Endovascular Thrombectomy Alone for Large Vessel Occlusion: A Cost-Effectiveness Evaluation Based on Meta-Analyses.

BACKGROUND: The benefit of intravenous thrombolysis with alteplase before endovascular thrombectomy (EVT) for acute ischemic stroke due to large vessel occlusion remains debated. In this study, we analyzed the cost-effectiveness of EVT alone versus intravenous alteplase before EVT in patients directly admitted to EVT-capable stroke centers from the Dutch health care payer perspective. METHODS: A decision analysis was performed using a Markov model with 15-year simulated follow-up to estimate total costs, quality-adjusted life years, and an incremental cost-effectiveness ratio of intravenous alteplase before EVT compared with EVT alone. A hypothetical cohort of 10 000 patients with large vessel occlusion aged 70 years was run in Monte Carlo simulation. Functional outcome of each treatment was derived from pooled results of 6 randomized controlled trials (RCTs). Uncertainty was assessed by probabilistic analyses, scenario analyses, and 1-way sensitivity analyses. RESULTS: Using functional outcomes obtained from 6 RCTs (intention-to-treat population), intravenous alteplase before EVT resulted in 0.05 quality-adjusted life years gained at an additional $2817 compared with EVT alone, resulting in the incremental cost-effectiveness ratio of $62 287. Probabilistic analyses showed that intravenous alteplase before EVT had a probability of 45% and 54%, respectively, of being cost-effective at the $52 500 and $84 000 thresholds. Restricting functional outcomes from our post hoc modified as-treated analysis of 6 RCTs (scenario 1), European RCTs (scenario 2), or a Dutch RCT (scenario 3), intravenous alteplase before EVT was cost-effective in 64%, 81%, and 50% of simulations at the $52 500 threshold, and 79%, 91%, and 67% of simulations at the $84 000 threshold. CONCLUSIONS: Intravenous alteplase before EVT was not cost-effective in patients with large vessel occlusion in the Netherlands at the $52 500 threshold but possibly cost-effective at the $84 000 threshold. Variable functional outcomes at 3 months based on different trial populations affected the cost-effectiveness of intravenous alteplase before EVT.

Multi-dimensional predictors of first drinking initiation and regular drinking onset in adolescence: A prospective longitudinal study.

Early adolescent drinking onset is linked to myriad negative consequences. Using the National Consortium on Alcohol and NeuroDevelopment in Adolescence (NCANDA) baseline to year 8 data, this study (1) leveraged best subsets selection and Cox Proportional Hazards regressions to identify the most robust predictors of adolescent first and regular drinking onset, and (2) examined the clinical utility of drinking onset in forecasting later binge drinking and withdrawal effects. Baseline predictors included youth psychodevelopmental characteristics, cognition, brain structure, family, peer, and neighborhood domains. Participants (N=538) were alcohol-naïve at baseline. The strongest predictors of first and regular drinking onset were positive alcohol expectancies (Hazard Ratios [HRs]=1.67-1.87), easy home alcohol access (HRs=1.62-1.67), more parental solicitation (e.g., inquiring about activities; HRs=1.72-1.76), and less parental control and knowledge (HRs=.72-.73). Robust linear regressions showed earlier first and regular drinking onset predicted earlier transition into binge and regular binge drinking (ßs=0.57-0.95). Zero-inflated Poisson regressions revealed that delayed first and regular drinking increased the likelihood (Incidence Rate Ratios [IRR]=1.62 and IRR=1.29, respectively) of never experiencing withdrawal. Findings identified behavioral and environmental factors predicting temporal paths to youthful drinking, dissociated first from regular drinking initiation, and revealed adverse sequelae of younger drinking initiation, supporting efforts to delay drinking onset.

Identifying high school risk factors that forecast heavy drinking onset in understudied young adults.

Heavy alcohol drinking is a major, preventable problem that adversely impacts the physical and mental health of US young adults. Studies seeking drinking risk factors typically focus on young adults who enrolled in 4-year residential college programs (4YCP) even though most high school graduates join the workforce, military, or community colleges. We examined 106 of these understudied young adults (USYA) and 453 4YCPs from the National Consortium on Alcohol and NeuroDevelopment in Adolescence (NCANDA) by longitudinally following their drinking patterns for 8 years from adolescence to young adulthood. All participants were no-to-low drinkers during high school. Whereas 4YCP individuals were more likely to initiate heavy drinking during college years, USYA participants did so later. Using mental health metrics recorded during high school, machine learning forecasted individual-level risk for initiating heavy drinking after leaving high school. The risk factors differed between demographically matched USYA and 4YCP individuals and between sexes. Predictors for USYA drinkers were sexual abuse, physical abuse for girls, and extraversion for boys, whereas 4YCP drinkers were predicted by the ability to recognize facial emotion and, for boys, greater openness. Thus, alcohol prevention programs need to give special consideration to those joining the workforce, military, or community colleges, who make up the majority of this age group.

Multi-level prediction of substance use: Interaction of white matter integrity, resting-state connectivity and inhibitory control measured repeatedly in every-day life.

Substance use disorders are characterized by inhibition deficits related to disrupted connectivity in white matter pathways, leading via interaction to difficulties in resisting substance use. By combining neuroimaging with smartphone-based ecological momentary assessment (EMA), we questioned how biomarkers moderate inhibition deficits to predict use. Thus, we aimed to assess white matter integrity interaction with everyday inhibition deficits and related resting-state network connectivity to identify multi-dimensional predictors of substance use. Thirty-eight patients treated for alcohol, cannabis or tobacco use disorder completed 1 week of EMA to report substance use five times and complete Stroop inhibition testing twice daily. Before EMA tracking, participants underwent resting state functional MRI and diffusion tensor imaging (DTI) scanning. Regression analyses were conducted between mean Stroop performances and whole-brain fractional anisotropy (FA) in white matter. Moderation testing was conducted between mean FA within significant clusters as moderator and the link between momentary Stroop performance and use as outcome. Predictions between FA and resting-state connectivity strength in known inhibition-related networks were assessed using mixed modelling. Higher FA values in the anterior corpus callosum and bilateral anterior corona radiata predicted higher mean Stroop performance during the EMA week and stronger functional connectivity in occipital-frontal-cerebellar regions. Integrity in these regions moderated the link between inhibitory control and substance use, whereby stronger inhibition was predictive of the lowest probability of use for the highest FA values. In conclusion, compromised white matter structural integrity in anterior brain systems appears to underlie impairment in inhibitory control functional networks and compromised ability to refrain from substance use.

Federated learning with knowledge distillation for multi-organ segmentation with partially labeled datasets.

The state-of-the-art multi-organ CT segmentation relies on deep learning models, which only generalize when trained on large samples of carefully curated data. However, it is challenging to train a single model that can segment all organs and types of tumors since most large datasets are partially labeled or are acquired across multiple institutes that may differ in their acquisitions. A possible solution is Federated learning, which is often used to train models on multi-institutional datasets where the data is not shared across sites. However, predictions of federated learning can be unreliable after the model is locally updated at sites due to 'catastrophic forgetting'. Here, we address this issue by using knowledge distillation (KD) so that the local training is regularized with the knowledge of a global model and pre-trained organ-specific segmentation models. We implement the models in a multi-head U-Net architecture that learns a shared embedding space for different organ segmentation, thereby obtaining multi-organ predictions without repeated processes. We evaluate the proposed method using 8 publicly available abdominal CT datasets of 7 different organs. Of those datasets, 889 CTs were used for training, 233 for internal testing, and 30 volumes for external testing. Experimental results verified that our proposed method substantially outperforms other state-of-the-art methods in terms of accuracy, inference time, and the number of parameters.

Contributions of Cerebral White Matter Hyperintensities to Postural Instability in Aging With and Without Alcohol Use Disorder.

BACKGROUND: Both postural instability and brain white matter hyperintensities (WMHs) are noted markers of normal aging and alcohol use disorder (AUD). Here, we questioned what variables contribute to the sway path-WMH relationship in individuals with AUD and healthy control participants. METHODS: The data comprised 404 balance platform sessions, yielding sway path length and magnetic resonance imaging data acquired cross-sectionally or longitudinally in 102 control participants and 158 participants with AUD ages 25 to 80 years. Balance sessions were typically conducted on the same day as magnetic resonance imaging fluid-attenuated inversion recovery acquisitions, permitting WMH volume quantification. Factors considered in multiple regression analyses as potential contributors to the relationship between WMH volumes and postural instability were age, sex, socioeconomic status, education, pedal 2-point discrimination, systolic and diastolic blood pressure, body mass index, depressive symptoms, total alcohol consumed in the past year, and race. RESULTS: Initial analysis identified diagnosis, age, sex, and race as significant contributors to observed sway path-WMH relationships. Inclusion of these factors as predictors in multiple regression analyses substantially attenuated the sway path-WMH relationships in both AUD and healthy control groups. Women, irrespective of diagnosis or race, had shorter sway paths than men. Black participants, irrespective of diagnosis or sex, had shorter sway paths than non-Black participants despite having modestly larger WMH volumes than non-Black participants, which is possibly a reflection of the younger age of the Black sample. CONCLUSIONS: Longer sway paths were related to larger WMH volumes in healthy men and women with and without AUD. Critically, however, age almost fully accounted for these associations.

Contributions of cerebral white matter hyperintensities, age, and pedal perception to postural sway in people with HIV.

OBJECTIVE: With aging, people with HIV (PWH) have diminishing postural stability that increases liability for falls. Factors and neuromechanisms contributing to instability are incompletely known. Brain white matter abnormalities seen as hyperintense (WMH) signals have been considered to underlie instability in normal aging and PWH. We questioned whether sway-WMH relations endured after accounting for potentially relevant demographic, physiological, and HIV-related variables. DESIGN: Mixed cross-sectional/longitudinal data were acquired over 15 years in 141 PWH and 102 age-range matched controls, 25-80 years old. METHODS: Multimodal structural MRI data were quantified for seven total and regional WMH volumes. Static posturography acquired with a force platform measured sway path length separately with eyes closed and eyes open. Statistical analyses used multiple regression with mixed modeling to test contributions from non-MRI and nonpath data on sway path-WMH relations. RESULTS: In simple correlations, longer sway paths were associated with larger WMH volumes in PWH and controls. When demographic, physiological, and HIV-related variables were entered into multiple regressions, the sway-WMH relations under both vision conditions in the controls were attenuated when accounting for age and two-point pedal discrimination. Although the sway-WMH relations in PWH were influenced by age, 2-point pedal discrimination, and years with HIV infection, the sway-WMH relations endured for five of the seven regions in the eyes-open condition. CONCLUSION: The constellation of age-related increasing instability while standing, degradation of brain white matter integrity, and peripheral pedal neuropathy is indicative of advancing fraility and liability for falls as people age with HIV infection.

Frontal cortical volume deficits as enduring evidence of childhood abuse in community adults with AUD and HIV infection comorbidity.

Background: Childhood abuse is an underappreciated source of stress, associated with adverse mental and physical health consequences. Childhood abuse has been directly associated with risky behavior thereby increasing the likelihood of alcohol misuse and risk of HIV infection, conditions associated with brain structural and functional deficits. Here, we examined the neural and behavioral correlates of childhood trauma history in alcohol use disorder (AUD), HIV infection (HIV), and their comorbidity (AUD+HIV). Methods: Occurrence of childhood trauma was evaluated by retrospective interview. Cortical (frontal, temporal, parietal, and occipital), subcortical (hippocampus, amygdala), and regional frontal volumes were derived from structural MRI, adjusted for intracranial volume and age. Test scores of executive functioning, attention/working memory, verbal/visual learning, verbal/visual memory, and motor speed functional domains were standardized on age and education of a laboratory control group. Results: History of childhood abuse was associated with smaller frontal lobe volumes regardless of diagnosis. For frontal subregional volumes, history of childhood abuse was selectively associated with smaller orbitofrontal and supplementary motor volumes. In participants with a child abuse history, poorer verbal/visual memory performance was associated with smaller orbitofrontal and frontal middle volumes, whereas in those without childhood abuse, poorer verbal/visual memory performance was associated with smaller orbitofrontal, frontal superior, and supplemental motor volumes. Conclusions: Taken together, these results comport with and extend the findings that childhood abuse is associated with brain and behavioral sequelae in AUD, HIV, and AUD+HIV comorbidity. Further, these findings suggest that sequelae of abuse in childhood may be best conceptualized as a spectrum disorder as significant deficits may be present in those who may not meet criteria for a formal trauma-related diagnosis yet may be suffering enduring stress effects on brain structural and functional health.

Time to Treatment With Intravenous Thrombolysis Before Thrombectomy and Functional Outcomes in Acute Ischemic Stroke: A Meta-Analysis.

Importance: The benefit of intravenous thrombolysis (IVT) for acute ischemic stroke declines with longer time from symptom onset, but it is not known whether a similar time dependency exists for IVT followed by thrombectomy. Objective: To determine whether the benefit associated with IVT plus thrombectomy vs thrombectomy alone decreases with treatment time from symptom onset. Design, Setting, and Participants: Individual participant data meta-analysis from 6 randomized clinical trials comparing IVT plus thrombectomy vs thrombectomy alone. Enrollment was between January 2017 and July 2021 at 190 sites in 15 countries. All participants were eligible for IVT and thrombectomy and presented directly at thrombectomy-capable stroke centers (n = 2334). For this meta-analysis, only patients with an anterior circulation large-vessel occlusion were included (n = 2313). Exposure: Interval from stroke symptom onset to expected administration of IVT and treatment with IVT plus thrombectomy vs thrombectomy alone. Main Outcomes and Measures: The primary outcome analysis tested whether the association between the allocated treatment (IVT plus thrombectomy vs thrombectomy alone) and disability at 90 days (7-level modified Rankin Scale [mRS] score range, 0 [no symptoms] to 6 [death]; minimal clinically important difference for the rates of mRS scores of 0-2: 1.3%) varied with times from symptom onset to expected administration of IVT. Results: In 2313 participants (1160 in IVT plus thrombectomy group vs 1153 in thrombectomy alone group; median age, 71 [IQR, 62 to 78] years; 44.3% were female), the median time from symptom onset to expected administration of IVT was 2 hours 28 minutes (IQR, 1 hour 46 minutes to 3 hours 17 minutes). There was a statistically significant interaction between the time from symptom onset to expected administration of IVT and the association of allocated treatment with functional outcomes (ratio of adjusted common odds ratio [OR] per 1-hour delay, 0.84 [95% CI, 0.72 to 0.97], P = .02 for interaction). The benefit of IVT plus thrombectomy decreased with longer times from symptom onset to expected administration of IVT (adjusted common OR for a 1-step mRS score shift toward improvement, 1.49 [95% CI, 1.13 to 1.96] at 1 hour, 1.25 [95% CI, 1.04 to 1.49] at 2 hours, and 1.04 [95% CI, 0.88 to 1.23] at 3 hours). For a mRS score of 0, 1, or 2, the predicted absolute risk difference was 9% (95% CI, 3% to 16%) at 1 hour, 5% (95% CI, 1% to 9%) at 2 hours, and 1% (95% CI, -3% to 5%) at 3 hours. After 2 hours 20 minutes, the benefit associated with IVT plus thrombectomy was not statistically significant and the point estimate crossed the null association at 3 hours 14 minutes. Conclusions and Relevance: In patients presenting at thrombectomy-capable stroke centers, the benefit associated with IVT plus thrombectomy vs thrombectomy alone was time dependent and statistically significant only if the time from symptom onset to expected administration of IVT was short.

Brain structural covariance network features are robust markers of early heavy alcohol use.

BACKGROUND AND AIMS: Recently, we demonstrated that a distinct pattern of structural covariance networks (SCN) from magnetic resonance imaging (MRI)-derived measurements of brain cortical thickness characterized young adults with alcohol use disorder (AUD) and predicted current and future problematic drinking in adolescents relative to controls. Here, we establish the robustness and value of SCN for identifying heavy alcohol users in three additional independent studies. DESIGN AND SETTING: Cross-sectional and longitudinal studies using data from the Pediatric Imaging, Neurocognition and Genetics (PING) study (n = 400, age range = 14-22 years), the National Consortium on Alcohol and Neurodevelopment in Adolescence (NCANDA) (n = 272, age range = 17-22 years) and the Human Connectome Project (HCP) (n = 375, age range = 22-37 years). CASES: Cases were defined based on heavy alcohol use patterns or former alcohol use disorder (AUD) diagnoses: 50, 68 and 61 cases were identified. Controls had none or low alcohol use or absence of AUD: 350, 204 and 314 controls were selected. MEASUREMENTS: Graph theory metrics of segregation and integration were used to summarize SCN. FINDINGS: Mirroring our prior findings, and across the three data sets, cases had a lower clustering coefficient [area under the curve (AUC) = -0.029, P = 0.002], lower modularity (AUC = -0.14, P = 0.004), lower average shortest path length (AUC = -0.078, P = 0.017) and higher global efficiency (AUC = 0.007, P = 0.010). Local efficiency differences were marginal (AUC = -0.017, P = 0.052). That is, cases exhibited lower network segregation and higher integration, suggesting that adjacent nodes (i.e. brain regions) were less similar in thickness whereas spatially distant nodes were more similar. CONCLUSION: Structural covariance network (SCN) differences in the brain appear to constitute an early marker of heavy alcohol use in three new data sets and, more generally, demonstrate the utility of SCN-derived metrics to detect brain-related psychopathology.

Age-Accelerated Increase of White Matter Hyperintensity Volumes Is Exacerbated by Heavy Alcohol Use in People Living With HIV.

BACKGROUND: Antiretroviral treatment has enabled people living with HIV infection to have a near-normal life span. With longevity comes opportunities for engaging in risky behavior, including initiation of excessive drinking. Given that both HIV infection and alcohol use disorder (AUD) can disrupt brain white matter integrity, we questioned whether HIV infection, even if successfully treated, or AUD alone results in signs of accelerated white matter aging and whether HIV+AUD comorbidity further accelerates brain aging. METHODS: Longitudinal magnetic resonance imaging-FLAIR data were acquired over a 15-year period from 179 control individuals, 204 participants with AUD, 70 participants with HIV, and 75 participants with comorbid HIV+AUD. White matter hyperintensity (WMH) volumes were quantified and localized, and their functional relevance was examined with cognitive and motor testing. RESULTS: The 3 diagnostic groups each had larger WMH volumes than the control group. Although all 4 groups exhibited accelerating volume increases with aging, only the HIV groups showed faster WMH enlargement than control individuals; the comorbid group showed faster acceleration than the HIV-only group. Sex and HIV infection length, but not viral suppression status, moderated acceleration. Correlations emerged between WMH volumes and attention/working memory and executive function scores of the AUD and HIV groups and between WMH volumes and motor skills in the 3 diagnostic groups. CONCLUSIONS: Even treated HIV can show accelerated aging, possibly from treatment sequelae or legacy effects, and notably from AUD comorbidity. WMH volumes may be especially relevant for tracking HIV and AUD brain health because each condition is associated with liability for hypertensive processes, for which WMHs are considered a marker.

Outcome of patients with multivessel occlusion stroke after endovascular treatment.

INTRODUCTION: Little is known about the implications of multivessel occlusions (MVO) in large vessel occlusion stroke patients who undergo endovascular treatment (EVT). PATIENTS AND METHODS: We report data from the MR CLEAN Registry: a prospective, observational study on all stroke patients who underwent EVT in the Netherlands (March 2014-November 2017). We included patients with an intracranial target occlusion in the anterior circulation. An MVO was defined as an MCA occlusion (M1/M2) or intracranial ICA/ICA-T occlusion, with a concurrent second occlusion in the ACA or PCA territory confirmed on baseline CTA. To compare outcomes, we performed a 10:1 propensity score matching analysis with a logistic regression model including potential confounders. Outcome measures included 90-day functional outcome (modified Rankin Scale, mRS) and mortality. RESULTS: Of 2946 included patients, 71 patients (2.4%) had an MVO (87% concurrent ACA occlusion, 10% PCA occlusion, 3% ⩾3 occlusions). These patients were matched to 71 non-MVO patients. Before matching, MVO patients had a higher baseline NIHSS (median 18 vs 16, p = 0.001) and worse collateral status (absent collaterals: 17% vs 6%, p < 0.001) compared to non-MVO patients. After matching, MVO patients had worse functional outcome at 90 days (median mRS 5 vs 3, cOR 0.39; 95%CI 0.25-0.62). Mortality was higher in MVO patients (46% vs 27%, OR 2.11, 95%CI 1.24-3.57). DISCUSSION AND CONCLUSION: MVOs on baseline imaging were uncommon in LVO stroke patients undergoing EVT, but were associated with poor functional outcome.

White matter microstructural integrity continues to develop from adolescence to young adulthood in mice and humans: Same phenotype, different mechanism.

As direct evaluation of a mouse model of human neurodevelopment, adolescent and young adult mice and humans underwent MR diffusion tensor imaging to quantify age-related differences in microstructural integrity of brain white matter fibers. Fractional anisotropy (FA) was greater in older than younger mice and humans. Despite the cross-species commonality, the underlying developmental mechanism differed: whereas evidence for greater axonal extension contributed to higher FA in older mice, evidence for continuing myelination contributed to higher FA in human adolescent development. These differences occurred in the context of species distinctions in overall brain growth: whereas the continued growth of the brain and skull in the murine model can accommodate volume expansion into adulthood, human white matter volume and myelination continue growth into adulthood within a fixed intracranial volume. Appreciation of the similarities and differences in developmental mechanism can enhance the utility of animal models of brain white matter structure, function, and response to exogenous manipulation.

Imputing Brain Measurements Across Data Sets via Graph Neural Networks.

Publicly available data sets of structural MRIs might not contain specific measurements of brain Regions of Interests (ROIs) that are important for training machine learning models. For example, the curvature scores computed by Freesurfer are not released by the Adolescent Brain Cognitive Development (ABCD) Study. One can address this issue by simply reapplying Freesurfer to the data set. However, this approach is generally computationally and labor intensive (e.g., requiring quality control). An alternative is to impute the missing measurements via a deep learning approach. However, the state-of-the-art is designed to estimate randomly missing values rather than entire measurements. We therefore propose to re-frame the imputation problem as a prediction task on another (public) data set that contains the missing measurements and shares some ROI measurements with the data sets of interest. A deep learning model is then trained to predict the missing measurements from the shared ones and afterwards is applied to the other data sets. Our proposed algorithm models the dependencies between ROI measurements via a graph neural network (GNN) and accounts for demographic differences in brain measurements (e.g. sex) by feeding the graph encoding into a parallel architecture. The architecture simultaneously optimizes a graph decoder to impute values and a classifier in predicting demographic factors. We test the approach, called Demographic Aware Graph-based Imputation (DAGI), on imputing those missing Freesurfer measurements of ABCD (N=3760; minimum age 12 years) by training the predictor on those publicly released by the National Consortium on Alcohol and Neurodevelopment in Adolescence (NCANDA, N=540). 5-fold cross-validation on NCANDA reveals that the imputed scores are more accurate than those generated by linear regressors and deep learning models. Adding them also to a classifier trained in identifying sex results in higher accuracy than only using those Freesurfer scores provided by ABCD.

LSOR: Longitudinally-Consistent Self-Organized Representation Learning.

Interpretability is a key issue when applying deep learning models to longitudinal brain MRIs. One way to address this issue is by visualizing the high-dimensional latent spaces generated by deep learning via self-organizing maps (SOM). SOM separates the latent space into clusters and then maps the cluster centers to a discrete (typically 2D) grid preserving the high-dimensional relationship between clusters. However, learning SOM in a high-dimensional latent space tends to be unstable, especially in a self-supervision setting. Furthermore, the learned SOM grid does not necessarily capture clinically interesting information, such as brain age. To resolve these issues, we propose the first self-supervised SOM approach that derives a high-dimensional, interpretable representation stratified by brain age solely based on longitudinal brain MRIs (i.e., without demographic or cognitive information). Called Longitudinally-consistent Self-Organized Representation learning (LSOR), the method is stable during training as it relies on soft clustering (vs. the hard cluster assignments used by existing SOM). Furthermore, our approach generates a latent space stratified according to brain age by aligning trajectories inferred from longitudinal MRIs to the reference vector associated with the corresponding SOM cluster. When applied to longitudinal MRIs of the Alzheimer's Disease Neuroimaging Initiative (ADNI, N=632), LSOR generates an interpretable latent space and achieves comparable or higher accuracy than the state-of-the-art representations with respect to the downstream tasks of classification (static vs. progressive mild cognitive impairment) and regression (determining ADAS-Cog score of all subjects). The code is available at https://github.com/ouyangjiahong/longitudinal-som-single-modality.

An Explainable Geometric-Weighted Graph Attention Network for Identifying Functional Networks Associated with Gait Impairment.

One of the hallmark symptoms of Parkinson's Disease (PD) is the progressive loss of postural reflexes, which eventually leads to gait difficulties and balance problems. Identifying disruptions in brain function associated with gait impairment could be crucial in better understanding PD motor progression, thus advancing the development of more effective and personalized therapeutics. In this work, we present an explainable, geometric, weighted-graph attention neural network (xGW-GAT) to identify functional networks predictive of the progression of gait difficulties in individuals with PD. xGW-GAT predicts the multi-class gait impairment on the MDS-Unified PD Rating Scale (MDS-UPDRS). Our computational- and data-efficient model represents functional connectomes as symmetric positive definite (SPD) matrices on a Riemannian manifold to explicitly encode pairwise interactions of entire connectomes, based on which we learn an attention mask yielding individual- and group-level explainability. Applied to our resting-state functional MRI (rs-fMRI) dataset of individuals with PD, xGW-GAT identifies functional connectivity patterns associated with gait impairment in PD and offers interpretable explanations of functional subnetworks associated with motor impairment. Our model successfully outperforms several existing methods while simultaneously revealing clinically-relevant connectivity patterns. The source code is available at https://github.com/favour-nerrise/xGW-GAT.

Brain Volume in Fetal Alcohol Spectrum Disorders Over a 20-Year Span.

Importance: Anomalous brain development and mental health problems are prevalent in fetal alcohol spectrum disorders (FASD), but there is a paucity of longitudinal brain imaging research into adulthood. This study presents long-term follow-up of brain volumetrics in a cohort of participants with FASD. Objective: To test whether brain tissue declines faster with aging in individuals with FASD compared with control participants. Design, Setting, and Participants: This cohort study used magnetic resonance imaging (MRI) data collected from individuals with FASD and control individuals (age 13-37 years at first magnetic resonance imaging [MRI1] acquired 1997-2000) compared with data collected 20 years later (MRI2; 2018-2021). Participants were recruited for MRI1 through the University of Washington Fetal Alcohol Syndrome (FAS) Follow-Up Study. For MRI2, former participants were recruited by the University of Washington Fetal Alcohol and Drug Unit. Data were analyzed from October 2022 to August 2023. Main Outcomes and Measures: Intracranial volume (ICV) and regional cortical and cerebellar gray matter, white matter, and cerebrospinal fluid volumes were quantified automatically and analyzed, with group and sex as between-participant factors and age as a within-participant variable. Results: Of 174 individuals with MRI1 data, 48 refused participation, 36 were unavailable, and 24 could not be located. The remaining 66 individuals (37.9%) were rescanned for MRI2, including 26 controls, 18 individuals with nondysmorphic heavily exposed fetal alcohol effects (FAE; diagnosed prior to MRI1), and 22 individuals with FAS. Mean (SD) age was 22.9 (5.6) years at MRI1 and 44.7 (6.5) years at MRI2, and 35 participants (53%) were male. The FAE and FAS groups exhibited enduring stepped volume deficits at MRI1 and MRI2; volumes among control participants were greater than among participants with FAE, which were greater than volumes among participants with FAS (eg, mean [SD] ICV: control, 1462.3 [119.3] cc at MRI1 and 1465.4 [129.4] cc at MRI2; FAE, 1375.6 [134.1] cc at MRI1 and 1371.7 [120.3] cc at MRI2; FAS, 1297.3 [163.0] cc at MRI1 and 1292.7 [172.1] cc at MRI2), without diagnosis-by-age interactions. Despite these persistent volume deficits, the FAE participants and FAS participants showed patterns of neurodevelopment within reference ranges: increase in white matter and decrease in gray matter of the cortex and decrease in white matter and increase in gray matter of the cerebellum. Conclusions and Relevance: The findings of this cohort study support a nonaccelerating enduring, brain structural dysmorphic spectrum following prenatal alcohol exposure and a diagnostic distinction based on the degree of dysmorphia. FASD was not a progressive brain structural disorder by middle age, but whether accelerated decline occurs in later years remains to be determined.

Admission blood pressure and clinical outcomes in patients with acute ischaemic stroke treated with intravenous alteplase and endovascular treatment versus endovascular treatment alone: A MR CLEAN-NO IV substudy.

INTRODUCTION: High systolic blood pressure (SBP) is associated with poor functional outcome. We analysed whether the association of SBP with outcomes after endovascular treatment (EVT) is modified by prior intravenous thrombolysis (IVT). PATIENTS AND METHODS: This was a post-hoc analysis of MR CLEAN-NO IV, a randomised trial of IVT with alteplase followed by EVT versus EVT alone, within 4.5 h from stroke onset. SBP was recorded on hospital admission. The primary outcome was 90-day modified Rankin Scale (mRS) score and secondary outcomes included symptomatic intracranial haemorrhage (sICH) and successful reperfusion (eTICI 2b-3), analysed with (ordinal) logistic regression. Estimates were calculated per 10 mmHg change in SBP. We assessed whether IVT modified the associations of SBP with these outcomes using multiplicative interaction terms. RESULTS: Of 539 randomised patients, 266 received IVT. The association of SBP with mRS score was J-shaped, with an inflection point at 150 mmHg. Using 150 mmHg as a reference point, SBPs higher than 150 mmHg were associated with poor functional outcome (acOR: 1.23, 95% CI: 1.09-1.38), but lower SBPs were not (acOR: 1.14, 95% CI: 0.99-1.30). Higher SBP was not associated with the risk of sICH (aOR: 1.09, 95% CI: 0.93-1.27) nor with the probability of successful reperfusion (aOR: 1.00, 95% CI: 0.91-1.10). Our main result was that we found no effect modification by IVT (p-values for interaction, mRS = 0.94; sICH = 0.26; successful reperfusion = 0.58). DISCUSSION AND CONCLUSION: There was no effect modification of IVT with SBP for any of the clinical outcomes. Therefore, the level of SBP (if ⩽185/110 mmHg) should not guide IVT decisions in patients otherwise eligible for both IVT and EVT within the 4.5-h time window. TRIAL REGISTRATION: ISRCTN80619088, https://www.isrctn.com/ISRCTN80619088.

An Explainable Geometric-Weighted Graph Attention Network for Identifying Functional Networks Associated with Gait Impairment.

One of the hallmark symptoms of Parkinson's Disease (PD) is the progressive loss of postural reflexes, which eventually leads to gait difficulties and balance problems. Identifying disruptions in brain function associated with gait impairment could be crucial in better understanding PD motor progression, thus advancing the development of more effective and personalized therapeutics. In this work, we present an explainable, geometric, weighted-graph attention neural network (xGW-GAT) to identify functional networks predictive of the progression of gait difficulties in individuals with PD. xGW-GAT predicts the multi-class gait impairment on the MDS-Unified PD Rating Scale (MDS-UPDRS). Our computational- and data-efficient model represents functional connectomes as symmetric positive definite (SPD) matrices on a Riemannian manifold to explicitly encode pairwise interactions of entire connectomes, based on which we learn an attention mask yielding individual- and group-level explain-ability. Applied to our resting-state functional MRI (rs-fMRI) dataset of individuals with PD, xGW-GAT identifies functional connectivity patterns associated with gait impairment in PD and offers interpretable explanations of functional subnetworks associated with motor impairment. Our model successfully outperforms several existing methods while simultaneously revealing clinically-relevant connectivity patterns. The source code is available at https://github.com/favour-nerrise/xGW-GAT.