Robust, atlas-free, automatic segmentation of brain MRI in health and disease

White and Gray Matter Changes are Associated With Neurocognitive Decline in HIV Infection

OBJECTIVE

To investigate the relationship between neurocognitive deficits and structural changes on brain magnetic resonance imaging in people living with HIV (PLWH) with good virological control on combination antiretroviral therapy, compared with socioeconomically matched control participants recruited from the same communities.

METHODS

Brain magnetic resonance imaging scans, and clinical and neuropsychological data were obtained from virologically controlled PLWH (viral load of <50 c/mL and at least 1 year of combination antiretroviral therapy) and socioeconomically matched control participants. Magnetic resonance imaging was carried out on 3 T scanner with 8-channel head coils and segmented using Classification using Derivative-based Features. Multiple regression analysis was performed to examine the association between brain volume and various clinical and neuropsychiatric parameters adjusting for age, race, and sex. To evaluate longitudinal changes in brain volumes, a random coefficient model was used to evaluate the changes over time (age) adjusting for sex and race.

RESULTS

The cross-sectional study included 164 PLWH and 51 controls, and the longitudinal study included 68 PLWH and 20 controls with 2 or more visits (mean 2.2 years, range 0.8-5.1 years). Gray matter (GM) atrophy rate was significantly higher in PLWH compared with control participants, and importantly, the GM and global atrophy was associated with the various neuropsychological domain scores. Higher volume of white matter hyperintensities were associated with increased atherosclerotic cardiovascular disease risk score, and decreased executive functioning and memory domain scores in PLWH.

INTERPRETATION

These findings suggest ongoing neurological damage even in virologically controlled participants, with significant implications for clinical management of PLWH. ANN NEUROL 2024;95:941-950.

Subject-Specific Automatic Reconstruction of White Matter Tracts

MRI-based tractography is still underexploited and unsuited for routine use in brain tumor surgery due to heterogeneity of methods and functional-anatomical definitions and above all, the lack of a turn-key system. Standardization of methods is therefore desirable, whereby an objective and reliable approach is a prerequisite before the results of any automated procedure can subsequently be validated and used in neurosurgical practice. In this work, we evaluated these preliminary but necessary steps in healthy volunteers. Specifically, we evaluated the robustness and reliability (i.e., test-retest reproducibility) of tractography results of six clinically relevant white matter tracts by using healthy volunteer data (N = 136) from the Human Connectome Project consortium. A deep learning convolutional network-based approach was used for individualized segmentation of regions of interest, combined with an evidence-based tractography protocol and appropriate post-tractography filtering. Robustness was evaluated by estimating the consistency of tractography probability maps, i.e., averaged tractograms in normalized space, through the use of a hold-out cross-validation approach. No major outliers were found, indicating a high robustness of the tractography results. Reliability was evaluated at the individual level. First by examining the overlap of tractograms that resulted from repeatedly processed identical MRI scans (N = 10, 10 iterations) to establish an upper limit of reliability of the pipeline. Second, by examining the overlap for subjects that were scanned twice at different time points (N = 40). Both analyses indicated high reliability, with the second analysis showing a reliability near the upper limit. The robust and reliable subject-specific generation of white matter tracts in healthy subjects holds promise for future validation of our pipeline in a clinical population and subsequent implementation in brain tumor surgery.

Pseudo-Label Assisted nnU-Net enables automatic segmentation of 7T MRI from a single acquisition

Introduction

Automatic whole brain and lesion segmentation at 7T presents challenges, primarily from bias fields, susceptibility artifacts including distortions, and registration errors. Here, we sought to use deep learning algorithms (D/L) to do both skull stripping and whole brain segmentation on multiple imaging contrasts generated in a single Magnetization Prepared 2 Rapid Acquisition Gradient Echoes (MP2RAGE) acquisition on participants clinically diagnosed with multiple sclerosis (MS), bypassing registration errors.

Methods

Brain scans Segmentation from 3T and 7T scanners were analyzed with software packages such as FreeSurfer, Classification using Derivative-based Features (C-DEF), nnU-net, and a novel 3T-to-7T transfer learning method, Pseudo-Label Assisted nnU-Net (PLAn). 3T and 7T MRIs acquired within 9 months from 25 study participants with MS (Cohort 1) were used for training and optimizing. Eight MS patients (Cohort 2) scanned only at 7T, but with expert annotated lesion segmentation, was used to further validate the algorithm on a completely unseen dataset. Segmentation results were rated visually by experts in a blinded fashion and quantitatively using Dice Similarity Coefficient (DSC).

Results

Of the methods explored here, nnU-Net and PLAn produced the best tissue segmentation at 7T for all tissue classes. In both quantitative and qualitative analysis, PLAn significantly outperformed nnU-Net (and other methods) in lesion detection in both cohorts. PLAn's lesion DSC improved by 16% compared to nnU-Net.

Discussion

Limited availability of labeled data makes transfer learning an attractive option, and pre-training a nnUNet model using readily obtained 3T pseudo-labels was shown to boost lesion detection capabilities at 7T.

Cortical lesions uniquely predict motor disability accrual and form rarely in the absence of new white matter lesions in multiple sclerosis

Background and objectives

Cortical lesions (CL) are common in multiple sclerosis (MS) and associate with disability and progressive disease. We asked whether CL continue to form in people with stable white matter lesions (WML) and whether the association of CL with worsening disability relates to pre-existing or new CL.

Methods

A cohort of adults with MS were evaluated annually with 7 tesla (T) brain magnetic resonance imaging (MRI) and 3T brain and spine MRI for 2 years, and clinical assessments for 3 years. CL were identified on 7T images at each timepoint. WML and brain tissue segmentation were performed using 3T images at baseline and year 2.

Results

59 adults with MS had ≥1 7T follow-up visit (mean follow-up time 2±0.5 years). 9 had "active" relapsing-remitting MS (RRMS), defined as new WML in the year prior to enrollment. Of the remaining 50, 33 had "stable" RRMS, 14 secondary progressive MS (SPMS), and 3 primary progressive MS. 16 total new CL formed in the active RRMS group (median 1, range 0-10), 7 in the stable RRMS group (median 0, range 0-5), and 4 in the progressive MS group (median 0, range 0-1) (p=0.006, stable RR vs PMS p=0.88). New CL were not associated with greater change in any individual disability measure or in a composite measure of disability worsening (worsening Expanded Disability Status Scale or 9-hole peg test or 25-foot timed walk). Baseline CL volume was higher in people with worsening disability (median 1010μl, range 13-9888 vs median 267μl, range 0-3539, p=0.001, adjusted for age and sex) and in individuals with RRMS who subsequently transitioned to SPMS (median 2183μl, range 270-9888 vs median 321μl, range 0-6392 in those who remained RRMS, p=0.01, adjusted for age and sex). Baseline WML volume was not associated with worsening disability or transition from RRMS to SPMS.

Discussion

CL formation is rare in people with stable WML, even in those with worsening disability. CL but not WML burden predicts future worsening of disability, suggesting that the relationship between CL and disability progression is related to long-term effects of lesions that form in the earlier stages of disease, rather than to ongoing lesion formation.

Manual lesion segmentations for traumatic brain injury characterization

Traumatic brain injury (TBI) often results in heterogenous lesions that can be visualized through various neuroimaging techniques, such as magnetic resonance imaging (MRI). However, injury burden varies greatly between patients and structural deformations often impact usability of available analytic algorithms. Therefore, it is difficult to segment lesions automatically and accurately in TBI cohorts. Mislabeled lesions will ultimately lead to inaccurate findings regarding imaging biomarkers. Therefore, manual segmentation is currently considered the gold standard as this produces more accurate masks than existing automated algorithms. These masks can provide important lesion phenotype data including location, volume, and intensity, among others. There has been a recent push to investigate the correlation between these characteristics and the onset of post traumatic epilepsy (PTE), a disabling consequence of TBI. One motivation of the Epilepsy Bioinformatics Study for Antiepileptogenic Therapy (EpiBioS4Rx) is to identify reliable imaging biomarkers of PTE. Here, we report the protocol and importance of our manual segmentation process in patients with moderate-severe TBI enrolled in EpiBioS4Rx. Through these methods, we have generated a dataset of 127 validated lesion segmentation masks for TBI patients. These ground-truths can be used for robust PTE biomarker analyses, including optimization of multimodal MRI analysis via inclusion of lesioned tissue labels. Moreover, our protocol allows for analysis of the refinement process. Though tedious, the methods reported in this work are necessary to create reliable data for effective training of future machine-learning based lesion segmentation methods in TBI patients and subsequent PTE analyses.

Masseter muscle volume as a disease marker in adult-onset myotonic dystrophy type 1

The advent of clinical trials in myotonic dystrophy type 1 (DM1) necessitates the identification of reliable outcome measures to quantify different disease manifestations using minimal number of assessments. In this study, clinical correlations of mean masseter volume (mMV) were explored to evaluate its potential as a marker of muscle involvement in adult-onset DM1 patients. We utilised data from a preceding study, pertaining to 39 DM1 patients and 20 age-matched control participants. In this study participants had undergone MRI of the brain, completed various clinical outcome measures and had CTG repeats measured by small-pool PCR. Manual segmentation of masseter muscles was performed by a single rater to estimate mMV. The masseter muscle was atrophied in DM1 patients when compared to controls (p<0.001). Significant correlations were found between mMV and estimated progenitor allele length (p = 0.001), modal allele length (p = 0.003), disease duration (p = 0.009) and and the Muscle Impairment Rating Scale (p = 0.008). After correction for lean body mass, mMV was also inversely correlated with self-reported myotonia (p = 0.014). This study demonstrates that changes in mMV are sensitive in reflecting the underlying disease process. Quantitative MRI methods demonstrate that data concerning both central and peripheral disease could be acquired from MR brain imaging studies in DM1 patients.

Advances in human intracranial electroencephalography research, guidelines and good practices

Since the second-half of the twentieth century, intracranial electroencephalography (iEEG), including both electrocorticography (ECoG) and stereo-electroencephalography (sEEG), has provided an intimate view into the human brain. At the interface between fundamental research and the clinic, iEEG provides both high temporal resolution and high spatial specificity but comes with constraints, such as the individual's tailored sparsity of electrode sampling. Over the years, researchers in neuroscience developed their practices to make the most of the iEEG approach. Here we offer a critical review of iEEG research practices in a didactic framework for newcomers, as well addressing issues encountered by proficient researchers. The scope is threefold: (i) review common practices in iEEG research, (ii) suggest potential guidelines for working with iEEG data and answer frequently asked questions based on the most widespread practices, and (iii) based on current neurophysiological knowledge and methodologies, pave the way to good practice standards in iEEG research. The organization of this paper follows the steps of iEEG data processing. The first section contextualizes iEEG data collection. The second section focuses on localization of intracranial electrodes. The third section highlights the main pre-processing steps. The fourth section presents iEEG signal analysis methods. The fifth section discusses statistical approaches. The sixth section draws some unique perspectives on iEEG research. Finally, to ensure a consistent nomenclature throughout the manuscript and to align with other guidelines, e.g., Brain Imaging Data Structure (BIDS) and the OHBM Committee on Best Practices in Data Analysis and Sharing (COBIDAS), we provide a glossary to disambiguate terms related to iEEG research.

Logistic Regression-Based Model Is More Efficient Than U-Net Model for Reliable Whole Brain Magnetic Resonance Imaging Segmentation

OBJECTIVES

Automated whole brain segmentation from magnetic resonance images is of great interest for the development of clinically relevant volumetric markers for various neurological diseases. Although deep learning methods have demonstrated remarkable potential in this area, they may perform poorly in nonoptimal conditions, such as limited training data availability. Manual whole brain segmentation is an incredibly tedious process, so minimizing the data set size required for training segmentation algorithms may be of wide interest. The purpose of this study was to compare the performance of the prototypical deep learning segmentation architecture (U-Net) with a previously published atlas-free traditional machine learning method, Classification using Derivative-based Features (C-DEF) for whole brain segmentation, in the setting of limited training data.

MATERIALS AND METHODS

C-DEF and U-Net models were evaluated after training on manually curated data from 5, 10, and 15 participants in 2 research cohorts: (1) people living with clinically diagnosed HIV infection and (2) relapsing-remitting multiple sclerosis, each acquired at separate institutions, and between 5 and 295 participants' data using a large, publicly available, and annotated data set of glioblastoma and lower grade glioma (brain tumor segmentation). Statistics was performed on the Dice similarity coefficient using repeated-measures analysis of variance and Dunnett-Hsu pairwise comparison.

RESULTS

C-DEF produced better segmentation than U-Net in lesion (29.2%-38.9%) and cerebrospinal fluid (5.3%-11.9%) classes when trained with data from 15 or fewer participants. Unlike C-DEF, U-Net showed significant improvement when increasing the size of the training data (24%-30% higher than baseline). In the brain tumor segmentation data set, C-DEF produced equivalent or better segmentations than U-Net for enhancing tumor and peritumoral edema regions across all training data sizes explored. However, U-Net was more effective than C-DEF for segmentation of necrotic/non-enhancing tumor when trained on 10 or more participants, probably because of the inconsistent signal intensity of the tissue class.

CONCLUSIONS

These results demonstrate that classical machine learning methods can produce more accurate brain segmentation than the far more complex deep learning methods when only small or moderate amounts of training data are available (n ≤ 15). The magnitude of this advantage varies by tissue and cohort, while U-Net may be preferable for deep gray matter and necrotic/non-enhancing tumor segmentation, particularly with larger training data sets (n ≥ 20). Given that segmentation models often need to be retrained for application to novel imaging protocols or pathology, the bottleneck associated with large-scale manual annotation could be avoided with classical machine learning algorithms, such as C-DEF.

A Macaque Brain Extraction Model Based on U-Net Combined with Residual Structure

Accurately extracting brain tissue is a critical and primary step in brain neuroimaging research. Due to the differences in brain size and structure between humans and nonhuman primates, the performance of the existing tools for brain tissue extraction, working on macaque brain MRI, is constrained. A new transfer learning training strategy was utilized to address the limitations, such as insufficient training data and unsatisfactory model generalization ability, when deep neural networks processing the limited samples of macaque magnetic resonance imaging(MRI). First, the project combines two human brain MRI data modes to pre-train the neural network, in order to achieve faster training and more accurate brain extraction. Then, a residual network structure in the U-Net model was added, in order to propose a ResTLU-Net model that aims to improve the generalization ability of multiple research sites data. The results demonstrated that the ResTLU-Net, combined with the proposed transfer learning strategy, achieved comparable accuracy for the macaque brain MRI extraction tasks on different macaque brain MRI volumes that were produced by various medical centers. The mean Dice of the ResTLU-Net was 95.81% (no need for denoise and recorrect), and the method required only approximately 30–60 s for one extraction task on an NVIDIA 1660S GPU.

Diagnosis of Multiple Sclerosis Disease in Brain Magnetic Resonance Imaging Based on the Harris Hawks Optimization Algorithm

The damaged areas of brain tissues can be extracted by using segmentation methods, most of which are based on the integration of machine learning and data mining techniques. An important segmentation method is to utilize clustering techniques, especially the fuzzy C-means (FCM) clustering technique, which is sufficiently accurate and not overly sensitive to imaging noise. Therefore, the FCM technique is appropriate for multiple sclerosis diagnosis, although the optimal selection of cluster centers can affect segmentation. They are difficult to select because this is an NP-hard problem. In this study, the Harris Hawks optimization (HHO) algorithm was used for the optimal selection of cluster centers in segmentation and FCM algorithms. The HHO is more accurate than other conventional algorithms such as the genetic algorithm and particle swarm optimization. In the proposed method, every membership matrix is assumed as a hawk or an HHO member. The next step is to generate a population of hawks or membership matrices, the most optimal of which is selected to find the optimal cluster centers to decrease the multiple sclerosis clustering error. According to the tests conducted on a number of brain MRIs, the proposed method outperformed the FCM clustering and other techniques such as the k-NN algorithm, support vector machine, and hybrid data mining methods in accuracy.

Overestimation of grey matter atrophy in glioblastoma patients following radio(chemo)therapy

Objective

Brain atrophy has the potential to become a biomarker for severity of radiation-induced side-effects. Particularly brain tumour patients can show great MRI signal changes over time caused by e.g. oedema, tumour progress or necrosis. The goal of this study was to investigate if such changes affect the segmentation accuracy of normal appearing brain and thus influence longitudinal volumetric measurements.

Materials and methods

T1-weighted MR images of 52 glioblastoma patients with unilateral tumours acquired before and three months after the end of radio(chemo)therapy were analysed. GM and WM volumes in the contralateral hemisphere were compared between segmenting the whole brain (full) and the contralateral hemisphere only (cl) with SPM and FSL. Relative GM and WM volumes were compared using paired t tests and correlated with the corresponding mean dose in GM and WM, respectively.

Results

Mean GM atrophy was significantly higher for full segmentation compared to cl segmentation when using SPM (mean ± std: ΔV_GM,full = − 3.1% ± 3.7%, ΔV_GM,cl = − 1.6% ± 2.7%; p < 0.001, d = 0.62). GM atrophy was significantly correlated with the mean GM dose with the SPM cl segmentation (r = − 0.4, p = 0.004), FSL full segmentation (r = − 0.4, p = 0.004) and FSL cl segmentation (r = -0.35, p = 0.012) but not with the SPM full segmentation (r = − 0.23, p = 0.1).

Conclusions

For accurate normal tissue volume measurements in brain tumour patients using SPM, abnormal tissue needs to be masked prior to segmentation, however, this is not necessary when using FSL.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10334-021-00922-3.

Distinguishing type II focal cortical dysplasias from normal cortex: A novel normative modeling approach

OBJECTIVE

Focal cortical dysplasias (FCDs) are a common cause of apparently non-lesional drug-resistant focal epilepsy. Visual detection of subtle FCDs on MRI is clinically important and often challenging. In this study, we implement a set of 3D local image filters adapted from computer vision applications to characterize the appearance of normal cortex surrounding the gray-white junction. We create a normative model to serve as the basis for a novel multivariate constrained outlier approach to automated FCD detection.

METHODS

Standardized MPRAGE, T₂ and FLAIR MR images were obtained in 15 patients with radiologically or histologically diagnosed FCDs and 30 healthy volunteers. Multiscale 3D local image filters were computed for each MR contrast then sampled onto the gray-white junction surface. Using an iterative Gaussianization procedure, we created a normative model of cortical variability in healthy volunteers, allowing for identification of outlier regions and estimates of similarity in normal cortex and FCD lesions. We used a constrained outlier approach following local normalization to automatically detect FCD lesions based on projection onto the mean FCD feature vector.

RESULTS

FCDs as well as some normal cortical regions such as primary sensorimotor and paralimbic regions appear as outliers. Regions such as the paralimbic regions and the anterior insula have similar features to FCDs. Our constrained outlier approach allows for automated FCD detection with 80% sensitivity and 70% specificity.

SIGNIFICANCE

A normative model using multiscale local image filters can be used to describe the normal cortical variability. Although FCDs appear similar to some cortical regions such as the anterior insula and paralimbic cortices, they can be identified using a constrained outlier detection approach. Our method for detecting outliers and estimating similarity is generic and could be extended to identification of other types of lesions or atypical cortical areas.

Progressive multifocal leukoencephalopathy lesion and brain parenchymal segmentation from MRI using serial deep convolutional neural networks

•

PML has characteristic dynamic changes in brain and lesion volume on MRI.

•

JCnet is an automated method for brain atrophy and lesion segmentation in PML.

•

JCnet improves PML lesion segmentation accuracy compared to conventional methods.

•

JCnet can accurately track PML lesion changes over time.

Progressive multifocal leukoencephalopathy (PML) is a rare opportunistic brain infection caused by the JC virus and associated with substantial morbidity and mortality. Accurate MRI assessment of PML lesion burden and brain parenchymal atrophy is of decisive value in monitoring the disease course and response to therapy. However, there are currently no validated automatic methods for quantification of PML lesion burden or associated parenchymal volume loss. Furthermore, manual brain or lesion delineations can be tedious, require the use of valuable time resources by radiologists or trained experts, and are often subjective. In this work, we introduce JCnet (named after the causative viral agent), an end-to-end, fully automated method for brain parenchymal and lesion segmentation in PML using consecutive 3D patch-based convolutional neural networks. The network architecture consists of multi-view feature pyramid networks with hierarchical residual learning blocks containing embedded batch normalization and nonlinear activation functions. The feature maps across the bottom-up and top-down pathways of the feature pyramids are merged, and an output probability membership generated through convolutional pathways, thus rendering the method fully convolutional. Our results show that this approach outperforms and improves longitudinal consistency compared to conventional, state-of-the-art methods of healthy brain and multiple sclerosis lesion segmentation, utilized here as comparators given the lack of available methods validated for use in PML. The ability to produce robust and accurate automated measures of brain atrophy and lesion segmentation in PML is not only valuable clinically but holds promise toward including standardized quantitative MRI measures in clinical trials of targeted therapies. Code is available at: https://github.com/omarallouz/JCnet.

Semi-Automatic Signature-Based Segmentation Method for Quantification of Neuromelanin in Substantia Nigra

In Parkinson's disease (PD), there is a reduction of neuromelanin (NM) in the substantia nigra (SN). Manual quantification of the NM volume in the SN is unpractical and time-consuming; therefore, we aimed to quantify NM in the SN with a novel semi-automatic segmentation method. Twenty patients with PD and twelve healthy subjects (HC) were included in this study. T1-weighted spectral pre-saturation with inversion recovery (SPIR) images were acquired on a 3T scanner. Manual and semi-automatic atlas-free local statistics signature-based segmentations measured the surface and volume of SN, respectively. Midbrain volume (MV) was calculated to normalize the data. Receiver operating characteristic (ROC) analysis was performed to determine the sensitivity and specificity of both methods. PD patients had significantly lower SN mean surface (37.7 ± 8.0 vs. 56.9 ± 6.6 mm²) and volume (235.1 ± 45.4 vs. 382.9 ± 100.5 mm³) than HC. After normalization with MV, the difference remained significant. For surface, sensitivity and specificity were 91.7 and 95 percent, respectively. For volume, sensitivity and specificity were 91.7 and 90 percent, respectively. Manual and semi-automatic segmentation methods of the SN reliably distinguished between PD patients and HC. ROC analysis shows the high sensitivity and specificity of both methods.