Topological Embedding of Human Brain Networks with Applications to Dynamics of Temporal Lobe Epilepsy.

We introduce a novel, data-driven topological data analysis (TDA) approach for embedding brain networks into a lower-dimensional space in quantifying the dynamics of temporal lobe epilepsy (TLE) obtained from resting-state functional magnetic resonance imaging (rs-fMRI). This embedding facilitates the orthogonal projection of 0D and 1D topological features, allowing for the visualization and modeling of the dynamics of functional human brain networks in a resting state. We then quantify the topological disparities between networks to determine the coordinates for embedding. This framework enables us to conduct a coherent statistical inference within the embedded space. Our results indicate that brain network topology in TLE patients exhibits increased rigidity in 0D topology but more rapid flections compared to that of normal controls in 1D topology.

Unified Topological Inference for Brain Networks in Temporal Lobe Epilepsy Using the Wasserstein Distance.

Persistent homology offers a powerful tool for extracting hidden topological signals from brain networks. It captures the evolution of topological structures across multiple scales, known as filtrations, thereby revealing topological features that persist over these scales. These features are summarized in persistence diagrams, and their dissimilarity is quantified using the Wasserstein distance. However, the Wasserstein distance does not follow a known distribution, posing challenges for the application of existing parametric statistical models. To tackle this issue, we introduce a unified topological inference framework centered on the Wasserstein distance. Our approach has no explicit model and distributional assumptions. The inference is performed in a completely data driven fashion. We apply this method to resting-state functional magnetic resonance images (rs-fMRI) of temporal lobe epilepsy patients collected from two different sites: the University of Wisconsin-Madison and the Medical College of Wisconsin. Importantly, our topological method is robust to variations due to sex and image acquisition, obviating the need to account for these variables as nuisance covariates. We successfully localize the brain regions that contribute the most to topological differences. A MATLAB package used for all analyses in this study is available at https://github.com/laplcebeltrami/PH-STAT.

Genetic and environmental influence on resting state networks in young male and female adults: a cartographer mapping study.

We propose a unique, minimal assumption, approach based on variance analyses (compared with standard approaches) to investigate genetic influence on individual differences on the functional connectivity of the brain using 65 monozygotic and 65 dizygotic healthy young adult twin pairs' low-frequency oscillation resting state functional Magnetic Resonance Imaging (fMRI) data from the Human Connectome Project. Overall, we found high number of genetically-influenced functional (GIF) connections involving posterior to posterior brain regions (occipital/temporal/parietal) implicated in low-level processes such as vision, perception, motion, categorization, dorsal/ventral stream visuospatial, and long-term memory processes, as well as high number across midline brain regions (cingulate) implicated in attentional processes, and emotional responses to pain. We found low number of GIF connections involving anterior to anterior/posterior brain regions (frontofrontal > frontoparietal, frontotemporal, frontooccipital) implicated in high-level processes such as working memory, reasoning, emotional judgment, language, and action planning. We found very low number of GIF connections involving subcortical/noncortical networks such as basal ganglia, thalamus, brainstem, and cerebellum. In terms of sex-specific individual differences, individual differences in males were more genetically influenced while individual differences in females were more environmentally influenced in terms of the interplay of interactions of Task positive networks (brain regions involved in various task-oriented processes and attending to and interacting with environment), extended Default Mode Network (a central brain hub for various processes such as internal monitoring, rumination, and evaluation of self and others), primary sensorimotor systems (vision, audition, somatosensory, and motor systems), and subcortical/noncortical networks. There were >8.5-19.1 times more GIF connections in males than females. These preliminary (young adult cohort-specific) findings suggest that individual differences in the resting state brain may be more genetically influenced in males and more environmentally influenced in females; furthermore, standard approaches may suggest that it is more substantially nonadditive genetics, rather than additive genetics, which contribute to the differences in sex-specific individual differences based on this young adult (male and female) specific cohort. Finally, considering the preliminary cohort-specific results, based on standard approaches, environmental influences on individual differences may be substantially greater than that of genetics, for either sex, frontally and brain-wide. [Correction added on 10 May 2023, after first online publication: added: functional Magnetic Resonance Imaging. Added: individual differences in, twice. Added statement between furthermore … based on standard approaches.].

Comparison of Language and Memory Lateralization by Functional MRI and Wada Test in Epilepsy.

The intracarotid sodium amobarbital procedure (ISAP or Wada test) lateralizes cerebral functions to the cerebral hemispheres preoperatively. Functional magnetic resonance imaging (fMRI) is increasingly used to characterize preoperative language and memory lateralization. In this study, concordance of fMRI with Wada was examined in patients with medically intractable seizures. The relationship of the distance between the epileptogenic focus to functional activation area with patients' post-operative deficits in language was also analyzed. 27 epilepsy patients with preoperative fMRI and Wada data were analyzed using established fMRI paradigms for language and memory. Activation of Broca's and Wernicke's areas were measured in three dimensions. Language and memory lateralization were determined, and standard neuropsychiatry Wada test procedures were used for comparison. The shortest distance between a language area to the border of surgical focus (LAD) was also measured and compared with postoperative language deficits. Our study found that concordance between fMRI and Wada testing was 0.41 (Kappa's 'fair to good' concordance) for language dominance and 0.1 (Kappa's 'poor' concordance) for memory. No significant correlation was found between LAD and post-op language deficit (p=0.439). A correlation was found between LAD and post-op memory deficit (p=0.049; the further distance from surgical lesion to language area is associated with less post-operative memory loss). Females demonstrated significantly increased postoperative seizure improvement (Fisher's p-value=0.0296; female=8; male=6). A significant association between handedness (right-handed subjects) and postoperative seizure improvement was found (p=0.02) as well as a significant trend for interaction of gender and handedness on postoperative seizure improvement (p=0.09). Overall, our results demonstrate fMRI as a useful preoperative adjunct to Wada testing for language lateralization in patients with medically intractable seizures.

Graph Theory Analysis of Functional Connectivity Combined with Machine Learning Approaches Demonstrates Widespread Network Differences and Predicts Clinical Variables in Temporal Lobe Epilepsy.

Understanding how global brain networks are affected in epilepsy may elucidate the pathogenesis of seizures and its accompanying neurobehavioral comorbidities. We investigated functional changes within neural networks in temporal lobe epilepsy (TLE) using graph theory analysis of resting-state connectivity. Twenty-seven TLE presurgical patients (age 41.0 ± 12.3 years) and 85 age, gender, and handedness equivalent healthy controls (HCs; age 39.7 ± 16.9 years) were enrolled. Eyes-closed resting-state functional magnetic resonance image scans were analyzed to compare network properties and functional connectivity (FC) changes. TLE subjects showed significantly higher global efficiency, lower clustering coefficient ratio, and lower shortest path lengths ratio than HCs, as an indication of a more synchronized, yet less segregated network. A trend of functional reorganization with a shift of network hubs to the contralateral hemisphere was noted in TLE subjects. Support vector machine (SVM) with linear kernel was trained to separate between neural networks in TLE and HC subjects based on graph measurements. SVM analysis allowed separation between TLE and HC networks with 80.66% accuracy using eight features of graph measurements. Support vector regression (SVR) was used to predict neurocognitive performance from graph metrics. An SVR linear predictor showed discriminative prediction accuracy for four key neurocognitive variables in TLE (absolute R value range: 0.61-0.75). Despite TLE, our results showed both local and global network topology differences that reflect widespread alterations in FC in TLE. Network differences are discriminative between TLE and HCs using data-driven analysis and predicted severity of neurocognitive sequelae in our cohort.

Cognitive slowing and its underlying neurobiology in temporal lobe epilepsy.

Cognitive slowing is a known but comparatively under-investigated neuropsychological complication of the epilepsies in relation to other known cognitive comorbidities such as memory, executive function and language. Here we focus on a novel metric of processing speed, characterize its relative salience compared to other cognitive difficulties in epilepsy, and explore its underlying neurobiological correlates. Research participants included 55 patients with temporal lobe epilepsy (TLE) and 58 healthy controls from the Epilepsy Connectome Project (ECP) who were administered a battery of tests yielding 14 neuropsychological measures, including selected tests from the NIH Toolbox-Cognitive Battery, and underwent 3T MRI and resting state fMRI. TLE patients exhibited a pattern of generalized cognitive impairment with very few lateralized abnormalities. Using the neuropsychological measures, machine learning (Support Vector Machine binary classification model) classified the TLE and control groups with 74% accuracy with processing speed (NIH Toolbox Pattern Comparison Processing Speed Test) the best predictor. In TLE, slower processing speed was associated predominantly with decreased local gyrification in regions including the rostral and caudal middle frontal gyrus, inferior precentral cortex, insula, inferior parietal cortex (angular and supramarginal gyri), lateral occipital cortex, rostral anterior cingulate, and medial orbital frontal regions, as well as three small regions of the temporal lobe. Slower processing speed was also associated with decreased connectivity between the primary visual cortices in both hemispheres and the left supplementary motor area, as well as between the right parieto-occipital sulcus and right middle insular area. Overall, slowed processing speed is an important cognitive comorbidity of TLE associated with altered brain structure and connectivity.

Using Low-Frequency Oscillations to Detect Temporal Lobe Epilepsy with Machine Learning.

The National Institutes of Health-sponsored Epilepsy Connectome Project aims to characterize connectivity changes in temporal lobe epilepsy (TLE) patients. The magnetic resonance imaging protocol follows that used in the Human Connectome Project, and includes 20 min of resting-state functional magnetic resonance imaging acquired at 3T using 8-band multiband imaging. Glasser parcellation atlas was combined with the FreeSurfer subcortical regions to generate resting-state functional connectivity (RSFC), amplitude of low-frequency fluctuations (ALFFs), and fractional ALFF measures. Seven different frequency ranges such as Slow-5 (0.01-0.027 Hz) and Slow-4 (0.027-0.073 Hz) were selected to compute these measures. The goal was to train machine learning classification models to discriminate TLE patients from healthy controls, and to determine which combination of the resting state measure and frequency range produced the best classification model. The samples included age- and gender-matched groups of 60 TLE patients and 59 healthy controls. Three traditional machine learning models were trained: support vector machine, linear discriminant analysis, and naive Bayes classifier. The highest classification accuracy was obtained using RSFC measures in the Slow-4 + 5 band (0.01-0.073 Hz) as features. Leave-one-out cross-validation accuracies were ∼83%, with receiver operating characteristic area-under-the-curve reaching close to 90%. Increased connectivity from right area posterior 9-46v in TLE patients contributed to the high accuracies. With increased sample sizes in the near future, better machine learning models will be trained not only to aid the diagnosis of TLE, but also as a tool to understand this brain disorder.

Psychomotor slowing is associated with anomalies in baseline and prospective large scale neural networks in youth with epilepsy.

Purpose: Psychomotor slowing is a common but understudied cognitive impairment in epilepsy. Here we test the hypothesis that psychomotor slowing is associated with alterations in brain status reflected through analysis of large scale structural networks. We test the hypothesis that children with epilepsy with cognitive slowing at diagnosis will exhibit a cross-sectional and prospective pattern of altered brain development. Methods: A total of 78 children (age 8-18) with new/recent onset idiopathic epilepsies underwent 1.5 T MRI with network analysis of cortical, subcortical and cerebellar volumes. Children with epilepsy were divided into slow and fast psychomotor speed groups (adjusted for age, intelligence and epilepsy syndrome). Results: At baseline, slow-speed performers (SSP) presented lower modularity, lower global efficiency, higher transitivity, and lower number of hubs than fast-speed performers (FSP). Community structure in SSP exhibited poor association between cortical regions and both subcortical structures and the cerebellum while FSP presented well-defined communities. Prospectively, SSP displayed lower modularity but higher global efficiency and transitivity compared to FSP. Modules in FSP showed higher integration between and within themselves compared to SSP. SSP showed hubs mainly from frontal and temporal regions while in FSP were spread among frontal, temporal, parietal, subcortical areas and the left cerebellum. Implications: Results suggest the presence of widespread alterations in large scale networks between fast- and slow-speed children with recent onset epilepsies both at baseline and 2 years later. Slower processing speed appears to be a marker of abnormal brain development antecedent to epilepsy onset as well as brain development over the 2 years following diagnosis.

Preoperative FMRI Associated with Decreased Mortality and Morbidity in Brain Tumor Patients.

BACKGROUND: Functional Magnetic Resonance Imaging (fMRI) is a presurgical planning technique used to localize functional cortex so as to maximize resection of diseased tissue and avoid viable tissue. In this retrospective study, we examined differences in morbidity and mortality of brain tumor patients who received preoperative fMRI in comparison to those who did not. METHODS: Brain tumor patients (n=206) were selected from a retrospective review of neurosurgical case logs from 2001-2009 at the University of Wisconsin-Madison. RESULTS: Univariate analysis showed improved mortality in the fMRI group and the fMRI+Electrical Cortical Stimulation Mapping (ECM) group compared to the No-fMRI group. Multivariate analyses showed improved mortality of the fMRI group and the fMRI+ECM group compared to the No-fMRI group, with age and tumor grade being the most significant influencers. Overall, the fMRI group showed survival benefits at 3 years; twice that of the No-fMRI group. Furthermore, patients with high-grade tumors showed significant survival benefits in the fMRI group, while patients with low-grade tumors did not (controlling for age and ECM). There was also a significant difference in the two groups with respect to morbidity, with patients receiving fMRI showing improved outcomes in the motor and language domains. CONCLUSIONS: This study analyzing a large retrospective series of brain tumor patients with and without the use of fMRI in the preoperative planning has resulted in improved mortality and morbidity outcomes with the use of fMRI. These results point to the importance of incorporating fMRI in presurgical planning in the clinical management of patients with brain tumors.

Predicting primary outcomes of brain tumor patients with advanced neuroimaging MRI measures.

BACKGROUND: Advanced neuroimaging measures along with clinical variables acquired during standard imaging protocols provide a rich source of information for brain tumor patient treatment and management. Machine learning analysis has had much recent success in neuroimaging applications for normal and patient populations and has potential, specifically for brain tumor patient outcome prediction. The purpose of this work was to construct, using the current patient population distribution, a high accuracy predictor for brain tumor patient outcomes of mortality and morbidity (i.e., transient and persistent language and motor deficits). The clinical value offered is a statistical tool to help guide treatment and planning as well as an investigation of the influential factors of the disease process. METHODS: Resting state fMRI, diffusion tensor imaging, and task fMRI data in combination with clinical and demographic variables were used to represent the tumor patient population (n = 62; mean age = 51.2 yrs.) in a machine learning analysis in order to predict outcomes. RESULTS: A support vector machine classifier with a t-test filter and recursive feature elimination predicted patient mortality (18-month interval) with 80.7% accuracy, language deficits (transient) with 74.2%, motor deficits with 71.0%, language outcomes (persistent) with 80.7% and motor outcomes with 83.9%. The most influential features of the predictors were resting fMRI connectivity, and fractional anisotropy and mean diffusivity measures in the internal capsule, brain stem and superior and inferior longitudinal fasciculi. CONCLUSIONS: This study showed that advanced neuroimaging data with machine learning methods can potentially predict patient outcomes and reveal influential factors driving the predictions.

Characterizing the relationship between functional MRI-derived measures and clinical outcomes in patients with vascular lesions.

OBJECT: Functional MRI (fMRI) has proven to be an effective component of pretreatment planning in patients harboring a variety of different brain lesions. The authors have recently reported significant relationships concerning distances between brain tumor borders and areas of functional activation (lesion-to-activation distance; LAD) with regard to patient morbidity and mortality. This study further examines the relationship between LAD, focusing on a host of vascular lesions and pre- and posttreatment morbidity. METHODS: This study included a sample population of patients with vascular lesions (n = 106), primarily arteriovenous malformations (AVMs) and cavernomas. These patients underwent pretreatment fMRI-based motor mapping (n = 72) or language mapping (n = 84). The impact of LAD and other variables derived from the patient medical record were analyzed with respect to functional deficits in terms of morbidity (weakness and/or aphasia). RESULTS: In patients with no pretreatment deficits, there was trend for a significant relationship between the Wernicke area LAD and posttreatment language deficits. In patients with or without pretreatment deficits, a trend toward significance was observed between sensorimotor LAD and posttreatment motor deficits. Additionally, lesion type (AVMs or cavernomas) affected posttreatment deficits, with more patients with cavernomas showing posttreatment language deficits than patients with AVMs. However, this difference was not observed for posttreatment motor deficits. CONCLUSIONS: These findings suggest that the proximity of a vascular lesion to sensorimotor and language areas is a relevant parameter in estimating patient prognosis in the perioperative period. Additionally, vascular lesion type and existence of pretreatment deficits play a significant role in outcomes.

Chronic neurological deficits in mice after perinatal hypoxia and ischemia correlate with hemispheric tissue loss and white matter injury detected by MRI.

We investigated the effects of perinatal hypoxia-ischemia (HI) on brain injury and neurological functional outcome at postnatal day (P)30 through P90. HI was induced by exposing P9 mice to 8% O(2) for 55 min using the Vannucci HI model. Following HI, mice were treated with either vehicle control or Na(+)/H(+) exchanger isoform 1 (NHE1) inhibitor HOE 642. The animals were examined by the accelerating rotarod test at P30 and the Morris water maze (MWM) test at P60. T(2)-weighted MRI was conducted at P90. Diffusion tensor imaging (DTI) was subsequently performed in ex vivo brains, followed by immunohistochemical staining for changes in myelin basic protein (MBP) and neurofilament protein expression in the corpus callosum (CC). Animals at P30 after HI showed deficits in motor and spatial learning. T(2) MRI detected a wide spectrum of brain injury in these animals. A positive linear correlation was observed between learning deficits and the degree of tissue loss in the ipsilateral hemisphere and hippocampus. Additionally, CC DTI fractional anisotropy (FA) values correlated with MBP expression. Both FA and MBP values correlated with performance on the MWM test. HOE 642-treated mice exhibited improved spatial learning and memory, and less white matter injury in the CC. These findings suggest that HI-induced cerebral atrophy and CC injury contribute to the development of deficits in learning and memory, and that inhibition of NHE1 is neuroprotective in part by reducing white matter injury. T(2)-weighted MRI and DTI are useful indicators of functional outcome after perinatal HI.

Children with new-onset epilepsy exhibit diffusion abnormalities in cerebral white matter in the absence of volumetric differences.

The purpose of this investigation was to examine the diffusion properties of cerebral white matter in children with recent onset epilepsy (n=19) compared to healthy controls (n=11). Subjects underwent DTI with quantification of mean diffusion (MD), fractional anisotropy (FA), axial diffusivity (D(ax)) and radial diffusivity (D(rad)) for regions of interest including anterior and posterior corpus callosum, fornix, cingulum, and internal and external capsules. Quantitative volumetrics were also performed for the corpus callosum and its subregions (anterior, midbody and posterior) and total lobar white and gray matter for the frontal, parietal, temporal and occipital lobes. The results demonstrated no group differences in total lobar gray or white matter volumes or volume of the corpus callosum and its subregions, but did show reduced FA and increased D(rad) in the posterior corpus callosum and cingulum. These results provide the earliest indication of microstructural abnormality in cerebral white matter among children with idiopathic epilepsies. This abnormality occurs in the context of normal volumetrics and suggests disruption in myelination processes.

Accuracy of an automated method to measure rotations of vertebrae from computerized tomography data.

STUDY DESIGN: In this phantom study, rotations of a vertebral body calculated from computerized tomography (CT) were compared to the actual rotations provided by a specially designed device incorporating a reduction gear. OBJECTIVE: The objective was to measure the accuracy of the CT and an automated software program to calculate rotations of lumbar vertebral bodies. BACKGROUND: Rotations of individual vertebrae secondary to a change in position or load can be measured in select patients by roentgen stereophotogrammetry or by using CT and a specially constructed table that creates the rotation of the torso. The purpose of this study was to measure the precision and accuracy of rotation measurements made with a CT scanner and an automated program to calculate rotation. METHODS: We constructed a phantom with a lumbar vertebra that can be rotated within a CT scanner. CT of the vertebra were obtained at angular positions of 0, 0.360 degrees , 1.080 degrees , 2.520 degrees , 5.400 degrees , 11.160 degrees , 29.160 degrees , 29.340 degrees , 29.520 degrees , 29.610 degrees , 29.700 degrees , 29.790 degrees , 29.880 degrees , and 29.889 degrees . With an automated program based on a pixel-shift algorithm, we calculated rotations of the vertebra between pairs of images. Accuracy was calculated as mean difference between the actual and the calculated rotation, and precision was calculated as the standard deviation of the differences. RESULTS: Differences between actual and calculated rotations varied from -0.083 degrees to 0.132 degrees . For rotations less than 15 degrees , mean error (accuracy) was -0.039 degrees , and the standard deviation (precision) was 0.029 degrees . For rotations greater than 15 degrees , the accuracy was 0.086 degrees , and the precision was 0.023 degrees . CONCLUSIONS: This study shows that rotations of lumbar vertebrae may be measured with CT, and an automated program to an accuracy and precision better than 0.1 degrees , comparable to that of roentgen stereophotogrammetry.

Ventral medulla pHi measured in vivo by 31P NMR is not regulated during hypercapnia in anesthetized rat.

Chemoreceptors in the ventral medulla contribute to the respiratory response to hypercapnia. Do they 'sense' intracellular pH (pHi)? We measured pHi in the ventral medulla or cortex (control) using 31P-NMR obtained via a novel 3 x 5 mm2 surface coil in anesthetized rats breathing air or 7% CO2. During air breathing over 240 min, pHi decreased slightly from 7.13 +/- 0.02 to 7.05 +/- 0.02 (SEM; n = 5; 2 cortex, 3 ventral medulla). During 180 min of hypercapnia, cortical pHi (n = 4) decreased from 7.17 +/- 0.02 to 6.87 +/- 0.01 by 90 min and recovered by 150 min. Ventral medulla pHi showed no such regulation. It decreased from 7.11 +/- 0.02 to 6.88 +/- 0.02 at 90 min and recovered only after cessation of hypercapnia (n = 5), results consistent with pHi being the chemoreceptor stimulus. However, non-chemoreceptor neurons that contribute to our medullary NMR signal also do not appear to regulate pHi in vitro. Regional differences in pHi regulation between cortex and ventral medulla may be due to both chemosensitive and non-chemosensitive neurons.