Automated model-based bias field correction of MR images of the brain

Similarity and quality metrics for MR image-to-image translation

Image-to-image translation can create large impact in medical imaging, as images can be synthetically transformed to other modalities, sequence types, higher resolutions or lower noise levels. To ensure patient safety, these methods should be validated by human readers, which requires a considerable amount of time and costs. Quantitative metrics can effectively complement such studies and provide reproducible and objective assessment of synthetic images. If a reference is available, the similarity of MR images is frequently evaluated by SSIM and PSNR metrics, even though these metrics are not or too sensitive regarding specific distortions. When reference images to compare with are not available, non-reference quality metrics can reliably detect specific distortions, such as blurriness. To provide an overview on distortion sensitivity, we quantitatively analyze 11 similarity (reference) and 12 quality (non-reference) metrics for assessing synthetic images. We additionally include a metric on a downstream segmentation task. We investigate the sensitivity regarding 11 kinds of distortions and typical MR artifacts, and analyze the influence of different normalization methods on each metric and distortion. Finally, we derive recommendations for effective usage of the analyzed similarity and quality metrics for evaluation of image-to-image translation models.

Development of A Radiomic Model for MGMT Promoter Methylation Detection in Glioblastoma Using Conventional MRI

The methylation of the O6-methylguanine-DNA methyltransferase (MGMT) promoter is a molecular marker associated with a better response to chemotherapy in patients with glioblastoma (GB). Standard pre-operative magnetic resonance imaging (MRI) analysis is not adequate to detect MGMT promoter methylation. This study aims to evaluate whether the radiomic features extracted from multiple tumor subregions using multiparametric MRI can predict MGMT promoter methylation status in GB patients. This retrospective single-institution study included a cohort of 277 GB patients whose 3D post-contrast T1-weighted images and 3D fluid-attenuated inversion recovery (FLAIR) images were acquired using two MRI scanners. Three separate regions of interest (ROIs) showing tumor enhancement, necrosis, and FLAIR hyperintensities were manually segmented for each patient. Two machine learning algorithms (support vector machine (SVM) and random forest) were built for MGMT promoter methylation prediction from a training cohort (196 patients) and tested on a separate validation cohort (81 patients), based on a set of automatically selected radiomic features, with and without demographic variables (i.e., patients' age and sex). In the training set, SVM based on the selected radiomic features of the three separate ROIs achieved the best performances, with an average of 83.0% (standard deviation: 5.7%) for accuracy and 0.894 (0.056) for the area under the curve (AUC) computed through cross-validation. In the test set, all classification performances dropped: the best was obtained by SVM based on the selected features extracted from the whole tumor lesion constructed by merging the three ROIs, with 64.2% (95% confidence interval: 52.8-74.6%) accuracy and 0.572 (0.439-0.705) for AUC. The performances did not change when the patients' age and sex were included with the radiomic features into the models. Our study confirms the presence of a subtle association between imaging characteristics and MGMT promoter methylation status. However, further verification of the strength of this association is needed, as the low diagnostic performance obtained in this validation cohort is not sufficiently robust to allow clinically meaningful predictions.

MGDUN: An interpretable network for multi-contrast MRI image super-resolution reconstruction

Magnetic resonance imaging (MRI) Super-Resolution (SR) aims to obtain high resolution (HR) images with more detailed information for precise diagnosis and quantitative image analysis. Deep unfolding networks outperform general MRI SR reconstruction methods by providing better performance and improved interpretability, which enhance the trustworthiness required in clinical practice. Additionally, current SR reconstruction techniques often rely on a single contrast or a simple multi-contrast fusion mechanism, ignoring the complex relationships between different contrasts. To address these issues, in this paper, we propose a Model-Guided multi-contrast interpretable Deep Unfolding Network (MGDUN) for medical image SR reconstruction, which explicitly incorporates the well-studied multi-contrast MRI observation model into an unfolding iterative network. Specifically, we manually design an objective function for MGDUN that can be iteratively computed by the half-quadratic splitting algorithm. The iterative MGDUN algorithm is unfolded into a special model-guided deep unfolding network that explicitly takes into account both the multi-contrast relationship matrix and the MRI observation matrix during the end-to-end optimization process. Extensive experimental results on the multi-contrast IXI dataset and the BraTs 2019 dataset demonstrate the superiority of our proposed model, with PSNR reaching 37.3366 and 35.9690 respectively. Our proposed MGDUN provides a promising solution for multi-contrast MR image super-resolution reconstruction. Code is available at https://github.com/yggame/MGDUN.

A cellular resolution atlas of Broca's area

Brain cells are arranged in laminar, nuclear, or columnar structures, spanning a range of scales. Here, we construct a reliable cell census in the frontal lobe of human cerebral cortex at micrometer resolution in a magnetic resonance imaging (MRI)-referenced system using innovative imaging and analysis methodologies. MRI establishes a macroscopic reference coordinate system of laminar and cytoarchitectural boundaries. Cell counting is obtained with a digital stereological approach on the 3D reconstruction at cellular resolution from a custom-made inverted confocal light-sheet fluorescence microscope (LSFM). Mesoscale optical coherence tomography enables the registration of the distorted histological cell typing obtained with LSFM to the MRI-based atlas coordinate system. The outcome is an integrated high-resolution cellular census of Broca's area in a human postmortem specimen, within a whole-brain reference space atlas.

An open-source tool for longitudinal whole-brain and white matter lesion segmentation

In this paper we describe and validate a longitudinal method for whole-brain segmentation of longitudinal MRI scans. It builds upon an existing whole-brain segmentation method that can handle multi-contrast data and robustly analyze images with white matter lesions. This method is here extended with subject-specific latent variables that encourage temporal consistency between its segmentation results, enabling it to better track subtle morphological changes in dozens of neuroanatomical structures and white matter lesions. We validate the proposed method on multiple datasets of control subjects and patients suffering from Alzheimer's disease and multiple sclerosis, and compare its results against those obtained with its original cross-sectional formulation and two benchmark longitudinal methods. The results indicate that the method attains a higher test-retest reliability, while being more sensitive to longitudinal disease effect differences between patient groups. An implementation is publicly available as part of the open-source neuroimaging package FreeSurfer.

Is musical engagement enough to keep the brain young?

Music-making and engagement in music-related activities have shown procognitive benefits for healthy and pathological populations, suggesting reductions in brain aging. A previous brain aging study, using Brain Age Gap Estimation (BrainAGE), showed that professional and amateur-musicians had younger appearing brains than non-musicians. Our study sought to replicate those findings and analyze if musical training or active musical engagement was necessary to produce an age-decelerating effect in a cohort of healthy individuals. We scanned 125 healthy controls and investigated if musician status, and if musical behaviors, namely active engagement (AE) and musical training (MT) [as measured using the Goldsmiths Musical Sophistication Index (Gold-MSI)], had effects on brain aging. Our findings suggest that musician status is not related to BrainAGE score, although involvement in current physical activity is. Although neither MT nor AE subscales of the Gold-MSI are predictive for BrainAGE scores, dispositional resilience, namely the ability to deal with challenge, is related to both musical behaviors and sensitivity to musical pleasure. While the study failed to replicate the findings in a previous brain aging study, musical training and active musical engagement are related to the resilience factor of challenge. This finding may reveal how such musical behaviors can potentially strengthen the brain's resilience to age, which may tap into a type of neurocognitive reserve.

Quality reporting of radiomics analysis in pituitary adenomas: promoting clinical translation

OBJECTIVE

To evaluate the quality of radiomics studies on pituitary adenoma according to the radiomics quality score (RQS) and Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD).

METHODS

PubMed MEDLINE and EMBASE were searched to identify radiomics studies on pituitary adenomas. From 138 articles, 20 relevant original research articles were included. Studies were scored based on RQS and TRIPOD guidelines.

RESULTS

Most included studies did not perform pre-processing; isovoxel resampling, signal intensity normalization, and N4 bias field correction were performed in only five (25%), eight (40%), and four (20%) studies, respectively. Only two (10%) studies performed external validation. The mean RQS and basic adherence rate were 2.8 (7.6%) and 26.6%, respectively. There was a low adherence rate for conducting comparison to "gold-standard" (20%), multiple segmentation (25%), and stating potential clinical utility (25%). No study stated the biological correlation, conducted a test-retest or phantom study, was a prospective study, conducted cost-effectiveness analysis, or provided open-source code and data, which resulted in low-level evidence. The overall adherence rate for TRIPOD was 54.6%, and it was low for reporting the title (5%), abstract (0%), explaining the sample size (10%), and suggesting a full prediction model (5%).

CONCLUSION

The radiomics reporting quality for pituitary adenoma is insufficient. Pre-processing is required for feature reproducibility and external validation is necessary. Feature reproducibility, clinical utility demonstration, higher evidence levels, and open science are required. Titles, abstracts, and full prediction model suggestions should be improved for transparent reporting.

ADVANCES IN KNOWLEDGE

Despite the rapidly increasing number of radiomics researches on pituitary adenoma, the quality of science in these researches is unknown. Our study indicates that the overall quality needs to be significantly improved in radiomics studies on pituitary adenoma, and since the concept of RQS and IBSI is still unfamiliar to clinicians and radiologist researchers, our study may help to reach higher technical and clinical impact in the future study.

Deep Learning in Large and Multi-Site Structural Brain MR Imaging Datasets

Large, multi-site, heterogeneous brain imaging datasets are increasingly required for the training, validation, and testing of advanced deep learning (DL)-based automated tools, including structural magnetic resonance (MR) image-based diagnostic and treatment monitoring approaches. When assembling a number of smaller datasets to form a larger dataset, understanding the underlying variability between different acquisition and processing protocols across the aggregated dataset (termed “batch effects”) is critical. The presence of variation in the training dataset is important as it more closely reflects the true underlying data distribution and, thus, may enhance the overall generalizability of the tool. However, the impact of batch effects must be carefully evaluated in order to avoid undesirable effects that, for example, may reduce performance measures. Batch effects can result from many sources, including differences in acquisition equipment, imaging technique and parameters, as well as applied processing methodologies. Their impact, both beneficial and adversarial, must be considered when developing tools to ensure that their outputs are related to the proposed clinical or research question (i.e., actual disease-related or pathological changes) and are not simply due to the peculiarities of underlying batch effects in the aggregated dataset. We reviewed applications of DL in structural brain MR imaging that aggregated images from neuroimaging datasets, typically acquired at multiple sites. We examined datasets containing both healthy control participants and patients that were acquired using varying acquisition protocols. First, we discussed issues around Data Access and enumerated the key characteristics of some commonly used publicly available brain datasets. Then we reviewed methods for correcting batch effects by exploring the two main classes of approaches: Data Harmonization that uses data standardization, quality control protocols or other similar algorithms and procedures to explicitly understand and minimize unwanted batch effects; and Domain Adaptation that develops DL tools that implicitly handle the batch effects by using approaches to achieve reliable and robust results. In this narrative review, we highlighted the advantages and disadvantages of both classes of DL approaches, and described key challenges to be addressed in future studies.

A Coarse-to-Fine Deformable Transformation Framework for Unsupervised Multi-Contrast MR Image Registration with Dual Consistency Constraint

Multi-contrast magnetic resonance (MR) image registration is useful in the clinic to achieve fast and accurate imaging-based disease diagnosis and treatment planning. Nevertheless, the efficiency and performance of the existing registration algorithms can still be improved. In this paper, we propose a novel unsupervised learning-based framework to achieve accurate and efficient multi-contrast MR image registration. Specifically, an end-to-end coarse-to-fine network architecture consisting of affine and deformable transformations is designed to improve the robustness and achieve end-to-end registration. Furthermore, a dual consistency constraint and a new prior knowledge-based loss function are developed to enhance the registration performances. The proposed method has been evaluated on a clinical dataset containing 555 cases, and encouraging performances have been achieved. Compared to the commonly utilized registration methods, including VoxelMorph, SyN, and LT-Net, the proposed method achieves better registration performance with a Dice score of 0.8397± 0.0756 in identifying stroke lesions. With regards to the registration speed, our method is about 10 times faster than the most competitive method of SyN (Affine) when testing on a CPU. Moreover, we prove that our method can still perform well on more challenging tasks with lacking scanning information data, showing the high robustness for the clinical application.

An Optimized Mouse Brain Atlas for Automated Mapping and Quantification of Neuronal Activity Using iDISCO+ and Light Sheet Fluorescence Microscopy

In recent years, the combination of whole-brain immunolabelling, light sheet fluorescence microscopy (LSFM) and subsequent registration of data with a common reference atlas, has enabled 3D visualization and quantification of fluorescent markers or tracers in the adult mouse brain. Today, the common coordinate framework version 3 developed by the Allen’s Institute of Brain Science (AIBS CCFv3), is widely used as the standard brain atlas for registration of LSFM data. However, the AIBS CCFv3 is based on histological processing and imaging modalities different from those used for LSFM imaging and consequently, the data differ in both tissue contrast and morphology. To improve the accuracy and speed by which LSFM-imaged whole-brain data can be registered and quantified, we have created an optimized digital mouse brain atlas based on immunolabelled and solvent-cleared brains. Compared to the AIBS CCFv3 atlas, our atlas resulted in faster and more accurate mapping of neuronal activity as measured by c-Fos expression, especially in the hindbrain. We further demonstrated utility of the LSFM atlas by comparing whole-brain quantitative changes in c-Fos expression following acute administration of semaglutide in lean and diet-induced obese mice. In combination with an improved algorithm for c-Fos detection, the LSFM atlas enables unbiased and computationally efficient characterization of drug effects on whole-brain neuronal activity patterns. In conclusion, we established an optimized reference atlas for more precise mapping of fluorescent markers, including c-Fos, in mouse brains processed for LSFM.

An efficient brain tumor image classifier by combining multi-pathway cascaded deep neural network and handcrafted features in MR images

Accurate segmentation and delineation of the sub-tumor regions are very challenging tasks due to the nature of the tumor. Traditionally, convolutional neural networks (CNNs) have succeeded in achieving most promising performance for the segmentation of brain tumor; however, handcrafted features remain very important in identification of tumor's boundary regions accurately. The present work proposes a robust deep learning-based model with three different CNN architectures along with pre-defined handcrafted features for brain tumor segmentation, mainly to find out more prominent boundaries of the core and enhanced tumor regions. Generally, automatic CNN architecture does not use the pre-defined handcrafted features because it extracts the features automatically. In this present work, several pre-defined handcrafted features are computed from four MRI modalities (T2, FLAIR, T1c, and T1) with the help of additional handcrafted masks according to user interest and fed to the convolutional features (automatic features) to improve the overall performance of the proposed CNN model for tumor segmentation. Multi-pathway CNN is explored in this present work along with single-pathway CNN, which extracts simultaneously both local and global features to identify the accurate sub-regions of the tumor with the help of handcrafted features. The present work uses a cascaded CNN architecture, where the outcome of a CNN is considered as an additional input information to next subsequent CNNs. To extract the handcrafted features, convolutional operation was applied on the four MRI modalities with the help of several pre-defined masks to produce a predefined set of handcrafted features. The present work also investigates the usefulness of intensity normalization and data augmentation in pre-processing stage in order to handle the difficulties related to the imbalance of tumor labels. The proposed method is experimented on the BraST 2018 datasets and achieved promising results than the existing (currently published) methods with respect to different metrics such as specificity, sensitivity, and dice similarity coefficient (DSC) for complete, core, and enhanced tumor regions. Quantitatively, a notable gain is achieved around the boundaries of the sub-tumor regions using the proposed two-pathway CNN along with the handcrafted features. Graphical Abstract This data is mandatory. Please provide.

An Optimized Mouse Brain Atlas for Automated Mapping and Quantification of Neuronal Activity Using iDISCO+ and Light Sheet Fluorescence Microscopy

In recent years, the combination of whole-brain immunolabelling, light sheet fluorescence microscopy (LSFM) and subsequent registration of data with a common reference atlas, has enabled 3D visualization and quantification of fluorescent markers or tracers in the adult mouse brain. Today, the common coordinate framework version 3 developed by the Allen's Institute of Brain Science (AIBS CCFv3), is widely used as the standard brain atlas for registration of LSFM data. However, the AIBS CCFv3 is based on histological processing and imaging modalities different from those used for LSFM imaging and consequently, the data differ in both tissue contrast and morphology. To improve the accuracy and speed by which LSFM-imaged whole-brain data can be registered and quantified, we have created an optimized digital mouse brain atlas based on immunolabelled and solvent-cleared brains. Compared to the AIBS CCFv3 atlas, our atlas resulted in faster and more accurate mapping of neuronal activity as measured by c-Fos expression, especially in the hindbrain. We further demonstrated utility of the LSFM atlas by comparing whole-brain quantitative changes in c-Fos expression following acute administration of semaglutide in lean and diet-induced obese mice. In combination with an improved algorithm for c-Fos detection, the LSFM atlas enables unbiased and computationally efficient characterization of drug effects on whole-brain neuronal activity patterns. In conclusion, we established an optimized reference atlas for more precise mapping of fluorescent markers, including c-Fos, in mouse brains processed for LSFM.

Quality assessment of meningioma radiomics studies: Bridging the gap between exploratory research and clinical applications

PURPOSE

To evaluate the quality of radiomics studies on meningiomas, using a radiomics quality score (RQS), Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD), and the Image Biomarker Standardization Initiative (IBSI).

METHODS

PubMed MEDLINE and Embase were searched to identify radiomics studies on meningiomas. Of 138 identified articles, 25 relevant original research articles were included. Studies were scored according to the RQS, TRIPOD guidelines, and items in IBSI.

RESULTS

Only four studies (16 %) performed external validation. The mean RQS was 5.6 out of 36 (15.4 %), and the basic adherence rate was 26.8 %. The adherence rate was low for stating biological correlation (4%), conducting calibration statistics (12 %), multiple segmentation (16 %), and stating potential clinical utility (16 %). None of the studies conducted a test‒retest or phantom study, stated a comparison to a 'gold standard', conducted prospective studies or cost-effectivity analysis, or opened code and data to the public, resulting in low RQS. The overall adherence rate for TRIPOD was 54.1 %, with low scores for reporting the title (4%), abstract (0%), blind assessment of the outcome (8%), and explaining the sample size (0%). According to IBSI items, only 6 (24 %), 6 (24 %), and 3 (12 %) studies performed N4 bias-field correction, isovoxel resampling, and grey-level discretization, respectively. No study performed skull stripping.

CONCLUSION

The quality of radiomics studies for meningioma is insufficient. Acknowledgement of RQS, TRIPOD, and IBSI reporting guidelines may improve the quality of meningioma radiomics studies and enable their clinical application.

A Sparse Volume Reconstruction Method for Fetal Brain MRI Using Adaptive Kernel Regression

Slice-to-volume reconstruction (SVR) method can deal well with motion artifacts and provide high-quality 3D image data for fetal brain MRI. However, the problem of sparse sampling is not well addressed in the SVR method. In this paper, we mainly focus on the sparse volume reconstruction of fetal brain MRI from multiple stacks corrupted with motion artifacts. Based on the SVR framework, our approach includes the slice-to-volume 2D/3D registration, the point spread function- (PSF-) based volume update, and the adaptive kernel regression-based volume update. The adaptive kernel regression can deal well with the sparse sampling data and enhance the detailed preservation by capturing the local structure through covariance matrix. Experimental results performed on clinical data show that kernel regression results in statistical improvement of image quality for sparse sampling data with the parameter setting of the structure sensitivity 0.4, the steering kernel size of 7 × 7 × 7 and steering smoothing bandwidth of 0.5. The computational performance of the proposed GPU-based method can be over 90 times faster than that on CPU.

Quick guide on radiology image pre-processing for deep learning applications in prostate cancer research

Purpose: Deep learning has achieved major breakthroughs during the past decade in almost every field. There are plenty of publicly available algorithms, each designed to address a different task of computer vision in general. However, most of these algorithms cannot be directly applied to images in the medical domain. Herein, we are focused on the required preprocessing steps that should be applied to medical images prior to deep neural networks. Approach: To be able to employ the publicly available algorithms for clinical purposes, we must make a meaningful pixel/voxel representation from medical images which facilitates the learning process. Based on the ultimate goal expected from an algorithm (classification, detection, or segmentation), one may infer the required pre-processing steps that can ideally improve the performance of that algorithm. Required pre-processing steps for computed tomography (CT) and magnetic resonance (MR) images in their correct order are discussed in detail. We further supported our discussion by relevant experiments to investigate the efficiency of the listed preprocessing steps. Results: Our experiments confirmed how using appropriate image pre-processing in the right order can improve the performance of deep neural networks in terms of better classification and segmentation. Conclusions: This work investigates the appropriate pre-processing steps for CT and MR images of prostate cancer patients, supported by several experiments that can be useful for educating those new to the field (https://github.com/NIH-MIP/Radiology_Image_Preprocessing_for_Deep_Learning).

Quantitative definition of neurobehavior, vision, hearing and brain volumes in macaques congenitally exposed to Zika virus

Congenital Zika virus (ZIKV) exposure results in a spectrum of disease ranging from severe birth defects to delayed onset neurodevelopmental deficits. ZIKV-related neuropathogenesis, predictors of birth defects, and neurodevelopmental deficits are not well defined in people. Here we assess the methodological and statistical feasibility of a congenital ZIKV exposure macaque model for identifying infant neurobehavior and brain abnormalities that may underlie neurodevelopmental deficits. We inoculated five pregnant macaques with ZIKV and mock-inoculated one macaque in the first trimester. Following birth, growth, ocular structure/function, brain structure, hearing, histopathology, and neurobehavior were quantitatively assessed during the first week of life. We identified the typical pregnancy outcomes of congenital ZIKV infection, with fetal demise and placental abnormalities. We estimated sample sizes needed to define differences between groups and demonstrated that future studies quantifying brain region volumes, retinal structure, hearing, and visual pathway function require a sample size of 14 animals per group (14 ZIKV, 14 control) to detect statistically significant differences in at least half of the infant exam parameters. Establishing the parameters for future studies of neurodevelopmental outcomes following congenital ZIKV exposure in macaques is essential for robust and rigorous experimental design.

A contrast-adaptive method for simultaneous whole-brain and lesion segmentation in multiple sclerosis

Here we present a method for the simultaneous segmentation of white matter lesions and normal-appearing neuroanatomical structures from multi-contrast brain MRI scans of multiple sclerosis patients. The method integrates a novel model for white matter lesions into a previously validated generative model for whole-brain segmentation. By using separate models for the shape of anatomical structures and their appearance in MRI, the algorithm can adapt to data acquired with different scanners and imaging protocols without retraining. We validate the method using four disparate datasets, showing robust performance in white matter lesion segmentation while simultaneously segmenting dozens of other brain structures. We further demonstrate that the contrast-adaptive method can also be safely applied to MRI scans of healthy controls, and replicate previously documented atrophy patterns in deep gray matter structures in MS. The algorithm is publicly available as part of the open-source neuroimaging package FreeSurfer.

Accurate and robust whole-head segmentation from magnetic resonance images for individualized head modeling

Transcranial brain stimulation (TBS) has been established as a method for modulating and mapping the function of the human brain, and as a potential treatment tool in several brain disorders. Typically, the stimulation is applied using a one-size-fits-all approach with predetermined locations for the electrodes, in electric stimulation (TES), or the coil, in magnetic stimulation (TMS), which disregards anatomical variability between individuals. However, the induced electric field distribution in the head largely depends on anatomical features implying the need for individually tailored stimulation protocols for focal dosing. This requires detailed models of the individual head anatomy, combined with electric field simulations, to find an optimal stimulation protocol for a given cortical target. Considering the anatomical and functional complexity of different brain disorders and pathologies, it is crucial to account for the anatomical variability in order to translate TBS from a research tool into a viable option for treatment. In this article we present a new method, called CHARM, for automated segmentation of fifteen different head tissues from magnetic resonance (MR) scans. The new method compares favorably to two freely available software tools on a five-tissue segmentation task, while obtaining reasonable segmentation accuracy over all fifteen tissues. The method automatically adapts to variability in the input scans and can thus be directly applied to clinical or research scans acquired with different scanners, sequences or settings. We show that an increase in automated segmentation accuracy results in a lower relative error in electric field simulations when compared to anatomical head models constructed from reference segmentations. However, also the improved segmentations and, by implication, the electric field simulations are affected by systematic artifacts in the input MR scans. As long as the artifacts are unaccounted for, this can lead to local simulation differences up to 30% of the peak field strength on reference simulations. Finally, we exemplarily demonstrate the effect of including all fifteen tissue classes in the field simulations against the standard approach of using only five tissue classes and show that for specific stimulation configurations the local differences can reach 10% of the peak field strength.

IOUC-3DSFCNN: Segmentation of Brain Tumors via IOU Constraint 3D Symmetric Full Convolution Network with Multimodal Auto-context

Accurate segmentation of brain tumors from magnetic resonance (MR) images play a pivot role in assisting diagnoses, treatments and postoperative evaluations. However, due to its structural complexities, e.g., fuzzy tumor boundaries with irregular shapes, accurate 3D brain tumor delineation is challenging. In this paper, an intersection over union (IOU) constraint 3D symmetric full convolutional neural network (IOUC-3DSFCNN) model fused with multimodal auto-context is proposed for the 3D brain tumor segmentation. IOUC-3DSFCNN incorporates 3D residual groups into the classic 3DU-Net to further deepen the network structure to obtain more abstract voxel features under a five-layer cohesion architecture to ensure the model stability. The IOU constraint is used to address the issue of extremely unbalanced tumor foreground and background regions in MR images. In addition, to obtain more comprehensive and stable 3D brain tumor profiles, the multimodal auto-context information is fused into the IOUC-3DSFCNN model to achieve end-to-end 3D brain tumor profiles. Extensive confirmatory and comparative experiments conducted on the benchmark BRATS 2017 dataset demonstrate that the proposed segmentation model is superior to classic 3DU-Net-relevant and other state-of-the-art segmentation models, which can achieve accurate 3D tumor profiles on multimodal MRI volumes even with blurred tumor boundaries and big noise.

Automated postoperative muscle assessment of hip arthroplasty patients using multimodal imaging joint segmentation

BACKGROUND AND OBJECTIVE

In patients treated with hip arthroplasty, the muscular condition and presence of inflammatory reactions are assessed using magnetic resonance imaging (MRI). As MRI lacks contrast for bony structures, computed tomography (CT) is preferred for clinical evaluation of bone tissue and orthopaedic surgical planning. Combining the complementary information of MRI and CT could improve current clinical practice for diagnosis, monitoring and treatment planning. In particular, the different contrast of these modalities could help better quantify the presence of fatty infiltration to characterise muscular condition and assess implant failure. In this work, we combine CT and MRI for joint bone and muscle segmentation and we propose a novel Intramuscular Fat Fraction estimation method for the quantification of muscle atrophy.

METHODS

Our multimodal framework is able to segment healthy and pathological musculoskeletal structures as well as implants, and develops into three steps. First, input images are pre-processed to improve the low quality of clinically acquired images and to reduce the noise associated with metal artefact. Subsequently, CT and MRI are non-linearly aligned using a novel approach which imposes rigidity constraints on bony structures to ensure realistic deformation. Finally, taking advantage of a multimodal atlas we created for this task, a multi-atlas based segmentation delineates pelvic bones, abductor muscles and implants on both modalities jointly. From the obtained segmentation, a multimodal estimation of the Intramuscular Fat Fraction can be automatically derived.

RESULTS

Evaluation of the segmentation in a leave-one-out cross-validation study on 22 hip sides resulted in an average Dice score of 0.90 for skeletal and 0.84 for muscular structures. Our multimodal Intramuscular Fat Fraction was benchmarked on 27 different cases against a standard radiological score, showing stronger association than a single modality approach in a one-way ANOVA F-test analysis.

CONCLUSIONS

The proposed framework represents a promising tool to support image analysis in hip arthroplasty, being robust to the presence of implants and associated image artefacts. By allowing for the automated extraction of a muscle atrophy imaging biomarker, it could quantitatively inform the decision-making process about patient's management.

Quantitative assessment of myelination patterns in preterm neonates using T2-weighted MRI

Myelination is considered to be an important developmental process during human brain maturation and closely correlated with gestational age. Quantitative assessment of the myelination status requires dedicated imaging, but the conventional T₂-weighted scans routinely acquired during clinical imaging of neonates carry signatures that are thought to be associated with myelination. In this work, we develop a quatitative marker of progressing myelination for assessment preterm neonatal brain maturation based on novel automatic segmentation method for myelin-like signals on T₂-weighted magnetic resonance images. Firstly we define a segmentation protocol for myelin-like signals. We then develop an expectation-maximization framework to obtain the automatic segmentations of myelin-like signals with explicit class for partial volume voxels whose locations are configured in relation to the composing pure tissues via second-order Markov random fields. The proposed segmentation achieves high Dice overlaps of 0.83 with manual annotations. The automatic segmentations are then used to track volumes of myelinated tissues in the regions of the central brain structures and brainstem. Finally, we construct a spatio-temporal growth models for myelin-like signals, which allows us to predict gestational age at scan in preterm infants with root mean squared error 1.41 weeks.

Ten Years of BrainAGE as a Neuroimaging Biomarker of Brain Aging: What Insights Have We Gained?

With the aging population, prevalence of neurodegenerative diseases is increasing, thus placing a growing burden on individuals and the whole society. However, individual rates of aging are shaped by a great variety of and the interactions between environmental, genetic, and epigenetic factors. Establishing biomarkers of the neuroanatomical aging processes exemplifies a new trend in neuroscience in order to provide risk-assessments and predictions for age-associated neurodegenerative and neuropsychiatric diseases at a single-subject level. The "Brain Age Gap Estimation (BrainAGE)" method constitutes the first and actually most widely applied concept for predicting and evaluating individual brain age based on structural MRI. This review summarizes all studies published within the last 10 years that have established and utilized the BrainAGE method to evaluate the effects of interaction of genes, environment, life burden, diseases, or life time on individual neuroanatomical aging. In future, BrainAGE and other brain age prediction approaches based on structural or functional markers may improve the assessment of individual risks for neurological, neuropsychiatric and neurodegenerative diseases as well as aid in developing personalized neuroprotective treatments and interventions.

Differences in functional connectivity along the anterior-posterior axis of human hippocampal subfields

There is a paucity of information about how human hippocampal subfields are functionally connected to each other and to neighbouring extra-hippocampal cortices. In particular, little is known about whether patterns of functional connectivity (FC) differ down the anterior-posterior axis of each subfield. Here, using high resolution structural MRI we delineated the hippocampal subfields in healthy young adults. This included the CA fields, separating DG/CA4 from CA3, separating the pre/parasubiculum from the subiculum, and also segmenting the uncus. We then used high resolution resting state functional MRI to interrogate FC. We first analysed the FC of each hippocampal subfield in its entirety, in terms of FC with other subfields and with the neighbouring regions, namely entorhinal, perirhinal, posterior parahippocampal and retrosplenial cortices. Next, we analysed FC for different portions of each hippocampal subfield along its anterior-posterior axis, in terms of FC between different parts of a subfield, FC with other subfield portions, and FC of each subfield portion with the neighbouring cortical regions of interest. We found that intrinsic functional connectivity between the subfields aligned generally with the tri-synaptic circuit but also extended beyond it. Our findings also revealed that patterns of functional connectivity between the subfields and neighbouring cortical areas differed markedly along the anterior-posterior axis of each hippocampal subfield. Overall, these results contribute to ongoing efforts to characterise human hippocampal subfield connectivity, with implications for understanding hippocampal function.

Functional connectivity along the anterior-posterior axis of hippocampal subfields in the ageing human brain

While age‐related volumetric changes in human hippocampal subfields have been reported, little is known about patterns of subfield functional connectivity (FC) in the context of healthy ageing. Here we investigated age‐related changes in patterns of FC down the anterior–posterior axis of each subfield. Using high resolution structural MRI we delineated the dentate gyrus (DG), CA fields (including separating DG from CA3), the subiculum, pre/parasubiculum, and the uncus in healthy young and older adults. We then used high resolution resting state functional MRI to measure FC in each group and to directly compare them. We first examined the FC of each subfield in its entirety, in terms of FC with other subfields and with neighboring cortical regions, namely, entorhinal, perirhinal, posterior parahippocampal, and retrosplenial cortices. Next, we analyzed subfield to subfield FC within different portions along the hippocampal anterior–posterior axis, and FC of each subfield portion with the neighboring cortical regions of interest. In general, the FC of the older adults was similar to that observed in the younger adults. We found that, as in the young group, the older group displayed intrinsic FC between the subfields that aligned with the tri‐synaptic circuit but also extended beyond it, and that FC between the subfields and neighboring cortical areas differed markedly along the anterior–posterior axis of each subfield. We observed only one significant difference between the young and older groups. Compared to the young group, the older participants had significantly reduced FC between the anterior CA1‐subiculum transition region and the transentorhinal cortex, two brain regions known to be disproportionately affected during the early stages of age‐related tau accumulation. Overall, these results contribute to ongoing efforts to characterize human hippocampal subfield connectivity, with implications for understanding hippocampal function and its modulation in the ageing brain.

A modality-adaptive method for segmenting brain tumors and organs-at-risk in radiation therapy planning

In this paper we present a method for simultaneously segmenting brain tumors and an extensive set of organs-at-risk for radiation therapy planning of glioblastomas. The method combines a contrast-adaptive generative model for whole-brain segmentation with a new spatial regularization model of tumor shape using convolutional restricted Boltzmann machines. We demonstrate experimentally that the method is able to adapt to image acquisitions that differ substantially from any available training data, ensuring its applicability across treatment sites; that its tumor segmentation accuracy is comparable to that of the current state of the art; and that it captures most organs-at-risk sufficiently well for radiation therapy planning purposes. The proposed method may be a valuable step towards automating the delineation of brain tumors and organs-at-risk in glioblastoma patients undergoing radiation therapy.

A Spatial Registration Toolbox for Structural MR Imaging of the Aging Brain

During aging the brain undergoes a series of structural changes, in size, shape as well as tissue composition. In particular, cortical atrophy and ventricular enlargement are often present in the brain of elderly individuals. This poses serious challenges in the spatial registration of structural MR images. In this study, we addressed this open issue by proposing an enhanced framework for MR registration and segmentation. Our solution was compared with other approaches based on the tools available in SPM12, a widely used software package. Performance of the different methods was assessed on 229 T1-weighted images collected in healthy individuals, with age ranging between 55 and 90 years old. Our method showed a consistent improvement as compared to other solutions, especially for subjects with enlarged lateral ventricles. It also provided a superior inter-subject alignment in cortical regions, with the most marked improvement in the frontal lobe. We conclude that our method is a valid alternative to standard approaches based on SPM12, and is particularly suitable for the processing of structural MR images of brains with cortical atrophy and ventricular enlargement. The method is integrated in our software toolbox MRTool, which is freely available to the scientific community.

Automatic segmentation of the spinal cord and intramedullary multiple sclerosis lesions with convolutional neural networks

The spinal cord is frequently affected by atrophy and/or lesions in multiple sclerosis (MS) patients. Segmentation of the spinal cord and lesions from MRI data provides measures of damage, which are key criteria for the diagnosis, prognosis, and longitudinal monitoring in MS. Automating this operation eliminates inter-rater variability and increases the efficiency of large-throughput analysis pipelines. Robust and reliable segmentation across multi-site spinal cord data is challenging because of the large variability related to acquisition parameters and image artifacts. In particular, a precise delineation of lesions is hindered by a broad heterogeneity of lesion contrast, size, location, and shape. The goal of this study was to develop a fully-automatic framework - robust to variability in both image parameters and clinical condition - for segmentation of the spinal cord and intramedullary MS lesions from conventional MRI data of MS and non-MS cases. Scans of 1042 subjects (459 healthy controls, 471 MS patients, and 112 with other spinal pathologies) were included in this multi-site study (n = 30). Data spanned three contrasts (T1-, T2-, and T2∗-weighted) for a total of 1943 vol and featured large heterogeneity in terms of resolution, orientation, coverage, and clinical conditions. The proposed cord and lesion automatic segmentation approach is based on a sequence of two Convolutional Neural Networks (CNNs). To deal with the very small proportion of spinal cord and/or lesion voxels compared to the rest of the volume, a first CNN with 2D dilated convolutions detects the spinal cord centerline, followed by a second CNN with 3D convolutions that segments the spinal cord and/or lesions. CNNs were trained independently with the Dice loss. When compared against manual segmentation, our CNN-based approach showed a median Dice of 95% vs. 88% for PropSeg (p ≤ 0.05), a state-of-the-art spinal cord segmentation method. Regarding lesion segmentation on MS data, our framework provided a Dice of 60%, a relative volume difference of -15%, and a lesion-wise detection sensitivity and precision of 83% and 77%, respectively. In this study, we introduce a robust method to segment the spinal cord and intramedullary MS lesions on a variety of MRI contrasts. The proposed framework is open-source and readily available in the Spinal Cord Toolbox.

White matter microstructural correlates of relapse in alcohol dependence

Identification of neural correlates of relapse to alcohol after treatment is clinically important as it may inform better substance abuse treatment. Few studies have specifically analyzed the white matter microstructure in treatment seekers as it might relate to relapse risk versus long-term abstinence. Using 4 Tesla diffusion tensor imaging, we compared two groups of one-month-abstinent treatment-seekers, who were classified based on their drinking status between six and nine months after treatment initiation. We hypothesized that subsequent relapsers had greater white matter microstructural deficits in specific brain regions than long-term abstainers. At one month of abstinence, 37 future relapsers versus 25 future abstainers had lower fractional anisotropy (a measure of axonal organization and membrane integrity) in the corpus callosum and right stria terminalis/fornix, higher diffusivity in the genu of the corpus callosum, left and right stria terminalis/fornix, and lower diffusivity in left anterior corona radiata. These differences existed despite similar lifetime and recent drinking and smoking histories in the groups. Longer smoking duration in relapsers was associated with lower fractional anisotropy in right stria terminalis/fornix. The study identified specific microstructural biomarkers of alcohol relapse risk in adults, contributing to the definition of a neurobiological relapse risk profile in alcohol use disorder.

Processing of structural neuroimaging data in young children: Bridging the gap between current practice and state-of-the-art methods

The structure of the brain is subject to very rapid developmental changes during early childhood. Pediatric studies based on Magnetic Resonance Imaging (MRI) over this age range have recently become more frequent, with the advantage of providing in vivo and non-invasive high-resolution images of the developing brain, toward understanding typical and atypical trajectories. However, it has also been demonstrated that application of currently standard MRI processing methods that have been developed with datasets from adults may not be appropriate for use with pediatric datasets. In this review, we examine the approaches currently used in MRI studies involving young children, including an overview of the rationale for new MRI processing methods that have been designed specifically for pediatric investigations. These methods are mainly related to the use of age-specific or 4D brain atlases, improved methods for quantifying and optimizing image quality, and provision for registration of developmental data obtained with longitudinal designs. The overall goal is to raise awareness of the existence of these methods and the possibilities for implementing them in developmental neuroimaging studies.

Joint registration and synthesis using a probabilistic model for alignment of MRI and histological sections

Nonlinear registration of 2D histological sections with corresponding slices of MRI data is a critical step of 3D histology reconstruction algorithms. This registration is difficult due to the large differences in image contrast and resolution, as well as the complex nonrigid deformations and artefacts produced when sectioning the sample and mounting it on the glass slide. It has been shown in brain MRI registration that better spatial alignment across modalities can be obtained by synthesising one modality from the other and then using intra-modality registration metrics, rather than by using information theory based metrics to solve the problem directly. However, such an approach typically requires a database of aligned images from the two modalities, which is very difficult to obtain for histology and MRI.

Here, we overcome this limitation with a probabilistic method that simultaneously solves for deformable registration and synthesis directly on the target images, without requiring any training data. The method is based on a probabilistic model in which the MRI slice is assumed to be a contrast-warped, spatially deformed version of the histological section. We use approximate Bayesian inference to iteratively refine the probabilistic estimate of the synthesis and the registration, while accounting for each other’s uncertainty. Moreover, manually placed landmarks can be seamlessly integrated in the framework for increased performance and robustness.

Experiments on a synthetic dataset of MRI slices show that, compared with mutual information based registration, the proposed method makes it possible to use a much more flexible deformation model in the registration to improve its accuracy, without compromising robustness. Moreover, our framework also exploits information in manually placed landmarks more efficiently than mutual information: landmarks constrain the deformation field in both methods, but in our algorithm, it also has a positive effect on the synthesis – which further improves the registration. We also show results on two real, publicly available datasets: the Allen and BigBrain atlases. In both of them, the proposed method provides a clear improvement over mutual information based registration, both qualitatively (visual inspection) and quantitatively (registration error measured with pairs of manually annotated landmarks).

Differentiable Processing of Objects, Associations, and Scenes within the Hippocampus

The hippocampus is known to be important for a range of cognitive functions, including episodic memory, spatial navigation, and thinking about the future. However, researchers have found it difficult to agree on the exact nature of this brain structure's contribution to cognition. Some theories emphasize the role of the hippocampus in associative processes. Another theory proposes that scene construction is its primary role. To directly compare these accounts of hippocampal function in human males and females, we devised a novel mental imagery paradigm where different tasks were closely matched for associative processing and mental construction, but either did or did not evoke scene representations, and we combined this with high-resolution functional MRI. The results were striking in showing that different parts of the hippocampus, along with distinct cortical regions, were recruited for scene construction or nonscene-evoking associative processing. The contrasting patterns of neural engagement could not be accounted for by differences in eye movements, mnemonic processing, or the phenomenology of mental imagery. These results inform conceptual debates in the field by showing that the hippocampus does not seem to favor one type of process over another; it is not a story of exclusivity. Rather, there may be different circuits within the hippocampus, each associated with different cortical inputs, which become engaged depending on the nature of the stimuli and the task at hand. Overall, our findings emphasize the importance of considering the hippocampus as a heterogeneous structure, and that a focus on characterizing how specific portions of the hippocampus interact with other brain regions may promote a better understanding of its role in cognition.SIGNIFICANCE STATEMENT The hippocampus is known to be important for a range of cognitive functions, including episodic memory, spatial navigation, and thinking about the future. However, researchers have found it difficult to agree on the exact nature of this brain structure's contribution to cognition. Here we used a novel mental imagery paradigm and high-resolution functional MRI to compare accounts of hippocampal function that emphasize associative processes with a theory that proposes scene construction as a primary role. The results were striking in showing that different parts of the hippocampus, along with distinct cortical regions, were recruited for scene construction or nonscene-evoking associative processing. We conclude that a greater emphasis on characterizing how specific portions of the hippocampus interact with other brain regions may promote a better understanding of its role in cognition.

Efficient segmentation and correction model for brain MR images with level set framework based on basis functions

With the wide application of MR images to detect disease in human's brain deeply, the shortcomings of the technology are necessarily waiting to be solved. For example, MR images always show serious intensity inhomogeneity called the bias field, which may prevent to deduce exact analysis of images. To eliminate the distraction, many methods are proposed. Though experimental results already have stood for the advantages of those methods, there are still lots of problems that cannot be neglected, such as bad segmentation, wrong correction and over-correction which has not attracted much attention yet. Among all those methods, the multiplicative intrinsic component optimization (MICO) model influenced us more. Based on the MICO model and split Bregman method, in this paper, we put forward a new model to segment and correct bias field moderately and simultaneously for MR images. Then, we applied our model to a large quantity of MR images, and gained lots of expected results. For a better observation, we compared our model with the MICO model in both segmentation and bias correction results, it can be seen from the experimental results that our model has performed well for the challenging intensity inhomogeneity problems. Many good characteristics like accuracy, efficiency and robustness also have been exhibited in numerical results and comparisons with the MICO model.

Optimal multi-object segmentation with novel gradient vector flow based shape priors

Shape priors have been widely utilized in medical image segmentation to improve segmentation accuracy and robustness. A major way to encode such a prior shape model is to use a mesh representation, which is prone to causing self-intersection or mesh folding. Those problems require complex and expensive algorithms to mitigate. In this paper, we propose a novel shape prior directly embedded in the voxel grid space, based on gradient vector flows of a pre-segmentation. The flexible and powerful prior shape representation is ready to be extended to simultaneously segmenting multiple interacting objects with minimum separation distance constraint. The segmentation problem of multiple interacting objects with shape priors is formulated as a Markov Random Field problem, which seeks to optimize the label assignment (objects or background) for each voxel while keeping the label consistency between the neighboring voxels. The optimization problem can be efficiently solved with a single minimum s-t cut in an appropriately constructed graph. The proposed algorithm has been validated on two multi-object segmentation applications: the brain tissue segmentation in MRI images and the bladder/prostate segmentation in CT images. Both sets of experiments showed superior or competitive performance of the proposed method to the compared state-of-the-art methods.

Geometric and dosimetric evaluations of atlas-based segmentation methods of MR images in the head and neck region

Owing to its excellent soft-tissue contrast, magnetic resonance (MR) imaging has found an increased application in radiation therapy (RT). By harnessing these properties for treatment planning, automated segmentation methods can alleviate the manual workload burden to the clinical workflow. We investigated atlas-based segmentation methods of organs at risk (OARs) in the head and neck (H&N) region using one approach that selected the most similar atlas from a library of segmented images and two multi-atlas approaches. The latter were based on weighted majority voting and an iterative atlas-fusion approach called STEPS. We built the atlas library from pre-treatment T1-weighted MR images of 12 patients with manual contours of the parotids, spinal cord and mandible, delineated by a clinician. Following a leave-one-out cross-validation strategy, we measured the geometric accuracy by calculating Dice similarity coefficients (DSC), standard and 95% Hausdorff distances (HD and HD95), and the mean surface distance (MSD), whereby the manual contours served as the gold standard. To benchmark the algorithm, we determined the inter-observer variability (IOV) between three observers. To investigate the dosimetric effect of segmentation inaccuracies, we implemented an auto-planning strategy within the treatment planning system Monaco (Elekta AB, Stockholm, Sweden). For each set of auto-segmented OARs, we generated a plan for a 9-beam step and shoot intensity modulated RT treatment, designed according to our institution's clinical H&N protocol. Superimposing the dose distributions on the gold standard OARs, we calculated dose differences to OARs caused by delineation differences between auto-segmented and gold standard OARs. We investigated the correlations between geometric and dosimetric differences. The mean DSC was larger than 0.8 and the mean MSD smaller than 2 mm for the multi-atlas approaches, resulting in a geometric accuracy comparable to previously published results and within the range of the IOV. While dosimetric differences could be as large as 23% of the clinical goal, treatment plans fulfilled all imposed clinical goals for the gold standard OARs. Correlations between geometric and dosimetric measures were low with R2  <  0.5. The geometric accuracy and the ability to achieve clinically acceptable treatment plans indicate the suitability of using atlas-based contours for RT treatment planning purposes. The low correlations between geometric and dosimetric measures suggest that geometric measures alone are not sufficient to predict the dosimetric impact of segmentation inaccuracies on treatment planning for the data utilised in this study.

Geometric and dosimetric evaluations of atlas-based segmentation methods of MR images in the head and neck region

Owing to its excellent soft-tissue contrast, magnetic resonance (MR) imaging has found an increased application in radiation therapy (RT). By harnessing these properties for treatment planning, automated segmentation methods can alleviate the manual workload burden to the clinical workflow. We investigated atlas-based segmentation methods of organs at risk (OARs) in the head and neck (H&N) region using one approach that selected the most similar atlas from a library of segmented images and two multi-atlas approaches. The latter were based on weighted majority voting and an iterative atlas-fusion approach called STEPS. We built the atlas library from pre-treatment T1-weighted MR images of 12 patients with manual contours of the parotids, spinal cord and mandible, delineated by a clinician. Following a leave-one-out cross-validation strategy, we measured the geometric accuracy by calculating Dice similarity coefficients (DSC), standard and 95% Hausdorff distances (HD and HD95), and the mean surface distance (MSD), whereby the manual contours served as the gold standard. To benchmark the algorithm, we determined the inter-observer variability (IOV) between three observers. To investigate the dosimetric effect of segmentation inaccuracies, we implemented an auto-planning strategy within the treatment planning system Monaco (Elekta AB, Stockholm, Sweden). For each set of auto-segmented OARs, we generated a plan for a 9-beam step and shoot intensity modulated RT treatment, designed according to our institution's clinical H&N protocol. Superimposing the dose distributions on the gold standard OARs, we calculated dose differences to OARs caused by delineation differences between auto-segmented and gold standard OARs. We investigated the correlations between geometric and dosimetric differences. The mean DSC was larger than 0.8 and the mean MSD smaller than 2 mm for the multi-atlas approaches, resulting in a geometric accuracy comparable to previously published results and within the range of the IOV. While dosimetric differences could be as large as 23% of the clinical goal, treatment plans fulfilled all imposed clinical goals for the gold standard OARs. Correlations between geometric and dosimetric measures were low with R²  <  0.5. The geometric accuracy and the ability to achieve clinically acceptable treatment plans indicate the suitability of using atlas-based contours for RT treatment planning purposes. The low correlations between geometric and dosimetric measures suggest that geometric measures alone are not sufficient to predict the dosimetric impact of segmentation inaccuracies on treatment planning for the data utilised in this study.

Evaluation of methods for volumetric analysis of pediatric brain data: The child pipeline versus adult-based approaches

Pediatric brain volumetric analysis based on Magnetic Resonance Imaging (MRI) is of particular interest in order to understand the typical brain development and to characterize neurodevelopmental disorders at an early age. However, it has been shown that the results can be biased due to head motion, inherent to pediatric data, and due to the use of methods based on adult brain data that are not able to accurately model the anatomical disparity of pediatric brains. To overcome these issues, we proposed childmetrix, a tool developed for the analysis of pediatric neuroimaging data that uses an age-specific atlas and a probabilistic model-based approach in order to segment the gray matter (GM) and white matter (WM). The tool was extensively validated on 55 scans of children between 5 and 6 years old (including 13 children with developmental dyslexia) and 10 pairs of test-retest scans of children between 6 and 8 years old and compared with two state-of-the-art methods using an adult atlas, namely icobrain (applying a probabilistic model-based segmentation) and Freesurfer (applying a surface model-based segmentation). The results obtained with childmetrix showed a better reproducibility of GM and WM segmentations and a better robustness to head motion in the estimation of GM volume compared to Freesurfer. Evaluated on two subjects, childmetrix showed good accuracy with 82-84% overlap with manual segmentation for both GM and WM, thereby outperforming the adult-based methods (icobrain and Freesurfer), especially for the subject with poor quality data. We also demonstrated that the adult-based methods needed double the number of subjects to detect significant morphological differences between dyslexics and typical readers. Once further developed and validated, we believe that childmetrix would provide appropriate and reliable measures for the examination of children's brain.

Brain tumor image segmentation using kernel dictionary learning

Automated brain tumor image segmentation with high accuracy and reproducibility holds a big potential to enhance the current clinical practice. Dictionary learning (DL) techniques have been applied successfully to various image processing tasks recently. In this work, kernel extensions of the DL approach are adopted. Both reconstructive and discriminative versions of the kernel DL technique are considered, which can efficiently incorporate multi-modal nonlinear feature mappings based on the kernel trick. Our novel discriminative kernel DL formulation allows joint learning of a task-driven kernel-based dictionary and a linear classifier using a K-SVD-type algorithm. The proposed approaches were tested using real brain magnetic resonance (MR) images of patients with high-grade glioma. The obtained preliminary performances are competitive with the state of the art. The discriminative kernel DL approach is seen to reduce computational burden without much sacrifice in performance.

From early stress to 12-month development in very preterm infants: Preliminary findings on epigenetic mechanisms and brain growth

Very preterm (VPT) infants admitted to Neonatal Intensive Care Unit (NICU) are at risk for altered brain growth and less-than-optimal socio-emotional development. Recent research suggests that early NICU-related stress contributes to socio-emotional impairments in VPT infants at 3 months through epigenetic regulation (i.e., DNA methylation) of the serotonin transporter gene (SLC6A4). In the present longitudinal study we assessed: (a) the effects of NICU-related stress and SLC6A4 methylation variations from birth to discharge on brain development at term equivalent age (TEA); (b) the association between brain volume at TEA and socio-emotional development (i.e., Personal-Social scale of Griffith Mental Development Scales, GMDS) at 12 months corrected age (CA). Twenty-four infants had complete data at 12-month-age. SLC6A4 methylation was measured at a specific CpG previously associated with NICU-related stress and socio-emotional stress. Findings confirmed that higher NICU-related stress associated with greater increase of SLC6A4 methylation at NICU discharge. Moreover, higher SLC6A4 discharge methylation was associated with reduced anterior temporal lobe (ATL) volume at TEA, which in turn was significantly associated with less-than-optimal GMDS Personal-Social scale score at 12 months CA. The reduced ATL volume at TEA mediated the pathway linking stress-related increase in SLC6A4 methylation at NICU discharge and socio-emotional development at 12 months CA. These findings suggest that early adversity-related epigenetic changes might contribute to the long-lasting programming of socio-emotional development in VPT infants through epigenetic regulation and structural modifications of the developing brain.

Brain GABA and Glutamate Concentrations Following Chronic Gabapentin Administration: A Convenience Sample Studied During Early Abstinence From Alcohol

Gabapentin (GBP), a GABA analog that may also affect glutamate (Glu) production, can normalize GABA and Glu tone during early abstinence from alcohol, effectively treating withdrawal symptoms and facilitating recovery. Using in vivo magnetic resonance spectroscopy, we tested the degree to which daily GBP alters regional brain GABA and Glu levels in short-term abstinent alcohol-dependent individuals. Regional metabolite levels were compared between 13 recently abstinent alcohol-dependent individuals who had received daily GBP for at least 1 week (GBP+) and 25 matched alcohol-dependent individuals who had not received GBP (GBP-). Magnetic resonance spectra from up to five different brain regions were analyzed to yield absolute GABA and Glu concentrations. GABA and Glu concentrations in the parieto-occipital cortex were not different between GBP- and GBP+. Glu levels in anterior cingulate cortex, dorsolateral prefrontal cortex, and basal ganglia did not differ between GBP- and GBP+. However, in a subgroup of individuals matched on age, sex, and abstinence duration, GBP+ had markedly lower Glu in the frontal white matter (WM) than GBP-, comparable to concentrations found in light/non-drinking controls. Furthermore, lower frontal WM Glu in GBP+ correlated with a higher daily GBP dose. Daily GBP treatment at an average of 1,600 mg/day for at least 1 week was not associated with altered cortical GABA and Glu concentrations during short-term abstinence from alcohol, but with lower Glu in frontal WM. GBP for the treatment of alcohol dependence may work through reducing Glu in WM rather than increasing cortical GABA.

Semisupervised learning using denoising autoencoders for brain lesion detection and segmentation

The work explores the use of denoising autoencoders (DAEs) for brain lesion detection, segmentation, and false-positive reduction. Stacked denoising autoencoders (SDAEs) were pretrained using a large number of unlabeled patient volumes and fine-tuned with patches drawn from a limited number of patients ([Formula: see text], 40, 65). The results show negligible loss in performance even when SDAE was fine-tuned using 20 labeled patients. Low grade glioma (LGG) segmentation was achieved using a transfer learning approach in which a network pretrained with high grade glioma data was fine-tuned using LGG image patches. The networks were also shown to generalize well and provide good segmentation on unseen BraTS 2013 and BraTS 2015 test data. The manuscript also includes the use of a single layer DAE, referred to as novelty detector (ND). ND was trained to accurately reconstruct nonlesion patches. The reconstruction error maps of test data were used to localize lesions. The error maps were shown to assign unique error distributions to various constituents of the glioma, enabling localization. The ND learns the nonlesion brain accurately as it was also shown to provide good segmentation performance on ischemic brain lesions in images from a different database.