Explore the vast range of titles available.

The increasing availability of biomedical data from large biobanks, electronic health records, medical imaging, wearable and ambient biosensors, and the lower cost of genome and microbiome sequencing have set the stage for the development of multimodal artificial intelligence solutions that capture the complexity of human health and disease. In this Review, we outline the key applications enabled, along with the technical and analytical challenges. We explore opportunities in personalized medicine, digital clinical trials, remote monitoring and care, pandemic surveillance, digital twin technology and virtual health assistants. Further, we survey the data, modeling and privacy challenges that must be overcome to realize the full potential of multimodal artificial intelligence in health. Multimodal artificial intelligence models could unlock many exciting applications in health and medicine; this Review outlines the most promising uses and the technical pitfalls to avoid.

Driven by innovative technologies, novel analytical methods, and collaborations unimaginable not long ago, our understanding of the role of genetic variation in stroke has advanced substantially in recent years. However, a vast amount of data have accumulated quickly, and increasingly complex methodologies used in studies make keeping up to date on relevant findings difficult. In addition to well known, highly penetrant rare mutations that cause mendelian disorders related to stroke, several common genetic variants have been associated with common stroke subtypes, some of which also affect disease severity and clinical outcome. Furthermore, common genetic variations in biological pathways that have an important role in the pathophysiology of cerebrovascular diseases—such as blood pressure and oxidative phosphorylation—have been implicated in stroke. Clinical and translational applications of these and future discoveries in stroke genetics include identification of novel targets for treatment and development of personalised approaches to stroke prevention and management.

Post-stroke seizures (PSS) are sequelae of intracerebral hemorrhage (ICH) that may negatively impact patient outcomes. Available literature suggests that inflammatory biomarkers contribute to epileptogenesis as well as mortality. This retrospective cohort study aimed to elucidate the prognostic ability of CCL2, IL6, and IL8 in PSS development. ICH data were collected at Yale New Haven Hospital from 2014 to 2021. Plasma biomarker levels were measured via cytometric bead array. Patients with EEG-recorded epileptiform discharges, EEG-recorded seizures, or clinical seizures after ICH were defined as seizures and epileptiform discharges (SED), while those without were defined as non-SED. SED was further divided into early and late, occurring before and 7 days post-ICH, respectively. Additionally, we examined if biomarkers were associated with poor outcome (modified Rankin Scale score of 3–6) and mortality at 90, 180, and 365 days post-ICH. We conducted univariable and multivariable logistic regression analyses and reported the findings as Odds Ratio (OR) and 95% CI. We examined 172 patients with ICH, of whom 33 had early SED, 29 had late SED, and 110 had no SED. In univariable analyses, CCL2, ICH volume, diabetes, and lobar ICH were significantly associated with late SED, whereas NIHSS score at admission, ICH volume, and lobar ICH were significantly associated with early SED (p < 0.05). Lower CCL2 levels were independent predictors of late SED in multivariable analyses (OR 0.58; 95% CI 0.41–0.80, p < 0.001), but not of early SED. Additionally, we identified an independent association between higher CCL2 levels and 90-day mortality in multivariable analysis for the combined early and late SED cohorts (OR 1.87; 95% CI 1.03–3.37, p = 0.038). Additional studies investigating additional aspects of biomarkers, such as their temporal profile post-stroke within 24 hours or beyond 72 hours, are needed.

Chronological age is an imperfect estimate of molecular aging. Epigenetic age, derived from DNA methylation data, provides a more nuanced representation of aging-related biological processes. We examine the bidirectional relationship between epigenetic age and brain health events (stroke, dementia, late-life depression) using data from 4,018 participants. Participants with a prior brain health event are 4% epigenetically older (β = 0.04, SE = 0.01), indicating these conditions are associated with accelerated aging beyond that captured by chronological age. Additionally, a one standard deviation increase in epigenetic age is associated with 70% higher odds of experiencing a brain health event in the next four years (OR = 1.70, 95% CI = 1.16–2.50), suggesting epigenetic age acceleration is not just a consequence but also a predictor of poor brain health. Mendelian Randomization analyses replicate these findings, supporting their causal nature. Our results support using epigenetic age as a biomarker to evaluate interventions aimed at preventing and promoting recovery after brain health events. Here, the authors examine the bidirectional relationship between epigenetic age and brain health events, suggesting that stroke and dementia accelerate our epigenetic age, while an accelerated epigenetic age increases the risk of these conditions.

Neuroimaging is crucial for assessing mass effect in brain-injured patients. Transport to an imaging suite, however, is challenging for critically ill patients. We evaluated the use of a low magnetic field, portable MRI (pMRI) for assessing midline shift (MLS). In this observational study, 0.064 T pMRI exams were performed on stroke patients admitted to the neuroscience intensive care unit at Yale New Haven Hospital. Dichotomous (present or absent) and continuous MLS measurements were obtained on pMRI exams and locally available and accessible standard-of-care imaging exams (CT or MRI). We evaluated the agreement between pMRI and standard-of-care measurements. Additionally, we assessed the relationship between pMRI-based MLS and functional outcome (modified Rankin Scale). A total of 102 patients were included in the final study (48 ischemic stroke; 54 intracranial hemorrhage). There was significant concordance between pMRI and standard-of-care measurements (dichotomous, κ  = 0.87; continuous, ICC  = 0.94). Low-field pMRI identified MLS with a sensitivity of 0.93 and specificity of 0.96. Moreover, pMRI MLS assessments predicted poor clinical outcome at discharge (dichotomous: adjusted OR 7.98, 95% CI 2.07–40.04 , p  = 0.005; continuous: adjusted OR 1.59, 95% CI 1.11–2.49, p  = 0.021). Low-field pMRI may serve as a valuable bedside tool for detecting mass effect.

Depression is a common source of human disability for which etiologic insights remain limited. Although abnormalities of monoamine neurotransmission, including dopamine, are theorized to contribute to the pathophysiology of depression, evidence linking dopamine-related genes to depression has been mixed. The current study sought to address this knowledge-gap by examining whether the combined effect of dopamine polymorphisms was associated with depressive symptomatology in both healthy individuals and individuals with depression. Data were drawn from three independent samples: (1) a discovery sample of healthy adult participants (n = 273); (2) a replication sample of adults with depression (n = 1,267); and (3) a replication sample of healthy adult participants (n = 382). A genetic risk score was created by combining functional polymorphisms from five genes involved in synaptic dopamine availability (COMT and DAT) and dopamine receptor binding (DRD1, DRD2, DRD3). In the discovery sample, the genetic risk score was associated with depressive symptomatology (β = -0.80, p = 0.003), with lower dopamine genetic risk scores (indicating lower dopaminergic neurotransmission) predicting higher levels of depression. This result was replicated with a similar genetic risk score based on imputed genetic data from adults with depression (β = -0.51, p = 0.04). Results were of similar magnitude and in the expected direction in a cohort of healthy adult participants (β = -0.86, p = 0.15). Sequence variation in multiple genes regulating dopamine neurotransmission may influence depressive symptoms, in a manner that appears to be additive. Further studies are required to confirm the role of genetic variation in dopamine metabolism and depression.

To create and validate an automated pipeline for detection of early signs of irreversible ischemic change from admission CTA in patients with large vessel occlusion (LVO) stroke. We retrospectively included 368 patients for training and 143 for external validation. All patients had anterior circulation LVO stroke, endovascular therapy with successful reperfusion, and follow-up diffusion-weighted imaging (DWI). We devised a pipeline to automatically segment Alberta Stroke Program Early CT Score (ASPECTS) regions and extracted their relative Hounsfield unit (rHU) values. We determined the optimal rHU cut points for prediction of final infarction in each ASPECT region, performed 10-fold cross-validation in the training set, and measured the performance via external validation in patients from another institute. We compared the model with an expert neuroradiologist for prediction of final infarct volume and poor functional outcome. We achieved a mean area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity of 0.69±0.13, 0.69±0.09, 0.61±0.23, and 0.72±0.11 across all regions and folds in cross-validation. In the external validation cohort, we achieved a median [interquartile] AUC, accuracy, sensitivity, and specificity of 0.71 [0.68-0.72], 0.70 [0.68-0.73], 0.55 [0.50-0.63], and 0.74 [0.73-0.77], respectively. The rHU-based ASPECTS showed significant correlation with DWI-based ASPECTS (rS = 0.39, p<0.001) and final infarct volume (rS = -0.36, p<0.001). The AUC for predicting poor functional outcome was 0.66 (95%CI: 0.57-0.75). The predictive capabilities of rHU-based ASPECTS were not significantly different from the neuroradiologist's visual ASPECTS for either final infarct volume or functional outcome. Our study demonstrates the feasibility of an automated pipeline and predictive model based on relative HU attenuation of ASPECTS regions on baseline CTA and its non-inferior performance in predicting final infarction on post-stroke DWI compared to an expert human reader.

Background/Objective There are limited data on the risks and benefits of using andexanet alfa (AA) in comparison with four-factor prothrombin complex concentrate (4F-PCC) to reverse factor Xa inhibitors (FXi) associated intracranial hemorrhage (ICH). We sought to describe our experience with AA or 4F-PCC in patients with oral FXi-related traumatic and spontaneous ICH. Methods We conducted a retrospective review of consecutive adult patients with FXi-related ICH who received AA or 4F-PCC. FXi-related ICH cases included traumatic and spontaneous intracranial hemorrhages. Our primary analysis evaluated ICH stability on head computed tomography scan (CT), defined as a similar amount of blood from the initial scan at the onset of ICH to subsequent scans, at 6-h and 24-h post-administration of AA or 4F-PCC. For the subset of spontaneous intraparenchymal hemorrhages, volume was measured at 6-h and 24-h post-reversal. In secondary analyses, we evaluated good functional outcome at discharge, defined as a Modified Rankin Score of less than 3, and the incidence of thrombotic events after AA or 4F-PCC adminstration, during hospitalization. Results A total of 44 patients (16 traumatic and 28 spontaneous ICH) with median age of 79 years [72–86], 36% females, with a FXi-related ICH, were included in this study. The majority of spontaneous ICHs were intraparenchymal 19 (68%). Twenty-eight patients (64%) received AA and 16 patients (36%) received 4F-PCC. There was no difference between AA and 4F-PCC in terms of CT stability at 6 h (21 [78%] vs 10 [71%], p  = 0.71) and 24 h (15 [88%] vs 6 [60%], p  = 0.15). In a subgroup of patients with spontaneous intraparenchymal hemorrhage, there was no difference in the degree of achieved hemostasis based on hematoma volume between AA and 4F-PCC at 6 h (9.3 mL [6.9–26.4] vs 10 mL [9.4–22.1], adjusted p  = 0. 997) and 24-h (9.2 mL [6.1–18.8] vs 9.9 [9.4–21.1], adjusted p  = 1). The number of patients with good outcome based on mRS on discharge were 10 (36%) and 6 (38%) in the AA and 4F-PCC groups, respectively (adjusted p  = 0.81). The incidence of thromboembolic events was similar in the AA and 4F-PCC groups (2 [7%] vs 0, p  = 0.53). Conclusion In this limited sample of patients, we found no difference in neuroimaging stability, functional outcome and thrombotic events when comparing AA and 4F-PCC in patients with FXi-related ICH. Since our analysis is likely underpowered, a multi-center collaborative network devoted to this question is warranted.

BackgroundAt least 60% of stroke, 40% of dementia and 35% of late-life depression (LLD) are attributable to modifiable risk factors, with great overlap due to shared pathophysiology. This study aims to systematically identify overlapping risk factors for these diseases and calculate their relative impact on a composite outcome.MethodsA systematic literature review was performed in PubMed, Embase and PsycInfo, between January 2000 and September 2023. We included meta-analyses reporting effect sizes of modifiable risk factors on the incidence of stroke, dementia and/or LLD. The most relevant meta-analyses were selected, and disability-adjusted life year (DALY) weighted beta (β)-coefficients were calculated for a composite outcome. The β-coefficients were normalised to assess relative impact.ResultsOur search yielded 182 meta-analyses meeting the inclusion criteria, of which 59 were selected to calculate DALY-weighted risk factors for a composite outcome. Identified risk factors included alcohol (normalised β-coefficient highest category: −34), blood pressure (130), body mass index (70), fasting plasma glucose (94), total cholesterol (22), leisure time cognitive activity (−91), depressive symptoms (57), diet (51), hearing loss (60), kidney function (101), pain (42), physical activity (−56), purpose in life (−50), sleep (76), smoking (91), social engagement (53) and stress (55).ConclusionsThis study identified overlapping modifiable risk factors and calculated the relative impact of these factors on the risk of a composite outcome of stroke, dementia and LLD. These findings could guide preventative strategies and serve as an empirical foundation for future development of tools that can empower people to reduce their risk of these diseases.PROSPERO registration numberCRD42023476939.

Frailty is an aging-related clinical phenotype defined as a state in which there is an increase in a person’s vulnerability for dependency and/or mortality when exposed to a stressor. While underlying mechanisms leading to the occurrence of frailty are complex, the importance of genetic factors has not been fully investigated. We conducted a large-scale genome-wide association study (GWAS) of frailty, as defined by the five criteria (weight loss, exhaustion, physical activity, walking speed, and grip strength) captured in the Fried Frailty Score (FFS), in 386,565 European descent participants enrolled in the UK Biobank (mean age 57 [SD 8] years, 208,481 [54%] females). We identified 37 independent, novel loci associated with the FFS ( p  < 5 × 10 –8 ), including seven loci without prior described associations with other traits. The variants associated with FFS were significantly enriched in brain tissues as well as aging-related pathways. Our post-GWAS bioinformatic analyses revealed significant genetic correlations between FFS and cardiovascular-, neurological-, and inflammation-related diseases/traits, and subsequent Mendelian Randomization analyses identified causal associations with chronic pain, obesity, diabetes, education-related traits, joint disorders, and depressive/neurological, metabolic, and respiratory diseases. The GWAS signals were replicated in the Health and Retirement Study (HRS, n  = 9,720, mean age 73 [SD 7], 5,582 [57%] females), where the polygenic risk score built from UKB GWAS was significantly associated with the FFS in HRS individuals (OR per SD of the score 1.27, 95% CI 1.22–1.31, p  = 1.3 × 10 –11 ). These results provide new insight into the biology of frailty by comprehensively evaluating its genetic architecture.