher likelihood of tissue damage. This also explains why the presence of “black holes” seen in T1W images of MS patients tends to correlate with the expanded disability status score.&#x0D; Conclusions: In summary, STAGE provides a comprehensive clinical imaging protocol that, combined with diffusion-weighted imaging (DWI) and FLAIR, can yield a standardized 10-minute (3T) o 15-minute (1.5T) imaging protocol of the entire brain across all manufacturers.

Microstructural changes in the brain in persons with mild COVID-19

BACKGROUND: The purpose of which was to investigate, using voxel-based MR morphometry, changes in the volume of brain structures in persons who had undergone a new coronavirus disease.&#x0D; MATERIALS AND METHODS: 34 patients with mild COVID-19 between 4 months and one year were examined. All study participants were scanned at two time points: before COVID-19 infection and re-imaging after infection. The comparison group consisted of 30 healthy volunteers who did not have COVID-19 and were comparable in terms of sex, age, timing and MRI protocol.&#x0D; RESULTS: In the main and control groups a significant increase in the volume of individual segments of the large hemisphere cortex, subcortical gray matter, and white matter of the brain was detected. Signs of atrophic process were registered in the cerebellar cortex and left frontal lobe. In the main group there was an increase of cortical volume practically in all cerebral lobes with some emphasis on the posterior parts of the right hemisphere. When analyzing changes of subcortical gray matter in the main group, there was reliable (p 0.05) increase of its total volume due to left caudate nucleus, pale ball, right amygdala body and ventral diencephalon. Increased volume of white matter of the brain in COVID-19 patients was detected in the left isthmus and rostral part of the right cingulate gyrus, left paracentral lobe, and right precuneus.&#x0D; CONCLUSION: the cerebellum is affected quite early by SARS-CoV-2, as we recorded atrophic changes in its cortex at 412 months. In COVID-19, the microstructural integrity, predominantly of gray matter and other brain segments, is impaired. Increased volume of these structures indicates the duration of the inflammatory process. An increase in the volume of these structures may indicate the duration of the inflammatory process.

Optimized multi-echo gradient-echo magnetic resonance imaging for gray and white matter segmentation in the lumbosacral cord at 3 T

Atrophy in the spinal cord (SC), gray (GM) and white matter (WM) is typically measured in-vivo by image segmentation on multi-echo gradient-echo magnetic resonance images. The aim of this study was to establish an acquisition and analysis protocol for optimal SC and GM segmentation in the lumbosacral cord at 3 T. Ten healthy volunteers underwent imaging of the lumbosacral cord using a 3D spoiled multi-echo gradient-echo sequence (Siemens FLASH, with 5 echoes and 8 repetitions) on a Siemens Prisma 3 T scanner. Optimal numbers of successive echoes and signal averages were investigated comparing signal-to-noise (SNR) and contrast-to-noise ratio (CNR) values as well as qualitative ratings for segmentability by experts. The combination of 5 successive echoes yielded the highest CNR between WM and cerebrospinal fluid and the highest rating for SC segmentability. The combination of 3 and 4 successive echoes yielded the highest CNR between GM and WM and the highest rating for GM segmentability in the lumbosacral enlargement and conus medullaris, respectively. For segmenting the SC and GM in the same image, we suggest combining 3 successive echoes. For SC or GM segmentation only, we recommend combining 5 or 3 successive echoes, respectively. Six signal averages yielded good contrast for reliable SC and GM segmentation in all subjects. Clinical applications could benefit from these recommendations as they allow for accurate SC and GM segmentation in the lumbosacral cord.

MobileUNetV3—A Combined UNet and MobileNetV3 Architecture for Spinal Cord Gray Matter Segmentation

The inspection of gray matter (GM) tissue of the human spinal cord is a valuable tool for the diagnosis of a wide range of neurological disorders. Thus, the detection and segmentation of GM regions in magnetic resonance images (MRIs) is an important task when studying the spinal cord and its related medical conditions. This work proposes a new method for the segmentation of GM tissue in spinal cord MRIs based on deep convolutional neural network (CNN) techniques. Our proposed method, called MobileUNetV3, has a UNet-like architecture, with the MobileNetV3 model being used as a pre-trained encoder. MobileNetV3 is light-weight and yields high accuracy compared with many other CNN architectures of similar size. It is composed of a series of blocks, which produce feature maps optimized using residual connections and squeeze-and-excitation modules. We carefully added a set of upsampling layers and skip connections to MobileNetV3 in order to build an effective UNet-like model for image segmentation. To illustrate the capabilities of the proposed method, we tested it on the spinal cord gray matter segmentation challenge dataset and compared it to a number of recent state-of-the-art methods. We obtained results that outperformed seven methods with respect to five evaluation metrics comprising the dice similarity coefficient (0.87), Jaccard index (0.78), sensitivity (87.20%), specificity (99.90%), and precision (87.96%). Based on these highly competitive results, MobileUNetV3 is an effective deep-learning model for the segmentation of GM MRIs in the spinal cord.

A fuzzy membership based comparison of the grey matter (GM) in cognitively normal (CN), mild cognitive impairment (MCI), and Alzheimer’s disease (AD) using brain images

Grey matter (GM) in human brain contains most of the important cells covering the regions involved in neurophysiological operations such as memory, emotions, decision making, etc. Alzheimer’s disease (AD) is a neurological disease that kills the brain cells in regions which are mostly involved in the neurophysiological operations. Mild Cognitive Impairment (MCI) is a stage between Cognitively Normal (CN) and AD, where a significant cognitive declination can be observed. The destruction of brain cells causes a reduction in the size of GM. Evaluation of changes in GM, may help in studying the overall brain transformations and accurate classification of different stages of AD. In this work, firstly skull of brain images is stripped for 5 different slices, then segmentation of GM is performed. Finally, the average number of pixels in grey region and the average atrophy in grey pixels per year is calculated and compared amongst CN, MCI, and AD patients of various ages and genders. It is observed that, for some subjects (in some particular ages) from different dementia stages, pattern of GM changes is almost identical. To solve this issue, we have used the concept of fuzzy membership functions to classify the dementia stages more accurately. It is observed from the comparison that average difference in the number of pixels between CN and MCI= 10.01%, CN and AD= 19.63%, MCI and AD= 10.72%. It can be also observed from the comparison that, the average atrophy in grey matter per year in CN= 1.92%, MCI= 3.13%, and AD= 4.33%.

Unravelling the impact of frontal lobe impairment for social dysfunction in myotonic dystrophy type 1

Myotonic dystrophy type 1 is an autosomal dominant multisystemic disorder affecting muscular and extra muscular systems, including the central nervous system. Cerebral involvement in myotonic dystrophy type 1 is associated with subtle cognitive and behavioural disorders, of major impact on socio-professional adaptation. The social dysfunction and its potential relation to frontal lobe neuropsychology remain under-evaluated in this pathology. The neuroanatomical network underpinning that disorder is yet to disentangle. Twenty-eight myotonic dystrophy type 1 adult patients (mean age: 46 years old) and 18 age and sex-matched healthy controls were included in the study. All patients performed an exhaustive neuropsychological assessment with a specific focus on frontal lobe neuropsychology (motivation, social cognition and executive functions). Among them, 18 myotonic dystrophy type 1 patients and 18 healthy controls had a brain MRI with T1 and T2 Flair sequences. Grey matter segmentation, Voxel-based morphometry and cortical thickness estimation were performed with Statistical Parametric Mapping Software SPM12 and Freesurfer software. Furthermore, T2 white matter lesions and subcortical structures were segmented with Automated Volumetry Software. Most patients showed significant impairment in executive frontal functions (auditory working memory, inhibition, contextualization and mental flexibility). Patients showed only minor difficulties in social cognition tests mostly in cognitive Theory of Mind, but with relative sparing of affective Theory of Mind and emotion recognition. Neuroimaging analysis revealed atrophy mostly in the parahippocampal and hippocampal regions and to a lesser extent in basal ganglia, regions involved in social navigation and mental flexibility, respectively. Social cognition scores were correlated with right parahippocampal gyrus atrophy. Social dysfunction in myotonic dystrophy type 1 might be a consequence of cognitive impairment regarding mental flexibility and social contextualization rather than a specific social cognition deficit such as emotion recognition. We suggest that both white matter lesions and grey matter disease could account for this social dysfunction, involving, in particular, the frontal-subcortical network and the hippocampal/arahippocampal regions, brain regions known, respectively, to integrate contextualization and social navigation.

Classification of Huntington's Disease Stage with Features Derived from Structural and Diffusion-Weighted Imaging.

The purpose of this study was to classify Huntington’s disease (HD) stage using support vector machines and measures derived from T1- and diffusion-weighted imaging. The effects of feature selection approach and combination of imaging modalities are assessed. Fourteen premanifest-HD individuals (Pre-HD; on average > 20 years from estimated disease onset), eleven early-manifest HD (Early-HD) patients, and eighteen healthy controls (HC) participated in the study. We compared three feature selection approaches: (i) whole-brain segmented grey matter (GM; voxel-based measure) or fractional anisotropy (FA) values; (ii) GM or FA values from subcortical regions-of-interest (caudate, putamen, pallidum); and (iii) automated selection of GM or FA values with the algorithm Relief-F. We assessed single- and multi-kernel approaches to classify combined GM and FA measures. Significant classifications were achieved between Early-HD and Pre-HD or HC individuals (accuracy: generally, 85% to 95%), and between Pre-HD and controls for the feature FA of the caudate ROI (74% accuracy). The combination of GM and FA measures did not result in higher performances. We demonstrate evidence on the high sensitivity of FA for the classification of the earliest Pre-HD stages, and successful distinction between HD stages.

Analysis of synthetic magnetic resonance images by multi-channel segmentation increases accuracy of volumetry in the putamen and decreases mis-segmentation in the dural sinuses.

Voxel-based morphometry (VBM) using magnetic resonance imaging (MR) has been used to estimate cortical atrophy associated with various diseases. However, there are mis-segmentations of segmented gray matter image in VBM. To study a twofold evaluation of single- and multi-channel segmentation using synthetic MR images: (1) mis-segmentation of segmented gray matter images in transverse and cavernous sinuses; and (2) accuracy and repeatability of segmented gray matter images. A total of 13 healthy individuals were scanned with 3D quantification using an interleaved Look-Locker acquisition sequence with a T2 preparation pulse (3D-QALAS) sequence on a 1.5-T scanner. Three of the 13 healthy participants were scanned five consecutive times for evaluation of repeatability. We used SyMRI software to create images with three contrasts: T1-weighted (T1W), T2-weighted (T2W), and proton density-weighted (PDW) images. Manual regions of interest (ROI) on T1W imaging were individually set as the gold standard in the transverse sinus, cavernous sinus, and putamen. Single-channel (T1W) and multi-channel (T1W + T2W, T1W + PDW, and T1W + T2W + PDW imaging) segmentations were performed with statistical parametric mapping 12 software. We found that mis-segmentations in both the transverse and cavernous sinuses were large in single-channel segmentation compared with multi-channel segmentations. Furthermore, the accuracy of segmented gray matter images in the putamen was high in both multi-channel T1W + PDW and T1W + T2W + PDW segmentations compared with other segmentations. Finally, the highest repeatability of left putamen volumetry was found with multi-channel segmentation T1WI + PDWI. Multi-channel segmentation with T1WI + PDWI provides good results for VBM compared with single-channel and other multi-channel segmentations.

The effect of lesion filling on brain network analysis in multiple sclerosis using structural magnetic resonance imaging

BackgroundGraph theoretical network analysis with structural magnetic resonance imaging (MRI) of multiple sclerosis (MS) patients can be used to assess subtle changes in brain networks. However, the presence of multiple focal brain lesions might impair the accuracy of automatic tissue segmentation methods, and hamper the performance of graph theoretical network analysis. Applying “lesion filling” by substituting the voxel intensities of a lesion with the voxel intensities of nearby voxels, thus creating an image devoid of lesions, might improve segmentation and graph theoretical network analysis. This study aims to determine if brain networks are different between MS subtypes and healthy controls (HC) and if the assessment of these differences is affected by lesion filling.MethodsThe study included 49 MS patients and 19 HC that underwent a T1w, and T2w-FLAIR MRI scan. Graph theoretical network analysis was performed from grey matter fractions extracted from the original T1w-images and T1w-images after lesion filling.ResultsArtefacts in lesion-filled T1w images correlated positively with total lesion volume (r = 0.84, p < 0.001) and had a major impact on grey matter segmentation accuracy. Differences in sensitivity for network alterations were observed between original T1w data and after application of lesion filling: graph theoretical network analysis obtained from lesion-filled T1w images produced more differences in network organization in MS patients.ConclusionLesion filling might reduce variability across subjects resulting in an increased detection rate of network alterations in MS, but also induces significant artefacts, and therefore should be applied cautiously especially in individuals with higher lesions loads.

Individual prediction and classification of cognitive impairment in patients with white matter lesions based on gray matter volume.

BackgroundVascular risk factors like white matter lesions (WMLs) are increasingly recognized as risk factors for vascular dementia (VaD) and can predict Alzheimer’s disease (AD) at least a decade before the clinical stage of the disease. This study aimed to predict cognitive decline and use machine learning techniques to classify older individuals (aged 50 years or older) with WMLs as having vascular mild cognitive impairment (VaMCI), VaD, or in good cognitive health (CH).MethodsA total of 79 individuals with WMLs were selected for this study and categorized into the following 3 groups: CH (n=25), VaMCI (n=33), and VaD (n=21). Data from the entire cohort was then divided into a training dataset (n=56) and testing dataset (n=23). The data were extracted from gray matter (GM) segmentations using voxel-based morphometry (VBM). A relevance vector regression (RVR) approach was used to test the relationship between the structural brain images and clinical scores. To predict the individual-level subtypes, we applied 2 different machine learning-based classifiers: support vector machine (SVM) and Gaussian process classification (GPC). All predictive models were trained on the training dataset and then validated on the testing dataset of age-matched participants.ResultsMulti-domain cognitive performance could be predicted based on the pattern of GM atrophy in older people with WMLs using a RVR approach. The classification of VaD versus CH (cross-validation accuracy =93.94%, test set accuracy =76.92%) and VaMCI versus CH (cross-validation accuracy =95.24%, test set accuracy =87.50%) could be successfully achieved using both SVM and GPC. However, SVM (cross-validation accuracy =67.57%, test set accuracy =70.59%) performed better than GPC in the classification of VaD versus VaMCI.ConclusionsBased on the patterns of gray matter and RVR-based model could achieve prediction of cognitive test scores, and SVM and GPC could classify the severity of cognitive impairment in older people with WMLs.

Optimization of magnetization transfer contrast for EPI FLAIR brain imaging.

To evaluate the impact of magnetization transfer (MT) on brain tissue contrast in turbo-spin-echo (TSE) and EPI fluid-attenuated inversion recovery (FLAIR) images, and to optimize an MT-prepared EPI FLAIR pulse sequence to match the tissue contrast of a clinical reference TSE FLAIR protocol. Five healthy volunteers underwent 3T brain MRI, including single slice TSE FLAIR, multi-slice TSE FLAIR, EPI FLAIR without MT-preparation, and MT-prepared EPI FLAIR with variations of the MT-preparation parameters, including number of preparation pulses, pulse amplitude, and resonance offset. Automated co-registration and gray matter (GM) versus white matter (WM) segmentation was performed using a T1-MPRAGE acquisition, and the GM versus WM signal intensity ratio (contrast ratio) was calculated for each FLAIR acquisition. Without MT preparation, EPI FLAIR showed poor tissue contrast (contrast ratio = 0.98), as did single slice TSE FLAIR. Multi-slice TSE FLAIR provided high tissue contrast (contrast ratio = 1.14). MT-prepared EPI FLAIR closely approximated the contrast of the multi-slice TSE FLAIR images for two combinations of the MT-preparation parameters (contrast ratio = 1.14). Optimized MT-prepared EPI FLAIR provided a 50% reduction in scan time compared to the reference TSE FLAIR acquisition. Optimized MT-prepared EPI FLAIR provides comparable brain tissue contrast to the multi-slice TSE FLAIR images used in clinical practice.

Predicting the Brain Age of Children with Cerebral Palsy Using a Two-Dimensional Convoluted Neural Network Prediction Model Without Gray and White Matter Segmentation

Predicting the Brain Age of Children with Cerebral Palsy Using a Two-Dimensional Convoluted Neural Network Prediction Model Without Gray and White Matter Segmentation

Brain opioid segments and striatal patterns of dopamine release induced by naloxone and morphine.

Opioid receptors are expressed throughout the brain and play a major role in regulating striatal dopamine (DA) release. Clinical studies have shown that naloxone (NAL, a nonspecific opioid antagonist) in individuals with opioid use disorder and morphine (MRP, a nonspecific opioid agonist) in healthy controls, resulted in DA release in the dorsal and ventral striatum, respectively. It is not known whether the underlying patterns of striatal DA release are associated with the striatal distribution of opioid receptors. We leveraged previously published PET datasets (collected in independent cohorts) to study the brain‐wide distribution of opioid receptors and to compare striatal opioid receptor availability with striatal DA release patterns. We identified three major gray matter segments based on availability maps of DA and opioid receptors: striatum, and primary and secondary opioid segments with high and intermediate opioid receptor availability, respectively. Patterns of DA release induced by NAL and MRP were inversely associated and correlated with kappa (NAL: r(68) = −0.81, MRP: r(68) = 0.54), and mu (NAL: r(68) = −0.62, MRP: r(68) = 0.46) opioid receptor availability. Kappa opioid receptor availability accounted for a unique part of variance in NAL‐ and MRP‐DA release patterns (ΔR 2 >0.14, p <.0001). In sum, distributions of opioid receptors distinguished major cortical and subcortical regions. Patterns of NAL‐ and MRP‐induced DA release had inverse associations with striatal opioid receptor availability. Our approach provides a pattern‐based characterization of drug‐induced DA targets and is relevant for modeling the role of opioid receptors in modulating striatal DA release.

S-CUDA: Self-cleansing unsupervised domain adaptation for medical image segmentation.

Medical image segmentation tasks hitherto have achieved excellent progresses with large-scale datasets, which empowers us to train potent deep convolutional neural networks (DCNNs). However, labeling such large-scale datasets is laborious and error-prone, which leads the noisy (or incorrect) labels to be an ubiquitous problem in the real-world scenarios. In addition, data collected from different sites usually exhibit significant data distribution shift (or domain shift). As a result, noisy label and domain shift become two common problems in medical imaging application scenarios, especially in medical image segmentation, which degrade the performance of deep learning models significantly. In this paper, we identify a novel problem hidden in medical image segmentation, which is unsupervised domain adaptation on noisy labeled data, and propose a novel algorithm named "Self-Cleansing Unsupervised Domain Adaptation" (S-CDUA) to address such issue. S-CUDA sets up a realistic scenario to solve the above problems simultaneously where training data (i.e., source domain) not only shows domain shift w.r.t. unsupervised test data (i.e., target domain) but also contains noisy labels. The key idea of S-CUDA is to learn noise-excluding and domain invariant knowledge from noisy supervised data, which will be applied on the highly corrupted data for label cleansing and further data-recycling, as well as on the test data with domain shift for supervised propagation. To this end, we propose a novel framework leveraging noisy-label learning and domain adaptation techniques to cleanse the noisy labels and learn from trustable clean samples, thus enabling robust adaptation and prediction on the target domain. Specifically, we train two peer adversarial networks to identify high-confidence clean data and exchange them in companions to eliminate the error accumulation problem and narrow the domain gap simultaneously. In the meantime, the high-confidence noisy data are detected and cleansed in order to reuse the contaminated training data. Therefore, our proposed method can not only cleanse the noisy labels in the training set but also take full advantage of the existing noisy data to update the parameters of the network. For evaluation, we conduct experiments on two popular datasets (REFUGE and Drishti-GS) for optic disc (OD) and optic cup (OC) segmentation, and on another public multi-vendor dataset for spinal cord gray matter (SCGM) segmentation. Experimental results show that our proposed method can cleanse noisy labels efficiently and obtain a model with better generalization performance at the same time, which outperforms previous state-of-the-art methods by large margin. Our code can be found at https://github.com/zzdxjtu/S-cuda.

A Semi-supervised Learning for Segmentation of Gigapixel Histopathology Images from Brain Tissues.

Automated segmentation of grey matter (GM) and white matter (WM) in gigapixel histopathology images is advantageous to analyzing distributions of disease pathologies, further aiding in neuropathologic deep phenotyping. Although supervised deep learning methods have shown good performance, its requirement of a large amount of labeled data may not be cost-effective for large scale projects. In the case of GM/WM segmentation, trained experts need to carefully trace the delineation in gigapixel images. To minimize manual labeling, we consider semi-surprised learning (SSL) and deploy one state-of-the-art SSL method (FixMatch) on WSIs. Then we propose a two-stage scheme to further improve the performance of SSL: the first stage is a self-supervised module to train an encoder to learn the visual representations of unlabeled data, subsequently, this well-trained encoder will be an initialization of consistency loss-based SSL in the second stage. We test our method on Amyloid-β stained histopathology images and the results outperform FixMatch with the mean IoU score at around 2% by using 6,000 labeled tiles while over 10% by using only 600 labeled tiles from 2 WSIs.Clinical relevance- this work minimizes the required labeling efforts by trained personnel. An improved GM/WM segmentation method could further aid in the study of brain diseases, such as Alzheimer's disease.

In vivo T1 and T2 relaxation time maps of brain tissue, skeletal muscle, and lipid measured in healthy volunteers at 50 mT.

PurposeLow‐field (B0 < 0.1 T) MRI has generated much interest as a means of increased accessibility via reduced cost and improved portability compared to conventional clinical systems (B0 ≥ 1.5 Tesla). Here we measure MR relaxation times at 50 mT and compare results with commonly used models based on both in vivo and ex vivo measurements.MethodsUsing 3D turbo spin echo readouts, T1 and T2 maps of the human brain and lower leg were acquired on a custom‐built 50 mT MRI scanner using inversion‐recovery and multi‐echo–based sequences, respectively. Image segmentation was performed based on a histogram analysis of the relaxation times.ResultsThe average T1 times of gray matter, white matter, and cerebrospinal fluid (CSF) were 327 ± 10 ms, 275 ± 5 ms, and 3695 ± 287 ms, respectively. Corresponding values of T2 were 102 ± 6 ms, 102 ± 6 ms, and 1584 ± 124 ms. T1 times in the calf muscle were measured to be 171 ± 11 ms and were 130 ± 5 ms in subcutaneous and bone marrow lipid. Corresponding T2 times were 39 ± 2 ms in muscle and 90 ± 13 ms in lipid.ConclusionsFor tissues except for CSF, the measured T1 times are much shorter than reported at higher fields and generally lie within the range of different models in the literature. As expected, T2 times are similar to those seen at typical clinical field strengths. Analysis of the relaxation maps indicates that segmentation of white and gray matter based purely on T1 or T2 will be quite challenging at low field given the relatively small difference in relaxation times.

White Matter, Gray Matter and Cerebrospinal Fluid Segmentation from Brain Magnetic Resonance Imaging Using Adaptive U-Net and Local Convolutional Neural Network

Abstract According to the World Alzheimer Report 2015, 46 million people are living with dementia in the world. The diagnosis of diseases helps doctors treating patients better. One of the signs of diseases is related to white matter, grey matter and cerebrospinal fluid. Therefore, the automatic segmentation of three tissues in brain imaging especially from magnetic resonance imaging (MRI) plays an important role in medical analysis. In this research, we proposed an effective approach to segment automatically these tissues in three-dimensional (3D) brain MRI. First, a deep learning model is used to segment the sure and unsure regions. In the unsure region, another deep learning model is used to classify each pixel. In the experiments, an adaptive U-net model is used to segment the sure and unsure regions, and the Local Convolutional Neural Network (CNN) model with multiple inputs is used to classify each pixel only in the unsure region. Our method was evaluated with a real image database, Internet Brain Segmentation Repository database, with 18 persons (IBSR 18) (https://www.nitrc.org/projects/ibsr) and compared with state of art methods being the results very promising.

A Heteroscedastic Uncertainty Model for Decoupling Sources of MRI Image Quality

Quality control (QC) of MR images is essential to ensure that downstream analyses such as segmentation can be performed successfully. Currently, QC is predominantly performed visually and subjectively, at significant time and operator cost. We aim to automate the process using a probabilistic network that estimates segmentation uncertainty through a heteroscedastic noise model, providing a measure of task-specific quality. By augmenting training images with k-space artefacts, we propose a novel CNN architecture to decouple sources of uncertainty related to the task and different k-space artefacts in a self-supervised manner. This enables the prediction of separate uncertainties for different types of data degradation. While the uncertainty predictions reflect the presence and severity of artefacts, the network provides more robust and generalisable segmentation predictions given the quality of the data. We show that models trained with artefact augmentation provide informative measures of uncertainty on both simulated artefacts and problematic real-world images identified by human-raters, both qualitatively and quantitatively in the form of error bars on volume measurements. Relating artefact uncertainty to segmentation Dice scores, we observe that our uncertainty predictions provide a better estimate of MRI quality from the point of view of the task (gray matter segmentation) compared to commonly used metrics of quality including signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR), hence providing a real-time quality metric indicative of segmentation quality.

Segmentation of WM, GM, and CSF from the Brain MRIs Using an Advanced Metaheuristic Approach

Segmentation of brain tissues from the Magnetic Resonance Image (MRI) is vital to detect and diagnose brain-related diseases. Precise brain MRI segmentation is challenging because of the unpredictable anatomical structure of the brain, existence of Intensity Inhomogeneity (IIH), Partial Volume (PV) effects, and noise. Most of the brain disorders can be identified exactly by the segmentation of brain tissues, such as White Matter (WM), Gray Matter (GM), and Cerebrospinal Fluid (CSF). The optimization algorithmic approach is widely used for the brain image segmentation. In this work, a robust metaheuristic Multi-level Thresholding (MT) technique named as Advanced Wind Driven Optimization (AWDO) is employed for the brain MRI segmentation. The optimal MT values can be attained by combining the AWDO with the very popular Otsu’s objective function. The results obtained by the proposed AWDO in segmenting various brain images at an optimized value of threshold are compared with the groundtruths and have given the high values for the metrics such as sensitivity, specificity, segmentation accuracy, dice coefficient, and Precision. The values of the objective function, Peak Signal to Noise Ratio (PSNR), Standard Deviation (SD), and Convergence time of the proposed method are also being obtained at various threshold levels. The results are compared with the results of some existing algorithms like Multiplicative Intrinsic Component Optimization (MICO), Electromagnetism Optimization (EMO), Histogram based Darwinian Particle Swarm Optimization (HDPSO), Fully Convolutional Neural Networks (FCNN), and Advance Ant Colony Optimization (AACO). The results justified the notable performance of the proposed approach in segmenting different brain tissues.