FastSurferVINN: Building resolution-independence into deep learning segmentation methods-A solution for HighRes brain MRI

"Recon-all-clinical": Cortical surface reconstruction and analysis of heterogeneous clinical brain MRI

Surface-based analysis of the cerebral cortex is ubiquitous in human neuroimaging with MRI. It is crucial for tasks like cortical registration, parcellation, and thickness estimation. Traditionally, such analyses require high-resolution, isotropic scans with good gray-white matter contrast, typically a T1-weighted scan with 1 mm resolution. This requirement precludes application of these techniques to most MRI scans acquired for clinical purposes, since they are often anisotropic and lack the required T1-weighted contrast. To overcome this limitation and enable large-scale neuroimaging studies using vast amounts of existing clinical data, we introduce recon-all-clinical, a novel methodology for cortical reconstruction, registration, parcellation, and thickness estimation for clinical brain MRI scans of any resolution and contrast. Our approach employs a hybrid analysis method that combines a convolutional neural network (CNN) trained with domain randomization to predict signed distance functions (SDFs), and classical geometry processing for accurate surface placement while maintaining topological and geometric constraints. The method does not require retraining for different acquisitions, thus simplifying the analysis of heterogeneous clinical datasets. We evaluated recon-all-clinical on multiple public datasets like ADNI, HCP, AIBL, OASIS and including a large clinical dataset of over 9,500 scans. The results indicate that our method produces geometrically precise cortical reconstructions across different MRI contrasts and resolutions, consistently achieving high accuracy in parcellation. Cortical thickness estimates are precise enough to capture aging effects, independently of MRI contrast, even though accuracy varies with slice thickness. Our method is publicly available at https://surfer.nmr.mgh.harvard.edu/fswiki/recon-all-clinical, enabling researchers to perform detailed cortical analysis on the huge amounts of already existing clinical MRI scans. This advancement may be particularly valuable for studying rare diseases and underrepresented populations where research-grade MRI data is scarce.

MPRAGElike: A novel approach to generate T1w images from multi-contrast gradient echo images for brain segmentation

PURPOSE

Brain segmentation and multi-parameter mapping (MPM) are important steps in neurodegenerative disease characterization. However, acquiring both a high-resolution T1w sequence like MPRAGE (standard input to brain segmentation) and an MPM in the same neuroimaging protocol increases scan time and patient discomfort, making it difficult to combine both in clinical examinations.

METHODS

A novel approach to synthesize T1w images from MPM images, named MPRAGElike, is proposed and compared to the standard technique used to produce synthetic MPRAGE images (synMPRAGE). Twenty-three healthy subjects were scanned with the same imaging protocol at three different 7T sites using universal parallel transmit RF pulses. SNR, CNR, and automatic brain segmentation results from both MPRAGElike and synMPRAGE were compared against an acquired MPRAGE.

RESULTS

The proposed MPRAGElike technique produced higher SNR values than synMPRAGE for all regions evaluated while also having higher CNR values for subcortical structures. MPRAGE was still the image with the highest SNR values overall. For automatic brain segmentation, MPRAGElike outperformed synMPRAGE when compared to MPRAGE (median Dice Similarity Coefficient of 0.90 versus 0.29 and Average Asymmetric Surface Distance of 0.33 versus 2.93 mm, respectively), in addition to being simple, flexible, and considerably more robust to low image quality than synMPRAGE.

CONCLUSION

The MPRAGElike technique can provide a better and more reliable alternative to synMPRAGE as a substitute for MPRAGE, especially when automatic brain segmentation is of interest and scan time is limited.

Scale-dependent brain age with higher-order statistics from structural magnetic resonance imaging

Inferring chronological age from magnetic resonance imaging (MRI) brain data has become a valuable tool for the early detection of neurodegenerative diseases. We present a method inspired by cosmological techniques for analyzing galaxy surveys, utilizing higher-order summary statistics with multivariate two- and three-point analyses in 3D Fourier space. This method offers physiological interpretability during the inference, allowing the detection of scales where brain anatomy differs across age groups, providing insights into brain aging processes. Similarly to the evolution of cosmic structures, the brain structure also evolves naturally but displays contrasting behaviors at different scales. On larger scales, structure loss occurs with age, possibly due to ventricular expansion, while smaller scales show increased structure, likely related to decreased cortical thickness and gray/white matter volume. Using MRI data from the OASIS-3 database of 869 sessions, our method predicts chronological age with a Mean Absolute Error (MAE) of 3.1 years, while providing information as a function of scale. A posterior density estimation shows that the 1- σ uncertainty for each individual varies between ~ 2 and 8 years, suggesting that, beyond sample variance, complex genetic or lifestyle-related factors may influence brain aging. We perform a twofold validation of the method. First, we apply the method to the Cam-CAN dataset, yielding a MAE of ~ 5.9 years for the age range from 18 to 88 years. Second, we apply the method to thousands of simulated MRI images generated with a state-of-the-art Latent Diffusion model. This work demonstrates the utility of interdisciplinary research, bridging cosmological methods and neuroscience.

Versatile MRI acquisition and processing protocol for population-based neuroimaging

Neuroimaging has an essential role in studies of brain health and of cerebrovascular and neurodegenerative diseases, requiring the availability of versatile magnetic resonance imaging (MRI) acquisition and processing protocols. We designed and developed a multipurpose high-resolution MRI protocol for large-scale and long-term population neuroimaging studies that includes structural, diffusion-weighted and functional MRI modalities. This modular protocol takes almost 1 h of scan time and is, apart from a concluding abdominal scan, entirely dedicated to the brain. The protocol links the acquisition of an extensive set of MRI contrasts directly to the corresponding fully automated data processing pipelines and to the required quality assurance of the MRI data and of the image-derived phenotypes. Since its successful implementation in the population-based Rhineland Study (ongoing, currently more than 11,000 participants, target participant number of 20,000), the proposed MRI protocol has proved suitable for epidemiological and clinical cross-sectional and longitudinal studies, including multisite studies. The approach requires expertise in magnetic resonance image acquisition, in computer science for the data management and the execution of processing pipelines, and in brain anatomy for the quality assessment of the MRI data. The protocol takes ~1 h of MRI acquisition and ~20 h of data processing to complete for a single dataset, but parallelization over multiple datasets using high-performance computing resources reduces the processing time. By making the protocol, MRI sequences and pipelines available, we aim to contribute to better comparability, interoperability and reusability of large-scale neuroimaging data.

A computational ontology framework for the synthesis of multi-level pathology reports from brain MRI scans

BackgroundConvolutional neural network (CNN) based volumetry of MRI data can help differentiate Alzheimer's disease (AD) and the behavioral variant of frontotemporal dementia (bvFTD) as causes of cognitive decline and dementia. However, existing CNN-based MRI volumetry tools lack a structured hierarchical representation of brain anatomy, which would allow for aggregating regional pathological information and automated computational inference.ObjectiveDevelop a computational ontology pipeline for quantifying hierarchical pathological abnormalities and visualize summary charts for brain atrophy findings, aiding differential diagnosis.MethodsUsing FastSurfer, we segmented brain regions and measured volume and cortical thickness from MRI scans pooled across multiple cohorts (N = 3433; ADNI, AIBL, DELCODE, DESCRIBE, EDSD, and NIFD), including healthy controls, prodromal and clinical AD cases, and bvFTD cases. Employing the Web Ontology Language (OWL), we built a semantic model encoding hierarchical anatomical information. Additionally, we created summary visualizations based on sunburst plots for visual inspection of the information stored in the ontology.ResultsOur computational framework dynamically estimated and aggregated regional pathological deviations across different levels of neuroanatomy abstraction. The disease similarity index derived from the volumetric and cortical thickness deviations achieved an AUC of 0.88 for separating AD and bvFTD, which was also reflected by distinct atrophy profile visualizations.ConclusionsThe proposed automated pipeline facilitates visual comparison of atrophy profiles across various disease types and stages. It provides a generalizable computational framework for summarizing pathologic findings, potentially enhancing the physicians' ability to evaluate brain pathologies robustly and interpretably.

Distinct patterns of cerebral and spinal pathology along the spectrum of ATXN2-related disorders

BACKGROUND

The ATXN2 gene contains a polymorphic CAG-rich region encoding a polyglutamine tract in ataxin- 2. Normal alleles have fewer than 27 CAG repeats, 27-34 repeats pose a risk for ALS (ATXN2-ALS), and > 34 repeats cause spinocerebellar ataxia type 2 (SCA2). The striking phenotypic differences between these two ATXN2-related conditions are not yet fully understood.

OBJECTIVE

To characterize and compare the distinguishing radiological signatures of ATXN2-ALS, SCA2, sporadic ALS (sALS) and healthy controls in vivo using quantitative computational neuroimaging techniques.

METHODS

Four groups were defined: healthy controls (n = 34), sALS (n = 17), ATXN2-ALS (n = 16), and SCA2 (n = 17). Cortical, subcortical, brainstem, cerebellar and spinal regions were segmented based on T1-weighted data using validated segmentation tools and their volumes estimated. Group-specific morphometric data were correlated with cerebral ATXN2 expression maps from the Allen Human Brain Atlas.

RESULTS

Study groups were age and sex-matched. sALS, ATXN2-ALS and SCA2 have distinct structural CNS signatures, with disease burden restricted to the precentral gyri in the sALS group, to the spinal cord and brainstem in the ATXN2-ALS group and more diffusely distributed in the subcortical structures in the SCA2 group. Brain ATXN2 expression correlated with the structural signature of SCA2, but not with that of ATXN2-ALS.

CONCLUSIONS

Neuroimaging signatures differ in ATXN2-ALS and SCA2, indicating distinct mechanisms of ATXN2-mediated neurodegeneration. sALS and ATXN2-ALS also exhibit distinct patterns of CNS involvement. The unique imaging signatures and clinical profiles along the spectrum of ATXN2-related disorders raise important questions regarding the pathophysiology of the disease and have practical clinical ramifications.

Brain Volume Loss After Stereotactic Laser Interstitial Thermal Therapy in Patients With Temporal Lobe Epilepsy

BACKGROUND AND PURPOSE

Temporal lobe epilepsy is the most common form of focal epilepsy. MR-guided laser interstitial thermal therapy (LITT) of the amygdalohippocampal complex has become an established therapy option in case of drug resistance. Long-term anatomic network effects on the brain due to deafferentiation have not yet been evaluated.

METHODS

We analyzed brain volumes of 11 patients with temporal lobe epilepsy before and 1-year after hippocampal LITT with FastSurfer segmenting T1-weighted data. Additionally, we performed visual ratings and measurements.

RESULTS

A total of 11 patients with temporal lobe epilepsy (7 left-sided, 4 right-sided) were included (5 females); the mean age years (±standard deviation) at surgery was 41.5 (±18.4) years. The mean postoperative defect size was 1427 (±517) mm3. Volumetry as well as visual ratings found a progressive volume loss after left-sided surgery in the ipsilateral temporal lobe, the contralateral (right) part of the thalamus, and especially contralateral (right) fusiform cortex. These changes could not be detected for right-sided surgery.

CONCLUSION

A (partial) ablation of the left (dominant) hippocampus appears to exert long-term effects on the right thalamus and right-sided temporal cortices. However, we could not observe this effect in the reverse direction. Volumetric studies for larger cohorts should be conducted to investigate these findings.

From histology to macroscale function in the human amygdala

The amygdala is a subcortical region in the mesiotemporal lobe that plays a key role in emotional and sensory functions. Conventional neuroimaging experiments treat this structure as a single, uniform entity, but there is ample histological evidence for subregional heterogeneity in microstructure and function. The current study characterized subregional structure-function coupling in the human amygdala, integrating post-mortem histology and in vivo MRI at ultra-high fields. Core to our work was a novel neuroinformatics approach that leveraged multiscale texture analysis as well as non-linear dimensionality reduction techniques to identify salient dimensions of microstructural variation in a 3D post-mortem histological reconstruction of the human amygdala. We observed two axes of subregional variation in this region, describing inferior-superior as well as mediolateral trends in microstructural differentiation that in part recapitulated established atlases of amygdala subnuclei. Translating our approach to in vivo MRI data acquired at 7 Tesla, we could demonstrate the generalizability of these spatial trends across 10 healthy adults. We then cross-referenced microstructural axes with functional blood-oxygen-level dependent (BOLD) signal analysis obtained during task-free conditions, and revealed a close association of structural axes with macroscale functional network embedding, notably the temporo-limbic, default mode, and sensory-motor networks. Our novel multiscale approach consolidates descriptions of amygdala anatomy and function obtained from histological and in vivo imaging techniques.

Pseudo-MRI Engine for MRI-Free Electromagnetic Source Imaging

Structural head MRIs are a crucial ingredient in MEG/EEG source imaging; they are used to define a realistically shaped volume conductor model, constrain the source space, and visualize the source estimates. However, individual MRIs are not always available, or they may be of insufficient quality for segmentation, leading to the use of a generic template MRI, matched MRI, or the application of a spherical conductor model. Such approaches deviate the model geometry from the true head structure and limit the accuracy of the forward solution. Here, we implemented an easy-to-use tool, pseudo-MRI engine, which utilizes the head-shape digitization acquired during a MEG/EEG measurement for warping an MRI template to fit the subject's head. To this end, the algorithm first removes outlier digitization points, densifies the point cloud by interpolation if needed, and finally warps the template MRI and its segmented surfaces to the individual head shape using the thin-plate-spline method. To validate the approach, we compared the geometry of segmented head surfaces, cortical surfaces, and canonical brain regions in the real and pseudo-MRIs of 25 subjects. We also tested the MEG source reconstruction accuracy with pseudo-MRIs against that obtained with the real MRIs from individual subjects with simulated and real MEG data. We found that the pseudo-MRI enables comparable source localization accuracy to the one obtained with the subject's real MRI. The study indicates that pseudo-MRI can replace the need for individual MRI scans in MEG/EEG source imaging for applications that do not require subcentimeter spatial accuracy.

FastSurfer-LIT: Lesion inpainting tool for whole-brain MRI segmentation with tumors, cavities, and abnormalities

Resection cavities, tumors, and other lesions can fundamentally alter brain structure and present as abnormalities in brain MRI. Specifically, quantifying subtle neuroanatomical changes in other, not directly affected regions of the brain is essential to assess the impact of tumors, surgery, chemo/radiotherapy, or drug treatments. However, only a limited number of solutions address this important task, while many standard analysis pipelines simply do not support abnormal brain images at all. In this paper, we present a method to perform sensitive neuroanatomical analysis of healthy brain regions in the presence of large lesions and cavities. Our approach called "FastSurfer Lesion Inpainting Tool" (FastSurfer-LIT) leverages the recently emerged Denoising Diffusion Probabilistic Models (DDPM) to fill lesion areas with healthy tissue that matches and extends the surrounding tissue. This enables subsequent processing with established MRI analysis methods such as the calculation of adjusted volume and surface measurements using FastSurfer or FreeSurfer. FastSurfer-LIT significantly outperforms previously proposed solutions on a large dataset of simulated brain tumors (N = 100) and synthetic multiple sclerosis lesions (N = 39) with improved Dice and Hausdorff measures, and also on a highly heterogeneous dataset with lesions and cavities in a manual assessment (N = 100). Finally, we demonstrate increased reliability to reproduce pre-operative cortical thickness estimates from corresponding post-operative temporo-mesial resection surgery MRIs. The method is publicly available at https://github.com/Deep-MI/LIT and will be integrated into the FastSurfer toolbox.

Functional and structural cerebellar-behavior relationships in aging

Healthy aging is associated with deficits in cognitive performance and brain changes, including in the cerebellum. Cerebellar communication with the cortex via closed-loop circuits through the thalamus have been established and these circuits are closely related to the functional topography of the cerebellum. In this study, we sought to elucidate relationships between cerebellar structure and function with cognition in healthy aging. We explored this relationship in 138 healthy adults (aged 35-86, 53% female) using resting-state functional connectivity MRI (fcMRI), cerebellar volume, and cognitive and motor assessments. Behavioral tasks assessed attention, processing speed, working memory, episodic memory, and motor abilities. We expected to find negative relationships between lobular volume with age, and positive relationships between specific lobular volumes with motor and cognitive behavior, respectively. We predicted lower cerebello-cortical fcMRI with increased age. Behaviorally, we expected higher cerebello-frontal and cerebello-association area fcMRI cerebellar connectivity to correlate with better behavioral performance. Correlations were conducted between cerebellar lobules I-IV, V, Crus I, Crus II, vermis VI and behavioral measures. We found lower volumes with increased age as well as both higher and lower cerebellar connectivity relationships with increased age, consistent with literature on functional connectivity and network segregation in aging. Further, we revealed unique associations between cerebellar structure and connectivity with comprehensive behavioral measures in a healthy aging population. Our findings further highlight the role of the cerebellum in aging.

Conservation of structural brain connectivity in people with multiple sclerosis

Multiple sclerosis (MS) is a neurodegenerative disease that affects the central nervous system. Structures affected in MS include the corpus callosum, connecting the hemispheres. Studies have shown that in mammalian brains, structural connectivity is organized according to a conservation principle, an inverse relationship between intra- and interhemispheric connectivity. The aim of this study was to replicate this conservation principle in subjects with MS and to explore how the disease interacts with it. A multicentric dataset has been analyzed including 513 people with MS and 208 healthy controls from seven different centers. Structural connectivity was quantified through various connectivity measures, and graph analysis was used to study the behavior of intra- and interhemispheric connectivity. The association between the intra- and the interhemispheric connectivity showed a similar strength for healthy controls (r = 0.38, p < 0.001) and people with MS (r = 0.35, p < 0.001). Intrahemispheric connectivity was associated with white matter fraction (r = 0.48, p < 0.0001), lesion volume (r = -0.44, p < 0.0001), and the Symbol Digit Modalities Test (r = 0.25, p < 0.0001). Results show that this conservation principle seems to hold for people with MS. These findings support the hypothesis that interhemispheric connectivity decreases at higher cognitive decline and disability levels, while intrahemispheric connectivity increases to maintain the balance.

Sensitivity of Advanced Magnetic Resonance Imaging to Progression over Six Months in Early Spinocerebellar Ataxia

BACKGROUND

Clinical trials for upcoming disease-modifying therapies of spinocerebellar ataxias (SCA), a group of rare movement disorders, lack endpoints sensitive to early disease progression, when therapeutics will be most effective. In addition, regulatory agencies emphasize the importance of biological outcomes.

OBJECTIVES

READISCA, a transatlantic clinical trial readiness consortium, investigated whether advanced multimodal magnetic resonance imaging (MRI) detects pathology progression over 6 months in preataxic and early ataxic carriers of SCA mutations.

METHODS

A total of 44 participants (10 SCA1, 25 SCA3, and 9 controls) prospectively underwent 3-T MR scanning at baseline and a median [interquartile range] follow-up of 6.2 [5.9-6.7] months; 44% of SCA participants were preataxic. Blinded analyses of annual changes in structural, diffusion MRI, MR spectroscopy, and the Scale for Assessment and Rating of Ataxia (SARA) were compared between groups using nonparametric testing. Sample sizes were estimated for 6-month interventional trials with 50% to 100% treatment effect size, leveraging existing large cohort data (186 SCA1, 272 SCA3) for the SARA estimate.

RESULTS

Rate of change in microstructural integrity (decrease in fractional anisotropy, increase in diffusivities) in the middle cerebellar peduncle, corona radiata, and superior longitudinal fasciculus significantly differed in SCAs from controls (P < 0.005), with high effect sizes (Cohen's d = 1-2) and moderate-to-high responsiveness (|standardized response mean| = 0.6-0.9) in SCAs. SARA scores did not change, and their rate of change did not differ between groups.

CONCLUSIONS

Diffusion MRI is sensitive to disease progression at very early-stage SCA1 and SCA3 and may provide a >5-fold reduction in sample sizes relative to SARA as endpoint for 6-month-long trials. © 2024 The Author(s). Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Automated brain segmentation and volumetry in dementia diagnostics: a narrative review with emphasis on FreeSurfer

Background

Dementia can be caused by numerous different diseases that present variable clinical courses and reveal multiple patterns of brain atrophy, making its accurate early diagnosis by conventional examinative means challenging. Although highly accurate and powerful, magnetic resonance imaging (MRI) currently plays only a supportive role in dementia diagnosis, largely due to the enormous volume and diversity of data it generates. AI-based software solutions/algorithms that can perform automated segmentation and volumetry analyses of MRI data are being increasingly used to address this issue. Numerous commercial and non-commercial software solutions for automated brain segmentation and volumetry exist, with FreeSurfer being the most frequently used.

Objectives

This Review is an account of the current situation regarding the application of automated brain segmentation and volumetry to dementia diagnosis.

Methods

We performed a PubMed search for “FreeSurfer AND Dementia” and obtained 493 results. Based on these search results, we conducted an in-depth source analysis to identify additional publications, software tools, and methods. Studies were analyzed for design, patient collective, and for statistical evaluation (mathematical methods, correlations).

Results

In the studies identified, the main diseases and cohorts represented were Alzheimer’s disease (n = 276), mild cognitive impairment (n = 157), frontotemporal dementia (n = 34), Parkinson’s disease (n = 29), dementia with Lewy bodies (n = 20), and healthy controls (n = 356). The findings and methods of a selection of the studies identified were summarized and discussed.

Conclusion

Our evaluation showed that, while a large number of studies and software solutions are available, many diseases are underrepresented in terms of their incidence. There is therefore plenty of scope for targeted research.

Sex-steroid hormones relate to cerebellar structure and functional connectivity across adulthood

Aging involves complex biological changes that affect disease susceptibility and aging trajectories. Although females typically live longer than males, they have a higher susceptibility to diseases like Alzheimer's, speculated to be influenced by menopause, and reduced ovarian hormone production. Understanding sex-specific differences is crucial for personalized medical interventions and gender equality in health. Our study aims to elucidate sex differences in regional cerebellar structure and connectivity during normal aging by investigating both structural and functional connectivity variations, with a focus on investigating these differences in the context of sex-steroid hormones. The study included 138 participants (mean age = 57(13.3) years, age range = 35-86 years, 54% women). The cohort was divided into three groups: 38 early middle-aged individuals (EMA) (mean age = 41(4.7) years), 48 late middle-aged individuals (LMA) (mean age = 58(4) years), and 42 older adults (OA) (mean age = 72(6.3) years). All participants underwent MRI scans, and saliva samples were collected for sex-steroid hormone quantification (17β-estradiol (E), progesterone (P), and testosterone (T)). We found less connectivity in females between Lobule I-IV and the cuneus, and greater connectivity in females between Crus I, Crus II, and the precuneus with increased age. Higher 17β-estradiol levels were linked to greater connectivity in Crus I and Crus II cerebellar subregions. Analyzing all participants together, testosterone was associated with both higher and lower connectivity in Lobule I-IV and Crus I, respectively, while higher progesterone levels were linked to lower connectivity in females. Structural differences were observed, with EMA males having larger volumes compared to LMA and OA groups, particularly in the right I-IV, right Crus I, right V, and right VI. EMA females showed higher volumes in the right lobules V and VI. These results highlight the significant role of sex hormones in modulating cerebellar connectivity and structure across adulthood, emphasizing the need to consider sex and hormonal status in neuroimaging studies to better understand age-related cognitive decline and neurological disorders.

Functional and Structural Cerebellar-Behavior Relationships in Aging

Healthy aging is associated with deficits in cognitive performance and brain changes, including in the cerebellum. Yet, the precise link between cerebellar function/structure and cognition in aging remains poorly understood. We explored this relationship in 138 healthy adults (aged 35-86, 53% female) using resting-state functional connectivity MRI (fcMRI), cerebellar volume, and cognitive and motor assessments in an aging sample. We expected to find negative relationships between lobular volume for with age, and positive relationships between specific lobular volumes with motor and cognition respectively. We predicted lower cerebellar fcMRI to cortical networks and circuits with increased age. Behaviorally, we expected higher cerebello-frontal fcMRI cerebellar connectivity with association areas to correlate with better behavioral performance. Behavioral tasks broadly assessed attention, processing speed, working memory, episodic memory, and motor abilities. Correlations were conducted between cerebellar lobules I-IV, V, Crus I, Crus II, vermis VI and behavioral measures. We found lower volumes with increased age as well as bidirectional cerebellar connectivity relationships with increased age, consistent with literature on functional connectivity and network segregation in aging. Further, we revealed unique associations for both cerebellar structure and connectivity with comprehensive behavioral measures in a healthy aging population. Our findings underscore cerebellar involvement in behavior during aging.

VINNA for neonates: Orientation independence through latent augmentations

A robust, fast, and accurate segmentation of neonatal brain images is highly desired to better understand and detect changes during development and disease, specifically considering the rise in imaging studies for this cohort. Yet, the limited availability of ground truth datasets, lack of standardized acquisition protocols, and wide variations of head positioning in the scanner pose challenges for method development. A few automated image analysis pipelines exist for newborn brain Magnetic Resonance Image (MRI) segmentation, but they often rely on time-consuming non-linear spatial registration procedures and require resampling to a common resolution, subject to loss of information due to interpolation and down-sampling. Without registration and image resampling, variations with respect to head positions and voxel resolutions have to be addressed differently. In deep learning, external augmentations such as rotation, translation, and scaling are traditionally used to artificially expand the representation of spatial variability, which subsequently increases both the training dataset size and robustness. However, these transformations in the image space still require resampling, reducing accuracy specifically in the context of label interpolation. We recently introduced the concept of resolution-independence with the Voxel-size Independent Neural Network framework, VINN. Here, we extend this concept by additionally shifting all rigid-transforms into the network architecture with a four degree of freedom (4-DOF) transform module, enabling resolution-aware internal augmentations (VINNA) for deep learning. In this work, we show that VINNA (i) significantly outperforms state-of-the-art external augmentation approaches, (ii) effectively addresses the head variations present specifically in newborn datasets, and (iii) retains high segmentation accuracy across a range of resolutions (0.5-1.0 mm). Furthermore, the 4-DOF transform module together with internal augmentations is a powerful, general approach to implement spatial augmentation without requiring image or label interpolation. The specific network application to newborns will be made publicly available as VINNA4neonates.

Predicting Brain Age and Gender from Brain Volume Data Using Variational Quantum Circuits

The morphology of the brain undergoes changes throughout the aging process, and accurately predicting a person's brain age and gender using brain morphology features can aid in detecting atypical brain patterns. Neuroimaging-based estimation of brain age is commonly used to assess an individual's brain health relative to a typical aging trajectory, while accurately classifying gender from neuroimaging data offers valuable insights into the inherent neurological differences between males and females. In this study, we aimed to compare the efficacy of classical machine learning models with that of a quantum machine learning method called a variational quantum circuit in estimating brain age and predicting gender based on structural magnetic resonance imaging data. We evaluated six classical machine learning models alongside a quantum machine learning model using both combined and sub-datasets, which included data from both in-house collections and public sources. The total number of participants was 1157, ranging from ages 14 to 89, with a gender distribution of 607 males and 550 females. Performance evaluation was conducted within each dataset using training and testing sets. The variational quantum circuit model generally demonstrated superior performance in estimating brain age and gender classification compared to classical machine learning algorithms when using the combined dataset. Additionally, in benchmark sub-datasets, our approach exhibited better performance compared to previous studies that utilized the same dataset for brain age prediction. Thus, our results suggest that variational quantum algorithms demonstrate comparable effectiveness to classical machine learning algorithms for both brain age and gender prediction, potentially offering reduced error and improved accuracy.

A computational pipeline towards large-scale and multiscale modeling of traumatic axonal injury

Contemporary biomechanical modeling of traumatic brain injury (TBI) focuses on either the global brain as an organ or a representative tiny section of a single axon. In addition, while it is common for a global brain model to employ real-world impacts as input, axonal injury models have largely been limited to inputs of either tension or compression with assumed peak strain and strain rate. These major gaps between global and microscale modeling preclude a systematic and mechanistic investigation of how tissue strain from impact leads to downstream axonal damage throughout the white matter. In this study, a unique subject-specific multimodality dataset from a male ice-hockey player sustaining a diagnosed concussion is used to establish an efficient and scalable computational pipeline. It is then employed to derive voxelized brain deformation, maximum principal strains and white matter fiber strains, and finally, to produce diverse fiber strain profiles of various shapes in temporal history necessary for the development and application of a deep learning axonal injury model in the future. The pipeline employs a structured, voxelized representation of brain deformation with adjustable spatial resolution independent of model mesh resolution. The method can be easily extended to other head impacts or individuals. The framework established in this work is critical for enabling large-scale (i.e., across the entire white matter region, head impacts, and individuals) and multiscale (i.e., from organ to cell length scales) modeling for the investigation of traumatic axonal injury (TAI) triggering mechanisms. Ultimately, these efforts could enhance the assessment of concussion risks and design of protective headgear. Therefore, this work contributes to improved strategies for concussion detection, mitigation, and prevention.

U-fiber analysis: a toolbox for automated quantification of U-fibers and white matter hyperintensities

Background

Whether white matter hyperintensities (WMHs) involve U-fibers is of great value in understanding the different etiologies of cerebral white matter (WM) lesions. However, clinical practice currently relies only on the naked eye to determine whether WMHs are in the vicinity of U-fibers, and there is a lack of good neuroimaging tools to quantify WMHs and U-fibers.

Methods

Here, we developed a multimodal neuroimaging toolbox named U-fiber analysis (UFA) that can automatically extract WMHs and quantitatively characterize the volume and number of WMHs in different brain regions. In addition, we proposed an anatomically constrained U-fiber tracking scheme and quantitatively characterized the microstructure diffusion properties, fiber length, and number of U-fibers in different brain regions to help clinicians to quantitatively determine whether WMHs in the proximal cortex disrupt the microstructure of U-fibers. To validate the utility of the UFA toolbox, we analyzed the neuroimaging data from 246 patients with cerebral small vessel disease (cSVD) enrolled at Zhongshan Hospital between March 2018 and November 2019 in a cross-sectional study.

Results

According to the manual judgment of the clinician, the patients with cSVD were divided into a WMHs involved U-fiber group (U-fiber-involved group, 51 cases) and WMHs not involved U-fiber group (U-fiber-spared group, 163 cases). There were no significant differences between the U-fiber-spared group and the U-fiber-involved group in terms of age (P=0.143), gender (P=0.462), education (P=0.151), Mini-Mental State Examination (MMSE) scores (P=0.151), and Montreal Cognitive Assessment (MoCA) scores (P=0.411). However, patients in the U-fiber-involved group had higher Fazekas scores (P<0.001) and significantly higher whole brain WMHs (P=0.046) and deep WMH volumes (P<0.001) compared to patients in the U-fiber-spared group. Moreover, the U-fiber-involved group had higher WMH volumes in the bilateral frontal [P(left) <0.001, P(right) <0.001] and parietal lobes [P(left) <0.001, P(right) <0.001]. On the other hand, patients in the U-fiber-involved group had higher mean diffusivity (MD) and axial diffusivity (AD) in the bilateral parietal [P(left, MD) =0.048, P(right, MD) =0.045, P(left, AD) =0.015, P(right, AD) =0.015] and right frontal-parietal regions [P(MD) =0.048, P(AD) =0.027], and had significantly reduced mean fiber length and number in the right parietal [P(length) =0.013, P(number) =0.028] and right frontal-parietal regions [P(length) =0.048] compared to patients in the U-fiber-spared group.

Conclusions

Our results suggest that WMHs in the proximal cortex may disrupt the microstructure of U-fibers. Our tool may provide new insights into the understanding of WM lesions of different etiologies in the brain.

Cortical Network Disruption Is Minimal in Early Stages of Psychosis

Background and Hypothesis

Schizophrenia is associated with white matter disruption and topological reorganization of cortical connectivity but the trajectory of these changes, from the first psychotic episode to established illness, is poorly understood. Current studies in first-episode psychosis (FEP) patients using diffusion magnetic resonance imaging (dMRI) suggest such disruption may be detectable at the onset of psychosis, but specific results vary widely, and few reports have contextualized their findings with direct comparison to young adults with established illness.

Study Design

Diffusion and T1-weighted 7T MR scans were obtained from N = 112 individuals (58 with untreated FEP, 17 with established schizophrenia, 37 healthy controls) recruited from London, Ontario. Voxel- and network-based analyses were used to detect changes in diffusion microstructural parameters. Graph theory metrics were used to probe changes in the cortical network hierarchy and to assess the vulnerability of hub regions to disruption. The analysis was replicated with N = 111 (57 patients, 54 controls) from the Human Connectome Project-Early Psychosis (HCP-EP) dataset.

Study Results

Widespread microstructural changes were found in people with established illness, but changes in FEP patients were minimal. Unlike the established illness group, no appreciable topological changes in the cortical network were observed in FEP patients. These results were replicated in the early psychosis patients of the HCP-EP datasets, which were indistinguishable from controls in most metrics.

Conclusions

The white matter structural changes observed in established schizophrenia are not a prominent feature in the early stages of this illness.

Computational Neurosurgery in Deep Brain Stimulation

Computational methods and technologies are critical for neurosurgery in general and in deep brain stimulation (DBS) in particular. They increasingly inform every aspect of clinical DBS therapy, from presurgical planning and hardware implantation to postoperative adjustment of stimulation parameters. Computational methods also occupy a prominent position within the DBS research sphere, where they facilitate efforts to better understand DBS' underlying mechanisms and optimize and individualize its delivery. This chapter provides a high-level overview of the various computational tools and methods that have been applied to DBS. First, we discuss the invaluable contribution of computational neuroimaging (primarily magnetic resonance imaging) to DBS, targeting and the role of postoperative methods of image analysis-specifically, electrode localization, volume of activated tissue modeling, and sweet-spot mapping-in precisely localizing DBS' targets in the brain and discerning optimal treatment loci. We then address the growing field of connectomics, which leverages specific magnetic resonance imaging (MRI) sequences and post-acquisition processing algorithms to explore how DBS operates at the level of brain-wide networks. Next, the search for electrophysiological and imaging-based biomarkers of optimal DBS therapy is explored. We lastly touch on the incipient field of spatial characterization analysis and discuss the ongoing development of adaptive, closed-loop DBS systems.

FastSurfer-HypVINN: Automated sub-segmentation of the hypothalamus and adjacent structures on high-resolutional brain MRI

The hypothalamus plays a crucial role in the regulation of a broad range of physiological, behavioral, and cognitive functions. However, despite its importance, only a few small-scale neuroimaging studies have investigated its substructures, likely due to the lack of fully automated segmentation tools to address scalability and reproducibility issues of manual segmentation. While the only previous attempt to automatically sub-segment the hypothalamus with a neural network showed promise for 1.0 mm isotropic T1-weighted (T1w) magnetic resonance imaging (MRI), there is a need for an automated tool to sub-segment also high-resolutional (HiRes) MR scans, as they are becoming widely available, and include structural detail also from multi-modal MRI. We, therefore, introduce a novel, fast, and fully automated deep-learning method named HypVINN for sub-segmentation of the hypothalamus and adjacent structures on 0.8 mm isotropic T1w and T2w brain MR images that is robust to missing modalities. We extensively validate our model with respect to segmentation accuracy, generalizability, in-session test-retest reliability, and sensitivity to replicate hypothalamic volume effects (e.g., sex differences). The proposed method exhibits high segmentation performance both for standalone T1w images as well as for T1w/T2w image pairs. Even with the additional capability to accept flexible inputs, our model matches or exceeds the performance of state-of-the-art methods with fixed inputs. We, further, demonstrate the generalizability of our method in experiments with 1.0 mm MR scans from both the Rhineland Study and the UK Biobank-an independent dataset never encountered during training with different acquisition parameters and demographics. Finally, HypVINN can perform the segmentation in less than a minute (graphical processing unit [GPU]) and will be available in the open source FastSurfer neuroimaging software suite, offering a validated, efficient, and scalable solution for evaluating imaging-derived phenotypes of the hypothalamus.

CSF, Blood, and MRI Biomarkers in Skogholt's Disease-A Rare Neurodegenerative Disease in a Norwegian Kindred

Skogholt's disease is a rare neurological disorder that is only observed in a small Norwegian kindred. It typically manifests in adulthood with uncharacteristic neurological symptoms from both the peripheral and central nervous systems. The etiology of the observed cerebral white matter lesions and peripheral myelin pathology is unclear. Increased cerebrospinal fluid (CSF) concentrations of protein have been confirmed, and recently, very high concentrations of CSF total and phosphorylated tau have been detected in Skogholt patients. The symptoms and observed biomarker changes in Skogholt's disease are largely nonspecific, and further studies are necessary to elucidate the disease mechanisms. Here, we report the results of neurochemical analyses of plasma and CSF, as well as results from the morphometric segmentation of cerebral magnetic resonance imaging. We analyzed the biomarkers Aβ1--42, Aβ1-40, Aβx-38, Aβx-40, Aβx-42, total and phosphorylated tau, glial fibrillary acidic protein, neurofilament light chain, platelet-derived growth factor receptor beta, and beta-trace protein. All analyzed CSF biomarkers, except neurofilament light chain and Aβ1/x-42, were increased several-fold. In blood, none of these biomarkers were significantly different between the Skogholt and control groups. MRI volumetric segmentation revealed decreases in the ventricular, white matter, and choroid plexus volumes in the Skogholt group, with an accompanying increase in white matter lesions. The cortical thickness and subcortical gray matter volumes were increased in the Skogholt group. Pathophysiological changes resulting from choroidal dysfunction and/or abnormal CSF turnover, which may cause the increases in CSF protein and brain biomarker levels, are discussed.

Sleep Differentially and Profoundly Impairs Recall Memory in a Patient with Fornix Damage

In March 2020, C.T., a kind, bright, and friendly young woman underwent surgery for a midline tumor involving her septum pellucidum and extending down into her fornices bilaterally. Following tumor diagnosis and surgery, C.T. experienced significant memory deficits: C.T.'s family reported that she could remember things throughout the day, but when she woke up in the morning or following a nap, she would expect to be in the hospital, forgetting all the information that she had learned before sleep. The current study aimed to empirically validate C.T.'s pattern of memory loss and explore its neurological underpinnings. On two successive days, C.T. and age-matched controls watched an episode of a TV show and took a nap or stayed awake before completing a memory test. Although C.T. performed numerically worse than controls in both conditions, sleep profoundly exacerbated her memory impairment, such that she could not recall any details following a nap. This effect was replicated in a second testing session. High-resolution MRI scans showed evidence of the trans-callosal surgical approach's impact on the mid-anterior corpus callosum, indicated that C.T. had perturbed white matter particularly in the right fornix column, and demonstrated that C.T.'s hippocampal volumes did not differ from controls. These findings suggest that the fornix is important for processing episodic memories during sleep. As a key output pathway of the hippocampus, the fornix may ensure that specific memories are replayed during sleep, maintain the balance of sleep stages, or allow for the retrieval of memories following sleep.

Automated hippocampal segmentation algorithms evaluated in stroke patients

Deep learning segmentation algorithms can produce reproducible results in a matter of seconds. However, their application to more complex datasets is uncertain and may fail in the presence of severe structural abnormalities-such as those commonly seen in stroke patients. In this investigation, six recent, deep learning-based hippocampal segmentation algorithms were tested on 641 stroke patients of a multicentric, open-source dataset ATLAS 2.0. The comparisons of the volumes showed that the methods are not interchangeable with concordance correlation coefficients from 0.266 to 0.816. While the segmentation algorithms demonstrated an overall good performance (volumetric similarity [VS] 0.816 to 0.972, DICE score 0.786 to 0.921, and Hausdorff distance [HD] 2.69 to 6.34), no single out-performing algorithm was identified: FastSurfer performed best in VS, QuickNat in DICE and average HD, and Hippodeep in HD. Segmentation performance was significantly lower for ipsilesional segmentation, with a decrease in performance as a function of lesion size due to the pathology-based domain shift. Only QuickNat showed a more robust performance in volumetric similarity. Even though there are many pre-trained segmentation methods, it is important to be aware of the possible decrease in performance for the segmentation results on the lesion side due to the pathology-based domain shift. The segmentation algorithm should be selected based on the research question and the evaluation parameter needed. More research is needed to improve current hippocampal segmentation methods.

Quantifying MR Head Motion in the Rhineland Study - A Robust Method for Population Cohorts

Head motion during MR acquisition reduces image quality and has been shown to bias neuromorphometric analysis. The quantification of head motion, therefore, has both neuroscientific as well as clinical applications, for example, to control for motion in statistical analyses of brain morphology, or as a variable of interest in neurological studies. The accuracy of markerless optical head tracking, however, is largely unexplored. Furthermore, no quantitative analysis of head motion in a general, mostly healthy population cohort exists thus far. In this work, we present a robust registration method for the alignment of depth camera data that sensitively estimates even small head movements of compliant participants. Our method outperforms the vendor-supplied method in three validation experiments: 1. similarity to fMRI motion traces as a low-frequency reference, 2. recovery of the independently acquired breathing signal as a high-frequency reference, and 3. correlation with image-based quality metrics in structural T1-weighted MRI. In addition to the core algorithm, we establish an analysis pipeline that computes average motion scores per time interval or per sequence for inclusion in downstream analyses. We apply the pipeline in the Rhineland Study, a large population cohort study, where we replicate age and body mass index (BMI) as motion correlates and show that head motion significantly increases over the duration of the scan session. We observe weak, yet significant interactions between this within-session increase and age, BMI, and sex. High correlations between fMRI and camera-based motion scores of proceeding sequences further suggest that fMRI motion estimates can be used as a surrogate score in the absence of better measures to control for motion in statistical analyses.

Low-field magnetic resonance image enhancement via stochastic image quality transfer

Low-field (<1T) magnetic resonance imaging (MRI) scanners remain in widespread use in low- and middle-income countries (LMICs) and are commonly used for some applications in higher income countries e.g. for small child patients with obesity, claustrophobia, implants, or tattoos. However, low-field MR images commonly have lower resolution and poorer contrast than images from high field (1.5T, 3T, and above). Here, we present Image Quality Transfer (IQT) to enhance low-field structural MRI by estimating from a low-field image the image we would have obtained from the same subject at high field. Our approach uses (i) a stochastic low-field image simulator as the forward model to capture uncertainty and variation in the contrast of low-field images corresponding to a particular high-field image, and (ii) an anisotropic U-Net variant specifically designed for the IQT inverse problem. We evaluate the proposed algorithm both in simulation and using multi-contrast (T1-weighted, T2-weighted, and fluid attenuated inversion recovery (FLAIR)) clinical low-field MRI data from an LMIC hospital. We show the efficacy of IQT in improving contrast and resolution of low-field MR images. We demonstrate that IQT-enhanced images have potential for enhancing visualisation of anatomical structures and pathological lesions of clinical relevance from the perspective of radiologists. IQT is proved to have capability of boosting the diagnostic value of low-field MRI, especially in low-resource settings.

Deep learning in neuroimaging data analysis: Applications, challenges, and solutions

Methods for the analysis of neuroimaging data have advanced significantly since the beginning of neuroscience as a scientific discipline. Today, sophisticated statistical procedures allow us to examine complex multivariate patterns, however most of them are still constrained by assuming inherent linearity of neural processes. Here, we discuss a group of machine learning methods, called deep learning, which have drawn much attention in and outside the field of neuroscience in recent years and hold the potential to surpass the mentioned limitations. Firstly, we describe and explain the essential concepts in deep learning: the structure and the computational operations that allow deep models to learn. After that, we move to the most common applications of deep learning in neuroimaging data analysis: prediction of outcome, interpretation of internal representations, generation of synthetic data and segmentation. In the next section we present issues that deep learning poses, which concerns multidimensionality and multimodality of data, overfitting and computational cost, and propose possible solutions. Lastly, we discuss the current reach of DL usage in all the common applications in neuroimaging data analysis, where we consider the promise of multimodality, capability of processing raw data, and advanced visualization strategies. We identify research gaps, such as focusing on a limited number of criterion variables and the lack of a well-defined strategy for choosing architecture and hyperparameters. Furthermore, we talk about the possibility of conducting research with constructs that have been ignored so far or/and moving toward frameworks, such as RDoC, the potential of transfer learning and generation of synthetic data.

An MRI Scans-Based Alzheimer's Disease Detection via Convolutional Neural Network and Transfer Learning

Alzheimer’s disease (AD) is the most common type (>60%) of dementia and can wreak havoc on the psychological and physiological development of sufferers and their carers, as well as the economic and social development. Attributed to the shortage of medical staff, automatic diagnosis of AD has become more important to relieve the workload of medical staff and increase the accuracy of medical diagnoses. Using the common MRI scans as inputs, an AD detection model has been designed using convolutional neural network (CNN). To enhance the fine-tuning of hyperparameters and, thus, the detection accuracy, transfer learning (TL) is introduced, which brings the domain knowledge from heterogeneous datasets. Generative adversarial network (GAN) is applied to generate additional training data in the minority classes of the benchmark datasets. Performance evaluation and analysis using three benchmark (OASIS-series) datasets revealed the effectiveness of the proposed method, which increases the accuracy of the detection model by 2.85−3.88%, 2.43−2.66%, and 1.8−40.1% in the ablation study of GAN and TL, as well as the comparison with existing works, respectively.