Search for: All records

Abstract Study objectivesThis study aimed to compare YASA’s automated sleep staging to manual staging in the context of a multi-night experimental sleep restriction protocol. MethodsSeventy-five adults (58% female; 57% nonwhite) participated in up to seven nights of laboratory-based polysomnography measurements. The study involved one adaptation night, followed by three nights of normal sleep (9 h time in bed) and three nights of sleep restriction (5.5 h time in bed). Condition order was counterbalanced. Manual sleep staging was performed by a registered polysomnographic technician. Sleep data were exported and processed using Python 3.12, MNE 1.8.0, and YASA 0.6.5 for automated sleep staging. ResultsAcross 483 valid sleep nights, there was 82.9% overall agreement between YASA scoring and manual scoring. Stage-specific agreement was 40.3% (N1), 85.1% (N2), 86.9% (N3), and 78.7% (REM). Agreement was higher during normal sleep nights than sleep restriction nights, particularly for wake, N1, and REM sleep classifications. ConclusionsYASA-based staging exhibited good overall agreement with manual scoring during normal sleep nights. However, caution is needed when interpreting N1 estimates, as well as in interpreting data during short sleep nights. Brief summary Current knowledge/study rationaleSleep stage scoring is typically conducted manually by at least one experienced technician, but this process is laborious and subject to biases. This study investigated whether a machine-learning-based open tool—YASA—could automatically stage polysomnography data with acceptable accuracy. Study impactAcross 430,813 epochs, YASA showed acceptable accuracy in sleep staging, making it an efficient tool for the sleep community. However, caution is still needed for some applications, such as interpreting YASA’s N1 estimates as well as data from short sleep nights. 

Abstract BackgroundCancer phylogenies are key to understanding tumor evolution. However, due to the uncertainty in phylogenetic estimation, one typically infers many, equally-plausible phylogenies from bulk DNA sequencing data of tumors, hindering downstream analysis that relies on correct phylogenies. ResultsTo resolve this challenge, we introduce Sapling, a method to solve two variants of theBackbone Tree Inference from Readsproblem, which seeks a small set of backbone trees on a subset of mutations that collectively summarize the space of plausible cancer phylogenies. We prove that the problems are NP-hard. ConclusionsOn simulated and real data, we demonstrate that Sapling is capable of inferring high-quality backbone trees that adequately summarize the space of plausible cancer phylogenies. In addition, we demonstrate that Sapling is able to infer full-size trees with higher likelihoods than state-of-the-art methods. 

Abstract Background:Public health resources are often allocated based on reported disease cases. However, for under-recognized infectious diseases such as tick-borne viruses, risk assessments should also account for ecological and socioeconomic factors that influence disease case reporting. This study identifies country-level predictors of tick-borne virus reporting and evaluates whether wealthier nations are more likely to report resource-intensive cases. Methods:We applied boosted regression trees, a robust machine learning algorithm, to a comprehensive global database of tick-borne viruses and 24 environmental and socioeconomic variables. Results:Countries with lower income inequality and greater expertise in verterinary, agricultural, or forestry sectors are more likely to report tick-borne virus cases. Wealthier nations with stronger institutional and professional capacity exhibit higher reporting rates, whereas countries affected by conflict or limited health infrastructure show underreporting. Climatic factors, particularly subartic environments, also contribute to reporting likelihood, complementing the effects of socioeconomic drivers. Conclusions:Disease reporting is shaped by both ecological context and socioeconomic capacity. Strengthening surveillance through targeted resource allocation and better integration of veterinary and public health expertise under the One Health framework could enhance global tick-borne disease mitigation. These findings provide valuable evidence to support the World Health Organization’s Global Arbovirus Initiative and emphasize the need for equitable disease surveillance across regions. 

Abstract Key messageThe 2024 ENSO event advanced the timing of spring phenological phases of native shrubs significantly more than non-native shrubs and native trees in a temperate deciduous woodland fragment in Wisconsin, USA. This suggests that, as spring temperatures warm, shrubs will likely play a pivotal role in forest dynamics including contributing to an earlier onset to the growing period and an early start to CO₂assimilation. ContextThe 2023/2024 El Niño Southern Oscillation (ENSO) event brought warmer than average temperatures to the Midwest USA. This presented a unique opportunity to examine how short-term warming might impact the phenology of temperate deciduous forest vegetation. AimTo quantify the impact of an ENSO-driven warm spring on the phenology of temperate deciduous forest vegetation in order to assess how trees and shrubs respond to short-term temperature anomalies. MethodsSpring phenology was recorded twice weekly (2018–2024) on 5 dominant tree species and 5 native and 4 non-native shrub species, in a woodland fragment on the University of Wisconsin Milwaukee campus. In addition, phenological transition dates were extracted from daily Green Chromatic Coordinate (GCC) data from a PhenoCam installed at the site. ResultsIn 2024, the average spring (March–May) temperature (8.7 ± 0.57 °C) was significantly warmer than the 2018–2023 average (6.7 ± 0.28 °C). Compared to the average of the previous 6 years, the timing of budburst in 2024 occurred significantly (p < 0.001) earlier on DOY 76, 82, and 98 for native shrubs, non-native shrubs, and trees, representing advances of 20, 17, and 18 days, respectively. The advance was greater in shrubs than trees suggesting that any future advance to the start of the growing-season in temperate deciduous forests resulting from warmer spring temperatures will likely be driven by early leafing species like shrubs. Notably, the rise in GCC in 2024 (DOY 116) occurred following budburst, indicating that PhenoCam imagery may not fully capture early vegetation phenology. ConclusionEarly leafing shrubs, and in particular native species, were more sensitive to warmer temperatures early in the season than non-native shrubs and native trees. Therefore, as temperatures warm in the future, the onset of growth in temperate deciduous forests is likely to be driven by the early spring phenophases of early leafing species. 

Abstract Purpose of ReviewThis review explores the metabolic pathways dysregulated in both atherosclerotic cardiovascular disease (ASCVD) and metabolic dysfunction-associated steatotic liver disease (MASLD), focusing on lipid, carbohydrate, amino acid, and energy metabolism and the specific alterations within major contributing cell types. Recent FindingsIn the setting of metabolic syndrome, lipid and carbohydrate overload impair hepatic metabolism, resulting in the accumulation of lipotoxic species and ensuing cellular damage, inflammation, oxidative stress, and cardiovascular consequences. Amino acid metabolism is emerging as a key regulator of cell fate and function in both MASLD and ASCVD. Mitochondrial dysfunction and cellular stress promote a pseudo-Warburg effect, shifting cells from efficient oxidative phosphorylation to anaerobic glycolysis and impairing homeostasis. Emerging therapies targeting hepatic metabolism to reduce cardiovascular risk and MASLD burden hold promise for future dual treatments. SummaryMASLD and ASCVD arise from common metabolic derangements that converge on shared cellular and molecular pathways. Defining these cross-tissue mechanisms may enable the development of integrated therapeutic approaches aimed at jointly mitigating hepatic and vascular injury, thus redefining treatment paradigms in cardiometabolic disease. 

Abstract BackgroundTraditional two-dimensional (2D) models do not adequately capture the complex cellular interactions, brain-specific architecture, and progressive pathology of Alzheimer’s disease (AD). Three-dimensional (3D) organoid and microfluidic technologies provide more physiologically relevant platforms for studying AD-associated neurovascular dysfunction. MethodsWe developed a membrane-free microfluidic endothelial barrier model integrated with neurospheroids derived from familial AD (FAD) neural progenitor cells. Human endothelial cells were cultured within perfusable microfluidic channels to establish a vascular-like interface rather than a fully specialized BBB endothelium. Pre-differentiated neurospheroids were grafted into the brain chamber. Endothelial barrier integrity, tight-junction expression, phosphorylated tau (pTau), and Aβ42/Aβ40 production and distribution between compartments were assessed using immunofluorescence imaging and ELISA. ResultsThe neurospheroid-grafted endothelial barrier construct captured key AD-associated phenotypes. ReN-AD-D4 models exhibited increased endothelial barrier permeability, reduced ZO-1 expression, and elevated pTau relative to controls. The platform supported endogenous Aβ generation, accumulation, and endothelial-associated deposition at the endothelial barrier. ELISA demonstrated differential Aβ42 and Aβ40 distribution, consistent with isoform-selective behavior reported in AD pathology. Collectively, these results indicate co-occurring neuronal and endothelial barrier alterations within the integrated 3D system. ConclusionThis microfluidic endothelial barrier–neurospheroid platform enables quantitative assessment of amyloid-β accumulation, spatial distribution, and compartmentalized secretion alongside tau pathology and endothelial barrier integrity changes. Integrating human endothelial monolayers with FAD-derived neurospheroids, the system is scalable and compatible with high-content imaging. Although it does not model BBB-specific transport mechanisms, it provides a robust framework for hypothesis-driven studies of neurovascular interactions and therapeutic screening applications. 

Abstract BackgroundEngineered vasculature is essential for the biofabrication of functional tissue mimics. To fabricate engineered vasculature, three-dimensional (3D) bioprinting has emerged as a promising approach due to its ability to form perfusable structures with customized geometries. Sacrificial ink extrusion, where sacrificial inks are printed into a crosslinkable hydrogel precursor support bath, is a versatile bioprinting modality for fabricating interconnected perfusable networks. However, the fabrication of self-supporting structures with a vessel-like shell remains challenging using conventional sacrificial ink extrusion approaches. To enable the fabrication of self-supporting, perfusable networks, we developed a 3D bioprinting approach termed Gelation of Uniform Interfacial Diffusant in Embedded 3D Printing (GUIDE-3DP). This approach leverages the diffusion of crosslinking initiators from a printed sacrificial ink into a gel precursor support bath to generate branched, perfusable networks with precise control over channel inner and outer diameters. MethodsHere, we present an end-to-end protocol for fabricating self-supporting vascular-like networks using the GUIDE-3DP method. We describe methods for freeform print path design, support bath and sacrificial ink preparation, 3D printing of perfusable structures, and seeding of printed structures with endothelial cells. Through this protocol, perfusable structures with complex branching geometries can be designed, fabricated, and endothelialized. DiscussionTo highlight the ability of GUIDE-3DP to fabricate self-supporting, perfusable networks with complex geometries, we demonstrate the fabrication of three representative structures: (1) an interconnected retinal vasculature network, (2) a hierarchical branched vascular network, and (3) a dual-material capillary-like network. We further demonstrate the endothelialization of printed structures with one or two cell types via single- or dual-material printing. Beyond vascular-like networks, this protocol is readily adaptable to design and fabricate mimics of other perfusable structures in the human body. Clinical trial numberNot applicable. 

Abstract Purpose of ReviewArtificial intelligence (AI) in healthcare has evolved dramatically from early expert systems, which were initially considered replacements for clinical judgment, to today’s collaborative frameworks that aim to augment physician decision-making. This evolution is particularly crucial in domains such as transplant surgery, where decisions carry irreversible consequences and require the integration of complex, often ambiguous data. Drawing on peer-reviewed literature from 2019 to 2025, we conducted a systematic review that analyzed key elements distinguishing successful human-AI partnerships from those that fail. Recent FindingsThe ideal balance incorporates human expertise into AI systems through weighted integration approaches, rather than binary accept-or-reject paradigms. This nuanced integration requires the ability to understand and adapt to the unique context of each case, underscoring the complexity and importance of the work of medical professionals and researchers. For transplant surgeons, whose practice exemplifies complex decision-making, current AI approaches built on binary logic struggle to capture the nuanced reasoning required for effective decision-making. Evidence from randomized controlled trials and multicenter validation studies demonstrates how human-AI collaborative systems achieve superior outcomes compared to either human or AI performance alone. SummaryThis systematic review traces the evolution of human-AI collaboration in medical decision-making, identifies gaps that limit true collaborative teaming, and examines how fuzzy logic systems offer a framework for supporting complex decisions while maintaining the clinical interpretability that surgeons require for confident decision-making 

Abstract BackgroundMicrobiome sequencing data are often collected from several body sites and exhibit dependencies. Our objective is to develop a model that enables joint analysis of data from different sites by capturing the underlying cross-site dependencies. The proposed model incorporates (i) latent factors shared across sites to explain common subject effects and to serve as the source of correlation between the sites and (ii) mixtures of latent factors to allow heterogeneity among the subjects in cross-site associations. ResultsOur simulation studies demonstrate that stronger associations between two sites lead to greater efficiency loss in regression analysis when such dependence is ignored in modeling. In a case study involving samples collected from a study on the female urogenital microbiome with aging, our model leads to the detection of covariate associations of the vaginal and urine microbiomes that are otherwise not statistically significant under a similar regression model applied to the two sites separately. ConclusionsWe propose a latent factor model for microbiome sequencing data collected from multiple sites. It captures the presumptive underlying cross-site associations without compromising estimation accuracy or inference efficiency in the absence of such associations. In addition, our proposed model improves predictive performance by enabling the prediction of microbial abundance at one site based on observations from another. We also provide an extended framework that allows for clustering of subjects (samples) and cluster-specific levels of paired association. Under this extended framework, clusters can be classified according to their association strengths.