Lesion network mapping predicts post-stroke behavioural deficits and improves localization

Frequency, phenotypes, and neuroimaging of early post stroke movement disorders: a prospective study

Background

A prospective observational study recruited patients with acute stroke. Patients were assessed for the presence of post-stroke movement disorders PSMDs during the first week of stroke. This study aimed to identify the frequency, clinical characteristics, and neuroimaging of early PSMDs (within the first week) and followed for 1 year.

Results

A total of 600 patients were recruited; 21 patients (3.5%) with PSMDs were detected. Thirteen (2.2%) patients presented with intention tremor/ataxia and eight (1.3%) presented with other movement disorders (most commonly, chorea and tremor). One patient presented with periodic left upper limb shaking with right subcortical watershed infarction, and one patient developed palatal myoclonus with right middle cerebral artery infarction. Patients with PSMDs had significantly lower stroke severity (NIHSS) and were more likely to have lacunar strokes (p < 0.001 and < 0.006, respectively) than patients without PSMDs. Early PSMDs were more associated with posterior circulation strokes (84.25%).

Conclusions

Early PSMDs are commonly hyperkinetic, more associated with small vessel disease, and less severe and posterior circulation strokes, implying their clinical importance for the proper management of stroke patients.

Post-stroke outcome prediction based on lesion-derived features

Stroke-induced deficits result from both focal structural damage and widespread network disruption. This study investigated whether simulated measures of network disruption, derived from structural lesions, could predict functional impairments in stroke patients. We extracted four lesion-derived feature sets: lesion masks, probabilistic structural disconnection maps (pSDMs), structural and indirectly estimated functional connectivity strengths between brain regions, and topological properties of functional and structural brain networks to predict motor, executive, and processing speed deficits in 340 S patients, employing PCA-based ridge regression with leave-one-out cross validation. The findings revealed that both structural disconnection map patterns and lesion masks were strong predictors of functional deficits. Lesion masks exhibited superior predictive performance relative to unthresholded pSDMs. Furthermore, applying a probability threshold to the pSDMs - retaining only disconnections present in a sufficient proportion of healthy subjects - significantly improved predictive performance. For motor deficits, thresholded SDMs (tSDMs) with thresholds of 0.9 and 0.5 produced the highest R2 values, 0.95 for left motor deficits and 0.69 for right motor deficits, respectively. In the case of executive function and processing speed, the highest R2 values were 0.58 and 0.64, achieved with tSDM thresholds of 0.3 and 0.5, respectively. Connectome-based features exhibited lower R2 values, with structural connection strength alterations showing stronger associations with post-stroke scores compared to changes in functional connectivity. Nodal parameters (degree and clustering coefficient) had lower explanatory power than the SDM features and lesion masks. Our findings underscore the effectiveness of lesion masks and thresholded SDMs in predicting post-stroke deficits. This study contributes to the growing body of evidence supporting the reliability of simulated structural networks as a complementary approach to lesion patterns and structural disconnection in predicting post-stroke outcomes.

Effectively capturing memory deficits: New advances in the use of the In-out-Test for cognitive Test post-stroke

OBJECTIVE

This study evaluated the reliability and validity of the In-Out-Test for detecting episodic memory deficits in stroke patients and explored its potential as a clinical test.

METHODS

A total of 75 stroke patients and 120 healthy controls underwent tests, including the Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), Picture-Based Memory Impairment Screen (PMIS), and In-Out-Test. Reliability metrics (Cronbach's α, inter-scorer reliability, test-retest reliability), criterion validity, corrected item-total correlation, hierarchical regression analysis and ROC curve analysis were performed to determine the sensitivity and specificity of the In-Out-Test.

RESULTS

Stroke patients scored lower across all tests (p < 0.001), with the largest difference in the In-Out-Test (d = 0.99). The In-Out-Test correlated strongly with other cognitive tests (r = 0.79-0.85 in stroke patients; r = 0.66-0.78 in controls). It explained an additional 4.5% of variance in MoCA-MIS scores (p < 0.001). Reliability was high (Cronbach's α = 0.835; inter-rater ICCs = 0.911-0.925; test-retest ICCs = 0.764-0.802). ROC analysis showed an AUC of 0.747, with a sensitivity of 0.708 and specificity of 0.680 at a cutoff of 10.5.

CONCLUSION

Preliminary findings indicated that the In-Out-Test showed potential in detecting episodic memory impairments in stroke patients, warranting further validation in larger cohorts.

Unraveling the influences of hemodynamic lag and intrinsic cerebrovascular reactivity on functional metrics in ischemic stroke

Resting-state functional magnetic resonance imaging (rs-fMRI) is a prominent tool for investigating functional deficits in stroke patients. However, the extent to which the hemodynamic lags (LAG) and the intrinsic cerebrovascular reactivity (iCVR) may affect the rs-fMRI metrics in different scales needs to be clarified for ischemic stroke. In this study, 73 ischemic stroke patients and 74 healthy controls (HC) were recruited to investigate how the correction of the LAG and/or iCVR would influence resting-state functional magnetic resonance imaging (rs-fMRI) metrics of three different spatial scales (local-scale, meso-scale and global-scale) in ischemic stroke. The analysis revealed that the Stroke pattern of all functional metrics using different correction strategies resembled the HC pattern. The highest overlap was observed in the Stroke pattern with correction for both LAG and iCVR, while the pattern without correction showed the lowest overlap. Most functional metrics after correction showed higher sensitivity in detecting between-group differences than those without correction. Moreover, our results were generally reproducible in an independent dataset. Collectively, these findings emphasize the necessity of considering LAG and iCVR effects to investigate stroke-related functional alterations, and highlight the significance of correction strategies for accurately interpreting the findings in rs-fMRI study of ischemic stroke.

Prediction of stroke severity: systematic evaluation of lesion representations

OBJECTIVE

To systematically evaluate which lesion-based imaging features and methods allow for the best statistical prediction of poststroke deficits across independent datasets.

METHODS

We utilized imaging and clinical data from three independent datasets of patients experiencing acute stroke (N1 = 109, N2 = 638, N3 = 794) to statistically predict acute stroke severity (NIHSS) based on lesion volume, lesion location, and structural and functional disconnection with the lesion location using normative connectomes.

RESULTS

We found that prediction models trained on small single-center datasets could perform well using within-dataset cross-validation, but results did not generalize to independent datasets (median R2 N1 = 0.2%). Performance across independent datasets improved using large single-center training data (R2 N2 = 15.8%) and improved further using multicenter training data (R2 N3 = 24.4%). These results were consistent across lesion attributes and prediction models. Including either structural or functional disconnection in the models outperformed prediction based on volume or location alone (P < 0.001, FDR-corrected).

INTERPRETATION

We conclude that (1) prediction performance in independent datasets of patients with acute stroke cannot be inferred from cross-validated results within a dataset, as performance results obtained via these two methods differed consistently, (2) prediction performance can be improved by training on large and, importantly, multicenter datasets, and (3) structural and functional disconnection allow for improved prediction of acute stroke severity.

Hayling and stroop tests tap dissociable deficits and network-level neural correlates

Although many executive function screens have been developed, it is not yet clear whether these assessments are equally effective in detecting post-stroke deficits of initiation and inhibition. This study presents a comparative analysis of the Stroop and Hayling tests aiming to evaluate whether these tests measure the same underlying cognitive functions and to identify the neural correlates of the deficits detected by both tasks. Sixty six stroke survivors and 70 healthy ageing controls completed the Hayling and Stroop tests. Stroke patients were found to exhibit qualitative performance differences across analogous Stroop and Hayling Test metrics intended to tap initiation and inhibition. The Stroop test was found to have high specificity to abnormal performance, but low sensitivity relative to the Hayling Test. Minimal overlap was present between the network-level correlates of analogous Stroop and Hayling Test metrics. Hayling Task strategy use metrics were significantly associated with distinct patterns of disconnection in stroke survivors, providing novel insight into the neural correlates of fine-grained behavioural patterns. Overall, these findings strongly suggest that the functions tapped by the Stroop and Hayling Test are both behaviourally and anatomically dissociable. The Hayling Test was found to offer improved sensitivity and detail relative to the Stroop test. This novel demonstration of the Hayling Test within the stroke population suggests that this task represents an effective measure for quantifying post-stroke initiation and inhibition deficits.

Common and distinct patterns underlying different linguistic tasks: multivariate disconnectome symptom mapping in poststroke patients

Numerous studies have been devoted to neural mechanisms of a variety of linguistic tasks (e.g. speech comprehension and production). To date, however, whether and how the neural patterns underlying different linguistic tasks are similar or differ remains elusive. In this study, we compared the neural patterns underlying 3 linguistic tasks mainly concerning speech comprehension and production. To address this, multivariate regression approaches with lesion/disconnection symptom mapping were applied to data from 216 stroke patients with damage to the left hemisphere. The results showed that lesion/disconnection patterns could predict both poststroke scores of speech comprehension and production tasks; these patterns exhibited shared regions on the temporal pole of the left hemisphere as well as unique regions contributing to the prediction for each domain. Lower scores in speech comprehension tasks were associated with lesions/abnormalities in the superior temporal gyrus and middle temporal gyrus, while lower scores in speech production tasks were associated with lesions/abnormalities in the left inferior parietal lobe and frontal lobe. These results suggested an important role of the ventral and dorsal stream pathways in speech comprehension and production (i.e. supporting the dual stream model) and highlighted the applicability of the novel multivariate disconnectome-based symptom mapping in cognitive neuroscience research.

Functional and structural lesion network mapping in neurological and psychiatric disorders: a systematic review

Background

The traditional approach to studying the neurobiological mechanisms of brain disorders and localizing brain function involves identifying brain abnormalities and comparing them to matched controls. This method has been instrumental in clinical neurology, providing insight into the functional roles of different brain regions. However, it becomes challenging when lesions in diverse regions produce similar symptoms. To address this, researchers have begun mapping brain lesions to functional or structural networks, a process known as lesion network mapping (LNM). This approach seeks to identify common brain circuits associated with lesions in various areas. In this review, we focus on recent studies that have utilized LNM to map neurological and psychiatric symptoms, shedding light on how this method enhances our understanding of brain network functions.

Methods

We conducted a systematic search of four databases: PubMed, Scopus, and Web of Science, using the term "Lesion network mapping." Our focus was on observational studies that applied lesion network mapping in the context of neurological and psychiatric disorders.

Results

Following our screening process, we included 52 studies, comprising a total of 6,814 subjects, in our systematic review. These studies, which utilized functional connectivity, revealed several regions and network overlaps across various movement and psychiatric disorders. For instance, the cerebellum was found to be part of a common network for conditions such as essential tremor relief, parkinsonism, Holmes tremor, freezing of gait, cervical dystonia, infantile spasms, and tics. Additionally, the thalamus was identified as part of a common network for essential tremor relief, Holmes tremor, and executive function deficits. The dorsal attention network was significantly associated with fall risk in elderly individuals and parkinsonism.

Conclusion

LNM has proven to be a powerful tool in localizing a broad range of neuropsychiatric, behavioral, and movement disorders. It holds promise in identifying new treatment targets through symptom mapping. Nonetheless, the validity of these approaches should be confirmed by more comprehensive prospective studies.

Linking Personalized Brain Atrophy to Schizophrenia Network and Treatment Response

BACKGROUND AND HYPOTHESIS

Schizophrenia manifests with marked heterogeneity in both clinical presentation and underlying biology. Modeling individual differences within clinical cohorts is critical to translate knowledge reliably into clinical practice. We hypothesized that individualized brain atrophy in patients with schizophrenia may explain the heterogeneous outcomes of repetitive transcranial magnetic stimulation (rTMS).

STUDY DESIGN

The magnetic resonance imaging (MRI) data of 797 healthy subjects and 91 schizophrenia patients (between January 1, 2015, and December 31, 2020) were retrospectively selected from our hospital database. The healthy subjects were used to establish normative reference ranges for cortical thickness as a function of age and sex. Then, a schizophrenia patient's personalized atrophy map was computed as vertex-wise deviations from the normative model. Each patient's atrophy network was mapped using resting-state functional connectivity MRI from a subgroup of healthy subjects (n = 652). In total 52 of the 91 schizophrenia patients received rTMS in a randomized clinical trial (RCT). Their longitudinal symptom changes were adopted to test the clinical utility of the personalized atrophy map.

RESULTS

The personalized atrophy maps were highly heterogeneous across patients, but functionally converged to a putative schizophrenia network that comprised regions implicated by previous group-level findings. More importantly, retrospective analysis of rTMS-RCT data indicated that functional connectivity of the personalized atrophy maps with rTMS targets was significantly associated with the symptom outcomes of schizophrenia patients.

CONCLUSIONS

Normative modeling can aid in mapping the personalized atrophy network associated with treatment outcomes of patients with schizophrenia.

Mapping lesion, structural disconnection, and functional disconnection to symptoms in semantic aphasia

Patients with semantic aphasia have impaired control of semantic retrieval, often accompanied by executive dysfunction following left hemisphere stroke. Many but not all of these patients have damage to the left inferior frontal gyrus, important for semantic and cognitive control. Yet semantic and cognitive control networks are highly distributed, including posterior as well as anterior components. Accordingly, semantic aphasia might not only reflect local damage but also white matter structural and functional disconnection. Here, we characterise the lesions and predicted patterns of structural and functional disconnection in individuals with semantic aphasia and relate these effects to semantic and executive impairment. Impaired semantic cognition was associated with infarction in distributed left-hemisphere regions, including in the left anterior inferior frontal and posterior temporal cortex. Lesions were associated with executive dysfunction within a set of adjacent but distinct left frontoparietal clusters. Performance on executive tasks was also associated with interhemispheric structural disconnection across the corpus callosum. In contrast, poor semantic cognition was associated with small left-lateralized structurally disconnected clusters, including in the left posterior temporal cortex. Little insight was gained from functional disconnection symptom mapping. These results demonstrate that while left-lateralized semantic and executive control regions are often damaged together in stroke aphasia, these deficits are associated with distinct patterns of structural disconnection, consistent with the bilateral nature of executive control and the left-lateralized yet distributed semantic control network.

Lesion network mapping for symptom localization: recent developments and future directions

PURPOSE OF REVIEW

Focal lesions causing specific neurological or psychiatric symptoms can occur in multiple different brain locations, complicating symptom localization. Here, we review lesion network mapping, a technique used to aid localization by mapping lesion-induced symptoms to brain circuits rather than individual brain regions. We highlight recent examples of how this technique is being used to investigate clinical entities and identify therapeutic targets.

RECENT FINDINGS

To date, lesion network mapping has successfully been applied to more than 40 different symptoms or symptom complexes. In each case, lesion locations were combined with an atlas of human brain connections (the human connectome) to map heterogeneous lesion locations causing the same symptom to a common brain circuit. This approach has lent insight into symptoms that have been difficult to localize using other techniques, such as hallucinations, tics, blindsight, and pathological laughter and crying. Further, lesion network mapping has recently been applied to lesions that improve symptoms, such as tremor and addiction, which may translate into new therapeutic targets.

SUMMARY

Lesion network mapping can be used to map lesion-induced symptoms to brain circuits rather than single brain regions. Recent findings have provided insight into long-standing clinical mysteries and identified testable treatment targets for circuit-based and symptom-based neuromodulation.

Mapping correlated neurological deficits after stroke to distributed brain networks

Understanding the relationships between brain organization and behavior is a central goal of neuroscience. Traditional teaching emphasizes that the human cerebrum includes many distinct areas for which damage or dysfunction would lead to a unique and specific behavioral syndrome. This teaching implies that brain areas correspond to encapsulated modules that are specialized for specific cognitive operations. However, empirically, local damage from stroke more often produces one of a small number of clusters of deficits and disrupts brain-wide connectivity in a small number of predictable ways (relative to the vast complexity of behavior and brain connectivity). Behaviors that involve shared operations show correlated deficits following a stroke, consistent with a low-dimensional behavioral space. Because of the networked organization of the brain, local damage from a stroke can result in widespread functional abnormalities, matching the low dimensionality of behavioral deficit. In alignment with this, neurological disease, psychiatric disease, and altered brain states produce behavioral changes that are highly correlated across a range of behaviors. We discuss how known structural and functional network priors in addition to graph theoretical concepts such as modularity and entropy have provided inroads to understanding this more complex relationship between brain and behavior. This model for brain disease has important implications for normal brain-behavior relationships and the treatment of neurological and psychiatric diseases.

Lesion network mapping of mania using different normative connectomes

Lesion network mapping is a neuroimaging technique that explores the network of regions functionally connected to lesions causing a common syndrome. The technique uses resting state functional connectivity from large databases of healthy individuals, i.e., connectomes, and has allowed for important insight into the potential network mechanisms underlying several neuropsychiatric disorders. However, concerns regarding reproducibility have arisen, that may be due to the use of different connectomes, with variable MRI acquisition parameters and preprocessing methods. Here, we tested the impact of using different connectomes on the results of lesion network mapping for mania. We found results were reliable and consistent independent of the connectome used.

Untapped Neuroimaging Tools for Neuro-Oncology: Connectomics and Spatial Transcriptomics

Simple Summary

Brain imaging, specifically magnetic resonance imaging (MRI), plays a key role in the clinical and research aspects of neuro-oncology. Novel neuroimaging techniques enable the transformation of a brain MRI into a so-called average brain. This allows projects using already acquired brain MRIs to perform group analyses and draw conclusions. Once the data are in this average brain, several types of analyses can be performed. For example, determining the most vulnerable locations for certain tumor types or perhaps even the underlying circuitry and gene expression that might cause predisposition to tumor growth. This information may further our understanding of tumor behavior, leading to better patient counseling, surgery timing, and treatment monitoring.

Abstract

Neuro-oncology research is broad and includes several branches, one of which is neuroimaging. Magnetic resonance imaging (MRI) is instrumental for the diagnosis and treatment monitoring of patients with brain tumors. Most commonly, structural and perfusion MRI sequences are acquired to characterize tumors and understand their behaviors. Thanks to technological advances, structural brain MRI can now be transformed into a so-called average brain accounting for individual morphological differences, which enables retrospective group analysis. These normative analyses are uncommonly used in neuro-oncology research. Once the data have been normalized, voxel-wise analyses and spatial mapping can be performed. Additionally, investigations of underlying connectomics can be performed using functional and structural templates. Additionally, a recently available template of spatial transcriptomics has enabled the assessment of associated gene expression. The few published normative analyses have shown relationships between tumor characteristics and spatial localization, as well as insights into the circuitry associated with epileptogenic tumors and depression after cingulate tumor resection. The wide breadth of possibilities with normative analyses remain largely unexplored, specifically in terms of connectomics and imaging transcriptomics. We provide a framework for performing normative analyses in oncology while also highlighting their limitations. Normative analyses are an opportunity to address neuro-oncology questions from a different perspective.

Lesion symptom mapping of domain-specific cognitive impairments using routine imaging in stroke

INTRODUCTION

This large-scale lesion-symptom mapping study investigates the necessary neuro-anatomical substrates of 5 cognitive domains frequently affected post stroke: Language, Attention, Praxis, Number, and Memory. This study aims to demonstrate the validity of using routine clinical brain imaging and standard bedside cognitive screening data from a large, real-world patient cohort for lesion-symptom mapping.

PATIENTS AND METHODS

Behavioural cognitive screening data from the Oxford Cognitive Screen and routine clinical neuroimaging from 573 acute patients was used in voxel-based lesion-symptom mapping analyses. Patients were classed as impaired or not on each of the subtests within 5 cognitive domains.

RESULTS

Distinct patterns of lesion damage were associated with different domains. Language functions were associated with damage to left hemisphere fronto-temporal areas. Visuo-spatial functions were associated with damage to posterior occipital areas (Visual Field) and the right temporo-parietal region (Visual Neglect). Different memory impairments were linked to distinct voxel clusters within the left insular and opercular cortices. Deficits which were not associated with localised voxels (e.g. executive function, praxis) represent distributed, bilateral functions.

DISCUSSION

The standardised, brief Oxford Cognitive Screen was able to reliably differentiate distinct neural correlates critically involved in supporting domain-specific cognitive abilities.

CONCLUSION

By demonstrating and replicating known brain anatomy correlates within real-life clinical cohorts using routinely collected scans and standard bedside screens, we open up VLSM techniques to a wealth of clinically relevant studies which can capitalise on using existing clinical data.

A novel stroke lesion network mapping approach: improved accuracy yet still low deficit prediction

Lesion network mapping estimates functional network abnormalities caused by a focal brain lesion. The method requires embedding the volume of the lesion into a normative functional connectome and using the average functional magnetic resonance imaging signal from that volume to compute the temporal correlation with all other brain locations. Lesion network mapping yields a map of potentially functionally disconnected regions. Although promising, this approach does not predict behavioural deficits well. We modified lesion network mapping by using the first principal component of the functional magnetic resonance imaging signal computed from the voxels within the lesioned area for temporal correlation. We measured potential improvements in connectivity strength, anatomical specificity of the lesioned network and behavioural prediction in a large cohort of first-time stroke patients at 2-weeks post-injury (n = 123). This principal component functional disconnection approach localized mainly cortical voxels of high signal-to-noise; and it yielded networks with higher anatomical specificity, and stronger behavioural correlation than the standard method. However, when examined with a rigorous leave-one-out machine learning approach, principal component functional disconnection approach did not perform better than the standard lesion network mapping in predicting neurological deficits. In summary, even though our novel method improves the specificity of disconnected networks and correlates with behavioural deficits post-stroke, it does not improve clinical prediction. Further work is needed to capture the complex adjustment of functional networks produced by focal damage in relation to behaviour.

Post-Stroke Cognitive Impairment: Pathophysiological Insights into Brain Disconnectome from Advanced Neuroimaging Analysis Techniques

The neurological symptoms of stroke have traditionally provided the foundation for functional mapping of the brain. However, there are many unresolved aspects in our understanding of cerebral activity, especially regarding high-level cognitive functions. This review provides a comprehensive look at the pathophysiology of post-stroke cognitive impairment in light of recent findings from advanced imaging techniques. Combining network neuroscience and clinical neurology, our research focuses on how changes in brain networks correlate with post-stroke cognitive prognosis. More specifically, we first discuss the general consequences of stroke lesions due to damage of canonical resting-state large-scale networks or changes in the composition of the entire brain. We also review emerging methods, such as lesion-network mapping and gradient analysis, used to study the aforementioned events caused by stroke lesions. Lastly, we examine other patient vulnerabilities, such as superimposed amyloid pathology and blood-brain barrier leakage, which potentially lead to different outcomes for the brain network compositions even in the presence of similar stroke lesions. This knowledge will allow a better understanding of the pathophysiology of post-stroke cognitive impairment and provide a theoretical basis for the development of new treatments, such as neuromodulation.

Estimated connectivity networks outperform observed connectivity networks when classifying people with multiple sclerosis into disability groups

BACKGROUND

Multiple Sclerosis (MS), a neurodegenerative and neuroinflammatory disease, causing lesions that disrupt the brain's anatomical and physiological connectivity networks, resulting in cognitive, visual and/or motor disabilities. Advanced imaging techniques like diffusion and functional MRI allow measurement of the brain's structural connectivity (SC) and functional connectivity (FC) networks, and can enable a better understanding of how their disruptions cause disability in people with MS (pwMS). However, advanced MRI techniques are used mainly for research purposes as they are expensive, time-consuming and require high-level expertise to acquire and process. As an alternative, the Network Modification (NeMo) Tool can be used to estimate SC and FC using lesion masks derived from pwMS and a reference set of controls' connectivity networks.

OBJECTIVE

Here, we test the hypothesis that estimated SC and FC (eSC and eFC) from the NeMo Tool, based only on an individual's lesion masks, can be used to classify pwMS into disability categories just as well as SC and FC extracted from advanced MRI directly in pwMS. We also aim to find the connections most important for differentiating between no disability vs evidence of disability groups.

MATERIALS AND METHODS

One hundred pwMS (age:45.5 ± 11.4 years, 66% female, disease duration: 12.97 ± 8.07 years) were included in this study. Expanded Disability Status Scale (EDSS) was used to assess disability, 67 pwMS had no disability (EDSS < 2). Observed SC and FC were extracted from diffusion and functional MRI directly in pwMS, respectively. The NeMo Tool was used to estimate the remaining structural connectome (eSC), by removing streamlines in a reference set of tractograms that intersected the lesion mask. The NeMo Tool's eSC was used then as input to a deep neural network to estimate the corresponding FC (eFC). Logistic regression with ridge regularization was used to classify pwMS into disability categories (no disability vs evidence of disability), based on demographics/clinical information (sex, age, race, disease duration, clinical phenotype, and spinal lesion burden) and either pairwise entries or regional summaries from one of the following matrices: SC, FC, eSC, and eFC. The area under the ROC curve (AUC) was used to assess the classification performance. Both univariate statistics and parameter coefficients from the classification models were used to identify features important to differentiating between the groups.

RESULTS

The regional eSC and eFC models outperformed their observed FC and SC counterparts (p-value < 0.05), while the pairwise eSC and SC performed similarly (p = 0.10). Regional eSC and eFC models had higher AUC (0.66-0.68) than the pairwise models (0.60-0.65), with regional eFC having highest classification accuracy across all models. Ridge regression coefficients for the regional eFC and regional observed FC models were significantly correlated (Pearson's r = 0.52, p-value < 10e-7). Decreased estimated SC node strength in default mode and ventral attention networks and increased eFC node strength in visual networks was associated with evidence of disability.

DISCUSSION

Here, for the first time, we use clinically acquired lesion masks to estimate both structural and functional connectomes in patient populations to better understand brain lesion-dysfunction mapping in pwMS. Models based on the NeMo Tool's estimates of SC and FC better classified pwMS by disability level than SC and FC observed directly in the individual using advanced MRI. This work provides a viable alternative to performing high-cost, advanced MRI in patient populations, bringing the connectome one step closer to the clinic.