BACKGROUND:Cerebral small vessel disease (CSVD) is a common incidental MRI finding in patients with transient ischemic attack (TIA) and stroke and has been linked to cognitive decline. This study investigated the prevalence of CSVD imaging biomarkers in TIA patients and their association with cognitive performance over 3 years. METHODS:We included 246 TIA patients from the INSPiRE-TMS study (ClinicalTrials.gov: NCT01586702). CSVD was assessed on baseline 3 T MRI using a composite score (0-4) including white matter hyperintensities (WMH), lacunes, cerebral microbleeds (CMBs), and enlarged perivascular spaces (PVS). Cognitive performance was evaluated using the Montreal Cognitive Assessment (MoCA) at baseline and annually for 3 years. RESULTS:At least one CSVD imaging biomarker was present in 58.5% of patients. Lacunes (36.6%) were the most common, followed by PVS (28.1%), WMH (19.5%), and CMBs (17.9%). Higher CSVD-score was independently associated with greater cognitive decline over 3 years (β = -0.52, 95% CI -0.95- -0.08, p = 0.020), along with older age (β = -0.08, 95% CI -0.13 to -0.03, p = 0.001). CMB burden was the strongest predictive component of the CSVD-score (β = 0.42, 95% CI -0.63 to -0.22, p < 0.001). CSVD-score was particularly associated with decline in the memory domain (adjusted β of -0.18, 95% CI -0.32 to -0.04, p = 0.015). CONCLUSION:CSVD imaging markers are present in over half of TIA patients and are independently associated with cognitive decline up to 3 years, with the strongest effect on memory. Whether the presence of CMBs is the strongest predictive imaging biomarker of cognitive decline in TIA patients requires confirmation in further studies.

Background and objectivesThis study investigates whether ultra-early follow-up imaging can reliably identify patients at risk for hematoma expansion (HE) in acute intracerebral hemorrhage (ICH).MethodsIn this multicenter study, we analyzed data from patients with primary ICH who underwent at least two non-contrast cranial computed tomography (NCCT) scans within 7 days of symptom onset or last known well. To define the optimal time window, hematoma growth dynamics were retrospectively assessed in a large cohort (n = 1,663). Based on these findings, a prospective sub-study included patients with repeated imaging within 200 min (n = 46). HE was defined as >6 mL or >33% volume increase between the admission and the second follow-up scan. The diagnostic performance of early volume increase was evaluated using receiver operating characteristic (ROC) analysis.ResultsThe highest proportion of patients with active hemorrhage was detected within the first 200 min in the initial phase of this study. In the prospective sub-study, percentage volume increase between admission and early follow-up imaging demonstrated excellent diagnostic performance for HE (AUC = 0.819). At an optimized cutoff, the model yielded a sensitivity of 0.885 and a positive predictive value (PPV) of 74%. Among patients with early expansion already visible at follow-up, 50% showed further volume increase on the final scan. A separate analysis limited to follow-up imaging within the first 120 min after symptom onset (n = 27) revealed a higher diagnostic accuracy, with an AUC of 0.846 (sensitivity 0.857; PPV 71%).ConclusionICH evolves rapidly in the first hours after onset. Follow-up imaging within the first 200 min can diagnose hematoma growth with high sensitivity and good accuracy. However, the inability to distinguish between ongoing and completed expansion underscores the need for additional imaging or clinical markers to support clinical decisions.

Introduction Following a cerebrovascular event, the associated risks for further major adverse cerebro- and cardiovascular events and death (MACE) and important aspects of cognitive, mental and patient-reported outcomes are currently not understood, particularly long-term. Here, we present the study design of the ongoing Berlin Long-term Observation of Vascular Events (BeLOVE) stroke stratum and report data of the first study phase.Methods and analysis BeLOVE is a prospective, longitudinal, observational, hospital-based cohort study. Its stroke stratum enrols adult patients with acute ischaemic stroke, transient ischaemic attack (TIA) or non-traumatic intracerebral haemorrhage. Patients undergo deep phenotyping including cerebral and cardiac MRI, ECG, echocardiography and bio-sampling including multi-omics analyses. Regular, standardised follow-ups take place annually over a period of up to 10 years and record the frequency of MACE as the primary outcome.Secondary outcomes include the frequency, progression and interactions of functional impairments, namely post-stroke cognition, pain, depression, seizures and their relationship to quality of life.The first study phase included 758 patients (median 69 years, 37% female). At 2-year follow-up, the cumulative incidence (95% CI) of the composite primary endpoint MACE was 0.107 (0.085 to 0.132) and that of first ischaemic stroke, first myocardial infarction and death were 0.066 (0.049 to 0.086), 0.015 (0.008 to 0.026) and 0.040 (0.027 to 0.056), respectively.Ethics and dissemination Each participant will provide informed written consent during the acute in-hospital phase. Data will be available for research purposes via a written request to the data use and access committee.Trial registration number German Clinical Trials Register: http://www.drks.de/DRKS00023323 on 4 November 2020.

BACKGROUND:White matter hyperintensity (WMH) segmentation using BIANCA (Brain Intensity AbNormality Classification Algorithm) in stroke populations is complicated by vascular lesions that share T2-hyperintense signal characteristics with WMH. Whether preprocessing decisions in the treatment of lesions affect segmentation accuracy has not yet been systematically evaluated in cerebrovascular cohorts involving multiple scanners. METHODS:We compared fixed probability thresholds and locally adaptive thresholding with LOCATE (LOCally Adaptive Threshold Estimation), and three lesion-handling approaches: lesions present (non removed), replaced with zero intensities (removed), and replaced with normal-appearing white matter intensities (NAWM; inpainted), using the BeLOVE cohort (Berlin Longterm Observation of Vascular Events) and the WMH Segmentation Challenge dataset. Phase I (n = 89) optimized thresholding via stratified 5-fold cross-validation. Phase II assessed preprocessing effects on segmentation accuracy (Phase II-A, n = 89) and volume agreement (Phase II-B, n = 211). RESULTS:BIANCA with LOCATE adaptive thresholding achieved moderate segmentation overlap (mean Dice 0.567), with lesion-level detection exceeding an F1 of 0.85. Preprocessing effects were statistically detectable but negligible in magnitude, with near-perfect agreement between all conditions. Stroke lesion volume was the highest-ranked predictor of volume differences between conditions; these scaled with lesion size but remained negligible in magnitude across all subgroups. CONCLUSIONS:BIANCA with LOCATE achieved moderate WMH segmentation performance with the best sensitivity-precision trade-off in this multi-scanner cerebrovascular cohort. Preprocessing effects were negligible at the group level. However, large lesions distort the FLAIR intensity distribution on which BIANCA relies for classification, which justifies lesion removal as a recommended preprocessing step.

BACKGROUND:Atrial fibrillation (AF) burden, the time spent in AF, is increasingly investigated. Data on its characteristics and temporal patterns during early in-hospital telemetry after ischaemic stroke remain limited. PATIENTS AND METHODS:Continuous ECG telemetry was analysed in consecutive patients with MRI-confirmed ischaemic stroke and AF admitted between October 2020 and January 2023 in this observational retrospective cohort trial. Peak AF burden within any 24-h window was categorised as low (≤1 h), medium (>1-6 h), high (> 6 h) or no recorded AF episodes. Associations with clinical and imaging features were assessed using multivariable regression. RESULTS:Among 392 patients with AF-related stroke (median monitoring duration 68 h, median AF burden 12.0 h), AF burden was low in 30 (7.7%), medium in 68 (17.3%), high in 193 (49.2%) and 101 (25.8%) had no recorded episodes. Left-insular stroke was negatively associated with high AF burden (aOR 0.38, 95% CI, 0.15-0.84, P < .05). There was a signal towards a positive association with right-insular stroke (aOR 1.75, 95% CI, 0.90-3.41, P = .09). The laterality difference persisted in continuous-scale analyses. First AF episodes clustered in the early evening among low-burden patients, with early morning episodes least common across all groups. Among patients with known stroke onset (n = 102), 49 exhibited AF confined to the first 48 h. CONCLUSION:AF burden varies widely during early in-hospital monitoring after ischaemic stroke and seems to differ according to insular lesion laterality. Our findings are hypothesis-generating and highlight the need for integrating lesion location, circadian and arrhythmia dynamics to characterise AF burden after stroke.

Abstract Background and aims Atrial fibrillation (AF) detected after ischemic stroke (AFDAS) is heterogenous. We hypothesized that AFDAS identified on admission ECG (ECG-AFDAS) represents an intermediate stage on the AF disease continuum between AFDAS detected only during prolonged cardiac monitoring on stroke unit (PCM-AFDAS) and AF known before the stroke (KAF). Methods We compared clinical, electrophysiological and neuroimaging features of consecutive patients with AF and MRI-confirmed ischemic stroke admitted to a comprehensive stroke center (10/2020–01/2023). AF subtype was treated as an ordered factor reflecting the suspected AF disease progression (PCM-AFDAS Results Among 409 ischemic stroke patients with AF (median age 81 years, 272 had KAF (66.5%), 56 ECG-AFDAS (13.7%), and 81 PCM-AFDAS (19.8%). Monotonic increases (i.e., lowest in PCM-AFDAS and highest in KAF) were observed for rate of diabetes and prior stroke, CHA₂DS₂-VASc score and AF burden. Monotonic decreases (i.e., highest in PCM-AFDAS and lowest in KAF) were observed for stroke lesion volume, rate of insular involvement, heart rate during AF, and LDL cholesterol (all P for trend <0.05). Across these variables, ECG-AFDAS consistently maintained an intermediate position relative to PCM-AFDAS and KAF. Conclusions Across many domains, ECG-AFDAS represents an intermediate AF phenotype between PCM-AFDAS and KAF. These findings support a continuum model of AFDAS and could inform future risk stratification strategies. Conflict of interest Markus G. Klammer, MD: nothing to disclose; Laura Reimann: nothing to disclose; Simone Lieschke, MD: nothing to disclose; Helena Stengl, MD: nothing to disclose; Maximilian Schoels, MD: nothing to disclose; Alexander Nelde, MSc: nothing to disclose; Kersten Villringer, MD: nothing to disclose; Christian Meisel, MD: nothing to disclose; Matthias Endres, MD: nothing to disclose; Jan F. Scheitz, MD: nothing to disclose

Abstract Background and aims White matter hyperintensities (WMH) as part of the small vessel disease score are increasingly subjected to automated WMH segmentation to enhance objectivity and comparability across studies. In this study, we investigated whether a deep-learning based automated segmentation model, TrUE-Net performs better than 0.85 thresholded BIANCA (Brain Intensity AbNormality Classification Algorithm), a k-nearest neighbour approach in identifying WMH lesions Methods We compared the robustness and performance of BIANCA_0.85 vs. TrUE-Net using data of the prospective Berlin Long-term Observation of Vascular Events=BeLOVE, () cohort and the WMH Segmentation Challenge dataset () using manually delineated WMH masks as ground truth. We assessed the overall performance, precision, sensitivity using three datasets. One consisting of a mixed dataset (n=109), data from the BeLOVE cohort (n=59) with mixed vascular lesions, examined at a 3T Siemens and 3 T Philips scanner and data from the Challenge dataset (n=50) with a variety of MRI scanners, comparing their DICE score (DC) and accounting for factors such as lesion volumes Results TrUE-Net showed better performance, precision, sensitivity in all 3 datasets (DC: 0.66±0.19/0.64±0.22/0.69±0.15) than BIANCA_0.85 (DC: 0.5±0.22/ 0.57±0.24/0.41±0.14) (figure1) and greater robustness towards different scanner types than BIANCA _0.85 (figure2). Out of the investigated factors (figure3), lesion volume differences had the most impact on the performance of TrUE-Net Conclusions TrUE-Net proves to be more accurate and robust in segmenting WMH than BIANCA_0.85, even with different MRI scanners, making it more reliable for analysing large amounts of data, both cross-sectional and longitudinal. Conflict of interest nothing to disclose regarding the topic of the submitted abstract: Chiara-Sophia Hübotter, Uchralt Temuulen, Joachim E. Weber, Katharina Schönrath, Ira Rohrpasser-Napierkowski, Tuncer, Mehmet, Matthias Endres, Ivana Galinovic, Kersten Villringer Figure 1 - belongs to Results Figure 2 - belongs to Results Figure 3 - belongs to Results

Abstract Background and Objectives Normal appearing white matter (NAWM) may already harbor subtle microstructural alterations not yet visible on conventional MRI. Quantitative Multi-Parametric Mapping (qMPM) such as Magnetization Transfer saturation (MTsat), longitudinal relaxation rate (R1), and Proton Density (PD) offer new possibilities for analyzing NAWM which are sensitive to demyelination, axonal loss, and edema. We aimed to characterize these alterations within white matter hyperintensities (WMH) and the perilesional NAWM (pNAWM), to gain insights into the underlying process of lesion progression. We also investigated their association with cerebrovascular risk factors (CVRF) and long-term cognitive performance. Methods This investigation included the cerebral MRI data of 245 participants from the prospective Berlin Longterm Observation of Vascular Events (BeLOVE) study. Furthermore, 121 participants’ cognitive performance was evaluated at baseline and longitudinally at 2 years follow-up using Montreal Cognitive Assessment (MoCA). Regions of interest (ROIs) of WMH, pNAWM at 1, 2, 3 mm were assessed in comparison to the mirrored contralesional white matter (cWM). Linear mixed effects models were employed to demonstrate the pairwise comparisons between each region using estimated marginal means and the association of MPM metrics with CVRFs. Linear regression was used to assess the association with cognitive performance. Results In 245 participants, (mean age 62 years, SD: 12 years; 29.8% females), MPM metrics demonstrated a clear spatial gradient of microstructural injury. MTsat and R1 values were lower in WMH compared to cWM (ß = -0.48 (-0.52 - -0.44) and ß = -0.07 (-0.08 - -0.06), p<0.001, respectively) and showed gradual recovery with increasing distance indicating a microstructural gradient in pNAWM. Conversely, PD values were higher in WMH and decreased peripherally (ß = 2.32 (2.05 – 2.61, p<0.001). No substantial associations were found between MPM parameters and CVRFs in our cohort. At baseline and 2-year follow-up, cognitive performance was associated with higher pNAWM R1 values, whereas MTsat were only moderately associated. Discussion Quantitative MPM reliably detects microstructural alterations not only within WMH, but also in pNAWM, confirming the high sensitivity of qMPM to subtle tissue pathology and support its utility as a promising biomarker for longitudinal studies and monitoring therapeutic effects.

Background: Post-stroke depression (PSD) affects up to one-third of stroke survivors, significantly impacting rehabilitation success and quality of life. However, its underlying pathophysiology remains unclear. Methods: We analyzed two independent, prospective ischemic stroke cohorts (PROSCIS-B [NCT01363856][1] and BAPTISe [NCT01954797][2]; total N=377) to identify brain regions and networks associated with depressive symptoms post-stroke. Lesion-symptom mapping (LSM) assessed associations between lesion location and depressive symptoms measured via the Center for Epidemiologic Studies Depression Scale (CES-D) up to 12 months post-stroke, while lesion network mapping (LNM) evaluated lesion connectivity with brain networks. We explored correlations between spatial similarity to the LNM-identified network and CES-D scores using linear regression models. Results: LSM revealed no significant associations between lesion location and depressive symptoms. In contrast, LNM showed that lesion connectivity to brain regions?including the frontal pole, middle and inferior frontal gyri, inferior temporal gyrus, supramarginal gyrus, angular gyrus, frontal orbital cortex, and thalamus?correlated with CES-D scores (r=0.12, p=0.02). These regions overlapped with canonical resting-state networks, such as the frontoparietal (Dice coefficient [DC] = 0.28), salience (DC = 0.27), and default-mode networks (DC = 0.20), as well as a previously published depression circuit (DC = 0.43). Conclusions: Lesion location alone was not associated with depressive symptoms post-stroke. However, lesion connectivity analysis revealed associations with brain networks, particularly the frontoparietal, salience, and default-mode networks, suggesting that disruption to these circuits may contribute to the development of PSD up to one-year post-stroke. ### Competing Interest Statement ME reports grants from Bayer and fees paid to the Charité from Bayer, Boehringer Ingelheim, BMS/Pfizer, Daiichi Sankyo, Amgen, GSK, Sanofi, Covidien, and Novartis, all outside of the submitted work. CO reports grants from Boehringer Ingelheim and Peak Profiling and honoraria for lectures and/or scientific advice from Boehringer-Ingelheim, Janssen, Limes Klinikgruppe, Neuraxpharm, Oberberg Kliniken and Peak Profiling. ### Funding Statement The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: AKu and AK are participants in the Berlin Institute of Health-Charité Clinical Scientist Program funded by the Charité-Universitätsmedizin Berlin and the Berlin Institute of Health. ME received funding from the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany's Excellence Strategy-EXC-2049-390688087 and Collaborative Research Center ReTune TRR 295- 424778381. ME received additional funding from Bundesministerium für Bildung und Forschung (BMBF; German Ministry for Education and Research) for the Center for Stroke Research Berlin. CO received funding from the German Research Foundation (OT 209/7-3; 14-1, 19-1, 21-1, EXC 2049), the European Commission (IMI2 859366), the German Federal Ministry of Education and Research (KS2017-067), the Berlin Institute of Health (B3010350), and the Wellcome Trust. AHN reports receiving research funding from the Corona Stiftung, the Else Kröner-Fresenius-Stiftung, and the German Center for Cardiovascular Research (DZHK), and was funded by the Berlin Institute of Health-Charité Clinical Scientist Program of the Charité-Universitätsmedizin Berlin and the Berlin Institute of Health. ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: All participants provided informed consent. The studies were conducted in accordance to the Declaration of Helsinki and were approved by the local ethics committee in Berlin. I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes Data supporting the results of this study is available upon reasonable request from the corresponding author. Code availability: Open source software was used for the pre-processing and analysis of the data, including: Lead-DBS (https://github.com/netstim/leaddbs), FSL 6.0.6.4 (https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/) and ANTsPy (https://github.com/ANTsX/ANTsPy). The code used to perform AIC, and cross-validations has been made publicly available at https://osf.io/pgnwh/?view_only=015d9716f84d478eb6e2c5cf7d41fff3. [1]: /lookup/external-ref?link_type=CLINTRIALGOV&access_num=NCT01363856&atom=%2Fmedrxiv%2Fearly%2F2025%2F01%2F06%2F2024.12.31.24319837.atom [2]: /lookup/external-ref?link_type=CLINTRIALGOV&access_num=NCT01954797&atom=%2Fmedrxiv%2Fearly%2F2025%2F01%2F06%2F2024.12.31.24319837.atom

Background: Cerebral small vessel disease (CSVD) is a common incidental finding on cerebral MRI in patients with transient ischemic attack (TIA) and stroke and has been linked to increased cerebrovascular risk and cognitive decline. This study aimed to investigate the prevalence of CSVD imaging biomarkers in TIA patients and evaluate their association with cognitive function over three years following the ischemic event. Methods: A cohort of 246 TIA patients from the INSPiRE-TMS (ClinicalTrials.gov: [NCT01586702][1]) study were included. The CSVD-score ? including white matter hyperintensities (WMH), lacunes, cerebral microbleeds (CMBs), and enlarged perivascular spaces (PVS) ? was assessed on baseline MRI. Cognitive performance was assessed via the Montreal Cognitive Assessment (MoCA) at baseline and annual outpatient visits up to 3 years. Results: CSVD was present in 58.5% of TIA patients. The most prevalent imaging biomarker was lacunes (36.6%), followed by PVS (28.1%), WMH (19.5%) and CMBs (17.9%). Cumulative CSVD-score (range 0-4) was an independently associated with cognitive decline up to 3 years (? = -0.53, 95% CI -0.97 ? -0.09, p = 0.018), alongside advanced age (? = -0.08, 95% CI -0.13 ? -0.03, p=0.001). CMB burden was the strongest predictive component of the CSVD-score (? = 0.42, 95% CI -0.63 ? -0.21, p < 0.001). Specifically, CSVD-score had a significant negative effect on the memory domain of cognitive function with an adjusted ? of -0.18 (95% CI -0.32 ? -0.04, p = 0.014). Conclusion: Imaging biomarkers of CSVD are present in more than half of TIA patients and are an independent predictor of cognitive decline up to 3 years, with the strongest effect on the memory domain of cognitive function. Whether the presence of CMBs is the strongest predictive imaging biomarker of cognitive decline in TIA patients requires confirmation in further studies. ### Competing Interest Statement The authors have declared no competing interest. ### Clinical Trial ClinicalTrials.gov of the INSPiRE-TMS study: [NCT01586702][1] ### Funding Statement The INSPiRE-TMS study was funded within the grant of the German Federal Ministry of Education and Research (BMBF) for the Center for Stroke Research Berlin and co-funded with unrestricted grants of Pfizer and the German Stroke Foundation. No additional funding was received towards this work. ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: The INSPIRE-TMS study protocol was created in accordance with the Declaration of Helsinki and approved by the ethics committee of the Charité Universitätsmedizin Berlin (EA2/084/11) with additional approval obtained by the respective ethics committees at all participating centres. I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes Data supporting the results of this study can be provided upon reasonable request. [1]: /lookup/external-ref?link_type=CLINTRIALGOV&access_num=NCT01586702&atom=%2Fmedrxiv%2Fearly%2F2025%2F03%2F11%2F2025.03.05.25323459.atom

Central Post Stroke Pain (CPSP) is a common, debilitating sequela of stroke that can occur within weeks to months after stroke onset, yet mechanistic insight and predictive biomarkers are lacking. To address this gap, we longitudinally studied patients with acute somatosensory stroke using structural and functional magnetic resonance imaging and assessed them for subsequent pain development. In the acute stage, patients who later developed CPSP had reduced grey matter concentration (GMC) in bilateral caudate nuclei and nuclei accumbens (NAc) and reduced functional connectivity from the contralesional caudate to the supplementary motor area (SMA) compared to those who remained pain-free. Reduced GMC of the contralesional caudate and reduced functional connectivity to SMA persisted after pain onset, furthermore, NAc exhibited reduced functional connectivity to bilateral sensorimotor cortices and SMA. Our findings provide evidence for neural changes that may predispose stroke patients to develop CPSP and point towards potential targets for pain prevention.

BACKGROUND:Predicting functional recovery after ischemic stroke is vital for guiding clinical care. This study investigated whether lesion network mapping (LNM), a technique for modeling symptom-specific brain networks, can improve outcome prediction of functional recovery up to one-year post-stroke. METHODS:We pooled data from two prospective stroke cohorts (1000Plus and PROSCIS-B; N = 565). Seven NIHSS-derived symptom networks were generated using LNM based on NIHSS sub-scores on admission (i.e., consciousness, language, motor, sensory, vision, neglect and ataxia). Lesion masks derived from MRI (within 7 days) were intersected with each symptom network to calculate individual network damage scores. Functional outcome was defined by the modified Rankin Scale (mRS) at 3 months (1000Plus) or 12 months (PROSCIS-B). Ordinal logistic regression models were performed to evaluate additional predictive value of LNM: Model 1 included age, lesion volume, and presence of selected neurological deficits; Model 2 included age, lesion volume, and NIHSS-derived network damage scores. Models were compared using pseudo-R2 and AIC. RESULTS:Patients had a mean age of 68 years and a median NIHSS of 3 (IQR 1-5). LNM revealed distinct, symptom-specific networks, with corresponding damage scores that were higher in patients exhibiting the respective deficits compared to those without. However, inclusion of these scores did not enhance the predictive accuracy of functional outcomes beyond that achieved with clinical variables alone (Model 1 vs. Model 2: pseudo-R2: 0.0468 vs. 0.0159; AIC:1730.598 vs. 1769.222). CONCLUSIONS:LNM-derived scores reflected symptom topography but did not enhance prediction of functional recovery. While promising as a mechanistic tool, the clinical utility of LNM-based damage metrics for prognostication remains limited and requires further validation.

BACKGROUND:Poststroke depression affects up to one-third of stroke survivors, significantly impacting recovery and quality of life. However, its pathophysiology remains unclear. METHODS:We analyzed 2 independent, prospective ischemic stroke cohorts (PROSCIS-B [Prospective Cohort of Incident Stroke Berlin] and BAPTISe [Biomarkers and Perfusion-Training-Induced Changes After Stroke]; n=377) enrolled at the Charité Hospital, Germany, to identify brain regions and networks associated with depressive symptoms poststroke. Lesion-symptom mapping assessed associations between lesion location and depressive symptoms measured by the Center for Epidemiological Studies Depression Scale at 6 (BAPTISe) or 12 (PROSCIS-B) months poststroke. Lesion-network mapping evaluated lesion connectivity with brain networks. A mixed-effects model, including cohort as a random effect, assessed the relationship between network similarity (Pearson correlation) and Center for Epidemiological Studies Depression Scale scores. Dice coefficients (DC) quantified spatial overlap with canonical resting-state networks. RESULTS:Lesion-symptom mapping showed no significant associations between lesion location and depressive symptoms. In contrast, lesion-network mapping revealed that lesion connectivity to brain regions including the frontal pole, middle and inferior frontal gyri, inferior temporal gyrus, supramarginal gyrus, angular gyrus, frontal orbital cortex, and thalamus weakly correlated with Center for Epidemiological Studies Depression Scale scores (β, 11.4 [95%CI, 1.8-21.1]; P=0.02). These regions overlapped with the frontoparietal (DC=0.28), salience (DC=0.27), and default mode (DC=0.20) networks, as well as a published depression circuit (DC=0.43). However, these findings did not replicate across data sets. CONCLUSIONS:Lesion location alone was not associated with poststroke depression. However, connectivity-based analyses implicated disruption of large-scale brain networks in the development of depressive symptoms. The failure to validate this association across data sets underscores the need for further studies with more comparable patient populations-particularly in terms of stroke severity and harmonized assessment time-points-to confirm these findings and their clinical relevance. REGISTRATION:URL: https://www.clinicaltrials.gov; Unique identifier: NCT01363856. URL: https://www.clinicaltrials.gov; Unique identifier: NCT01954797.

PurposeThis study aimed to evaluate the perfomance of Siemens Healthineers’ StrokeSegApp performance in automatically segmenting diffusion and perfusion lesions in patients with acute ischemic stroke and to assess its clinical utility in guiding mechanical thrombectomy decisions.MethodsThis retrospective study used MRI data of acute ischemic stroke patients from the prospective observational single-center 1000Plus study, acquired between September 2008 and June 2013 (clinicaltrials.org; NCT00715533) and manually segmented by radiologists as the ground truth. The performance of the StrokeSegApp was compared against this ground truth using the dice similarity coefficient (DSC) and Bland–Altman plots. The study also evaluated the application’s ability to recommend mechanical thrombectomy based on DEFUSE 2 and 3 trial criteria.ResultsThe StrokeSegApp demonstrated a mean DSC of 0.60 (95% CI: 0.57–0.63; n = 241) for diffusion deficit segmentation and 0.80 (95% CI: 0.76–0.85; n = 56) for perfusion deficit segmentation. The mean volume deviation was 0.49 mL for diffusion lesions and −7.69 mL for perfusion lesions. Out of 56 subjects meeting DEFUSE 2/3 criteria in the cohort, it correctly identified mechanical thrombectomy candidates with a sensitivity of 82.1% (95% CI: 63.1–93.9%) and a specificity of 96.4% (95% CI: 81.7–99.9%).ConclusionThe Siemens Healthineers’ StrokeSegApp provides accurate automated segmentation of ischemic stroke lesions, comparable to human experts as well as similar commercial software, and shows potential as a reliable tool in clinical decision-making for stroke treatment.

Blood-brain barrier (BBB) alterations may contribute to AD pathology through various mechanisms, including impaired amyloid-β (Aβ) clearance and neuroinflammation. Soluble platelet-derived growth factor receptor beta (sPDGFRβ) has emerged as a potential biomarker for BBB integrity. Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) offers a direct assessment of BBB permeability. However, the relationship between BBB dysfunction, cognitive impairment, and AD pathology remains unclear, with inconsistent findings in the literature. We conducted a cross-sectional study using data from the DELCODE and DESCRIBE cohorts to investigate BBB dysfunction in participants with normal cognition (NC), mild cognitive impairment (MCI), and AD dementia. BBB function was assessed using DCE-MRI and sPDGFRβ levels in cerebrospinal fluid and AD biomarkers Aβ and tau were measured. In a subset of patients, the CSF/plasma-ratio of albumin (QAlb) as a standard marker of BBB integrity and markers of neuroinflammation were analyzed. 91 participants (NC: 44, MCI: 21, AD: 26) were included in the analysis. The average age was 74.4 years, 42

Thalamic aphasia results from focal thalamic lesions that cause dysfunction of remote but functionally connected cortical areas due to language network perturbation. However, specific local and network-level neural substrates of thalamic aphasia remain incompletely understood. Using lesion symptom mapping, we demonstrate that lesions in the left ventrolateral and ventral anterior thalamic nucleus are most strongly associated with aphasia in general and with impaired semantic and phonemic fluency and complex comprehension in particular. Lesion network mapping (using a normative connectome based on fMRI data from 1000 healthy individuals) reveals a Thalamic aphasia network encompassing widespread left-hemispheric cerebral connections, with Broca's area showing the strongest associations, followed by the superior and middle frontal gyri, precentral and paracingulate gyri, and globus pallidus. Our results imply the critical involvement of the left ventrolateral and left ventral anterior thalamic nuclei in engaging left frontal cortical areas, especially Broca's area, during language processing. A lesion symptom mapping and lesion network mapping study suggests that lesions in the left ventral thalamic nuclei, linked to thalamic aphasia, map onto a common left-hemispheric network, with Broca's area showing the strongest associations.

Purpose To help radiologists examine the growing number of computed tomography (CT) scans, automatic anomaly detection is an ongoing focus of medical imaging research. Radiologists must analyze a CT scan by searching for any deviation from normal healthy anatomy. We propose an approach to detecting abnormalities in axial 2D CT slice images of the brain. Although much research has been done on detecting abnormalities in magnetic resonance images of the brain, there is little work on CT scans, where abnormalities are more difficult to detect due to the low image contrast that must be represented by the model used. Approach We use a generative adversarial network (GAN) to learn normal brain anatomy in the first step and compare two approaches to image reconstruction: training an encoder in the second step and using iterative optimization during inference. Then, we analyze the differences from the original scan to detect and localize anomalies in the brain. Results Our approach can reconstruct healthy anatomy with good image contrast for brain CT scans. We obtain median Dice scores of 0.71 on our hemorrhage test data and 0.43 on our test set with additional tumor images from publicly available data sources. We also compare our models to a state-of-the-art autoencoder and a diffusion model and obtain qualitatively more accurate reconstructions. Conclusions Without defining anomalies during training, a GAN-based network was used to learn healthy anatomy for brain CT scans. Notably, our approach is not limited to the localization of hemorrhages and tumors and could thus be used to detect structural anatomical changes and other lesions. (c) The Authors. Published by SPIE under a Creative Commons Attribution 4.0 International License.Distribution or reproduction of this work in whole or in part requires full attribution of the originalpublication, including its DOI. [DOI:10.1117/1.JMI.11.4.044508]

BACKGROUND:Atrial fibrillation detected after stroke (AFDAS) is considered to be a distinct entity influenced by cardiogenic and neurogenic factors. We hypothesized that patients with AFDAS have larger stroke lesions than patients without atrial fibrillation (AF) and with known AF (KAF). METHODS AND RESULTS:Consecutive patients with magnetic resonance imaging-confirmed acute ischemic stroke admitted to a university hospital between October 2020 and January 2023 were prospectively registered. We categorized patients as AFDAS, no AF or KAF upon hospital discharge. We manually segmented diffusion-weighted imaging lesions to determine lesion volume. We analyzed 1420 patients (median age, 78; 47.2% women, median National Institutes of Health Stroke Scale score, 3; median hospital stay, 5 days). Of these, 81 had AFDAS (5.7%), 329 had KAF (23.2%) and 1010 had no AF (71.1%). Lesion volume was larger in patients with AFDAS (median, 5.4 mL [interquartile range, 1.0-21.6]) compared with patients with no AF and KAF (median, 0.7 [interquartile range,0.2-4.4] and 2.0 [interquartile range,0.3-11.1] mL, respectively; both P<0.001). Lesion volume was independently associated with AFDAS compared with no AF (adjusted odds ratio, 1.37 [95% CI, 1.20-1.58] per log mL) and KAF (adjusted odds ratio, 1.22 [95% CI, 1.07-1.41] per log mL). Patients in the highest lesion volume quartile (>6.5 mL) were more likely to be diagnosed with AFDAS compared with the lowest quartile (<0.22 mL, 13.6% versus 2.1%; adjusted odds ratio, 5.88 [95% CI, 2.30-17.40]). These associations were more pronounced when excluding 151 patients with nonembolic lesion pattern and similar when excluding 199 patients with KAF on oral anticoagulation. CONCLUSIONS:Larger stroke lesions were independently associated with AFDAS diagnosis during index stroke hospitalization highlighting a potential neurogenic contribution to AFDAS pathogenesis.

OBJECTIVE:Among patients with acute stroke, we aimed to identify those who will later develop central post-stroke pain (CPSP) versus those who will not (non-pain sensory stroke [NPSS]) by assessing potential differences in somatosensory profile patterns and evaluating their potential as predictors of CPSP. METHODS:In a prospective longitudinal study on 75 acute stroke patients with somatosensory symptoms, we performed quantitative somatosensory testing (QST) in the acute/subacute phase (within 10 days) and on follow-up visits for 12 months. Based on previous QST studies, we hypothesized that QST values of cold detection threshold (CDT) and dynamic mechanical allodynia (DMA) would differ between CPSP and NPSS patients before the onset of pain. Mann-Whitney U-tests and mixed analysis of variances with Bonferroni corrections were performed to compare z-normalized QST scores between both groups. RESULTS:In total, 26 patients (34.7%) developed CPSP. In the acute phase, CPSP patients showed contralesional cold hypoesthesia compared to NPSS patients (p = 0.04), but no DMA differences. Additional exploratory analysis showed NPSS patients exhibit cold hyperalgesia on the contralesional side compared to the ipsilesional side, not seen in CPSP patients (p = 0.011). A gradient-boosting approach to predicting CPSP from QST patterns before pain onset had an overall accuracy of 84.6%, with a recall and precision of 75%. Notably, both in the acute and the chronic phase, approximately 80% of CPSP and NPSS patients showed bilateral QST abnormalities. INTERPRETATION:Cold perception differences between CPSP and NPSS patients appear early post stroke before pain onset. Prediction of CPSP through QST patterns seems feasible. ANN NEUROL 2025;97:507-520.

Introduction: Restricted retinal diffusion (RDR) has recently been recognized as a frequent finding on standard diffusion-weighted imaging (DWI) in central retinal artery occlusion (CRAO). However, data on early DWI signal evolution are missing. Patients and methods: Consecutive CRAO patients with DWI performed within 24 h after onset of visual impairment were included in a bicentric, retrospective cross-sectional study. Two blinded neuroradiologists assessed randomized DWI scans for the presence of retinal ischemia. RDR detection rates, false positive ratings, and interrater agreement were evaluated for predefined time groups. Results: Sixty eight CRAO patients (68.4 ± 16.8 years; 25 female) with 72 DWI scans (76.4% 3 T, 23.6% 1.5 T) were included. Mean time-delay between onset of CRAO and DWI acquisition was 13.4 ± 7.0 h. Overall RDR detection rates ranged from 52.8% to 62.5% with false positive ratings in 4.2%–8.3% of cases. RDR detection rates were higher in DWI performed 12–24 h after onset, when compared with DWI acquired within the first 12 h (79.5%vs 39.3%, p < 0.001). The share of false positive ratings was highest for DWI performed within the first 6 h of symptom onset (up to 14.3%). Interrater reliability was “moderate” for DWI performed within the first 18 h (κ = 0.57–0.58), but improved for DWI acquired between 18 and 24 h (κ = 0.94). Conclusion: DWI-based detection of retinal ischemia in early CRAO is likely to be time-dependent with superior diagnostic accuracy for DWI performed 12–24 h after onset of visual impairment.