Search

Abstract Background Integrating advanced machine-learning (ML) algorithms into clinical practice is challenging and requires interdisciplinary collaboration to develop transparent, interpretable, and ethically sound clinical decision support (CDS) tools. We aimed to design a ML-driven CDS tool to predict opioid overdose risk and gather feedback for its integration into the University of Florida Health (UFHealth) electronic health record (EHR) system. Methods We used user-centered design methods to integrate the ML algorithm into the EHR system. The backend and UI design sub-teams collaborated closely, both informed by user feedback sessions. We conducted seven user feedback sessions with five UF Health primary care physicians (PCPs) to explore aspects of CDS tools, including workflow, risk display, and risk mitigation strategies. After customizing the tool based on PCPs’ feedback, we held two rounds of one-on-one usability testing sessions with 8 additional PCPs to gather feedback on prototype alerts. These sessions informed iterative UI design and backend processes, including alert frequency and reappearance circumstances. Results The backend process development identified needs and requirements from our team, information technology, UFHealth, and PCPs. Thirteen PCPs (male = 62%, White = 85%) participated across 7 user feedback sessions and 8 usability testing sessions. During the user feedback sessions, PCPs (n = 5) identified flaws such as the term “high risk” of overdose potentially leading to unintended consequences (e.g., immediate addiction services referrals), offered suggestions, and expressed trust in the tool. In the first usability testing session, PCPs (n = 4) emphasized the need for natural risk presentation (e.g., 1 in 200) and suggested displaying the alert multiple times yearly for at-risk patients. Another 4 PCPs in the second usability testing session valued the UFHealth-specific alert for managing new or unfamiliar patients, expressed concerns about PCPs’ workload when prescribing to high-risk patients, and recommended incorporating the details page into training sessions to enhance usability. Conclusions The final backend process for our CDS alert aligns with PCP needs and UFHealth standards. Integrating feedback from PCPs in the early development phase of our ML-driven CDS tool helped identify barriers and facilitators in the CDS integration process. This collaborative approach yielded a refined prototype aimed at minimizing unintended consequences and enhancing usability.

Abstract Racial and ethnic minorities bear a disproportionate burden of type 2 diabetes (T2D) and its complications, with social determinants of health (SDoH) recognized as key drivers of these disparities. Implementing efficient and effective social needs management strategies is crucial. We propose a machine learning analytic pipeline to calculate the individualized polysocial risk score (iPsRS), which can identify T2D patients at high social risk for hospitalization, incorporating explainable AI techniques and algorithmic fairness optimization. We use electronic health records (EHR) data from T2D patients in the University of Florida Health Integrated Data Repository, incorporating both contextual SDoH (e.g., neighborhood deprivation) and person-level SDoH (e.g., housing instability). After fairness optimization across racial and ethnic groups, the iPsRS achieved a C statistic of 0.71 in predicting 1-year hospitalization. Our iPsRS can fairly and accurately screen patients with T2D who are at increased social risk for hospitalization.

Abstract Background Previous research on the risk of dementia associated with education attainment, smoking status, and alcohol use disorder (AUD) has yielded inconsistent results, indicating potential heterogeneous treatment effects (HTEs) of these factors on dementia risk. Thus, this study aimed to identify the important variables that may contribute to HTEs of these factors in older adults. Methods Using 2005–2021 data from the National Alzheimer’s Coordinating Center (NACC), we included older adults (≥ 65 years) with normal cognition at the first visit. The exposure of interest included college education or above, current smoking, and AUD and the outcome was all-cause dementia. We applied doubly robust learning to estimate risk differences (RD) and 95% confidence intervals (CI) between exposed and unexposed groups in the overall cohort and subgroups identified through a decision tree model. Results Of 10,062 participants included, 929 developed all-cause dementia over a median 4.4-year follow-up. College education or above was associated with a lower risk of all-cause dementia in the overall population (RD, -1.5%; 95%CI, -2.8 to -0.3), especially among the subpopulations without hypertension, regardless of the APOE4 status. Current smoking was not related to increased dementia risk overall (2.8%; -1.5 to 7.2) but was significantly associated with increased dementia risk among men with (21.1%, 3.1 to 39.1) and without (8.4%, 0.9 to 15.8) cerebrovascular disease. AUD was not related to increased dementia risk overall (2.0%; -7.7 to 11.7) but was significantly associated with increased dementia risk among men with neuropsychiatric disorders (31.5%; 7.4 to 55.7). Conclusions Our studies identified important factors contributing to HTEs of education, smoking, and AUD on risk of all-cause dementia, suggesting an individualized approach is needed to address dementia disparities.

Abstract Personal and family history of suicidal thoughts and behaviors (PSH and FSH, respectively) are significant risk factors associated with suicides. Research is limited in automatic identification of such data from clinical notes in Electronic Health Records. This study developed deep learning (DL) tools utilizing transformer models (Bio_ClinicalBERT and GatorTron) to detect PSH and FSH in clinical notes derived from three academic medical centers, and compared their performance with a rule-based natural language processing tool. For detecting PSH, the rule-based approach obtained an F1-score of 0.75 ± 0.07, while the Bio_ClinicalBERT and GatorTron DL tools scored 0.83 ± 0.09 and 0.84 ± 0.07, respectively. For detecting FSH, the rule-based approach achieved an F1-score of 0.69 ± 0.11, compared to 0.89 ± 0.10 for Bio_ClinicalBERT and 0.92 ± 0.07 for GatorTron. Across sites, the DL tools identified more than 80% of patients at elevated risk for suicide who remain undiagnosed and untreated.

Abstract Background Breast cancer, with its high morbidity and mortality rates, is a significant global health burden. Traditional treatments—surgery, chemotherapy, and radiotherapy—are widely used but come with drawbacks such as recurrence, metastasis, and significant side effects, including damage to healthy tissues. To address these limitations, new therapeutic strategies are being developed. Peroxidases (POD) can catalyze excess H2O2 in the tumor microenvironment to generate reactive oxygen species (ROS), which induce cancer cell apoptosis by disrupting redox homeostasis and modulating apoptosis-related proteins. However, natural enzymes face challenges like poor stability, high cost, and sensitivity to environmental conditions, limiting their application in breast cancer treatment. Nanozymes, nanomaterials with enzyme-like activity, offer a promising alternative by overcoming these limitations. Methods In this study, we successfully prepared Au@Pd nanozymes with peroxidase activity by depositing metallic Pd on Au nanoparticles (Au NPs) synthesized using a trisodium citrate reduction method and ascorbic acid reduction. The in vitro validation was conducted through a series of experiments, including ROS detection, flow cytometry, CCK-8 assay, DNA damage assessment, live/dead cell staining, Western blot (WB), and qPCR. Tumor treatment was performed via tail vein injection of the drug, followed by HE staining of the treated tissues and biochemical analysis of the blood. Results Au@Pd nanozymes can effectively accumulate at the tumor site through the EPR effect and exert peroxidase-like activity, catalyzing the excess H2O2 in the tumor microenvironment to produce ROS. This triggers apoptosis pathways and DNA damage, leading to the downregulation of the anti-apoptotic protein Bcl-2, upregulation of the pro-apoptotic protein Bax, and induction of apoptosis-related genes, demonstrating strong anti-tumor effects. Conclusions This study developed an efficient nanozyme-mediated catalytic therapy strategy targeting the tumor microenvironment for the treatment of breast cancer cells.

Colorectal cancer (CRC) is the third most diagnosed cancer and the second deadliest cancer worldwide representing a major public health problem. In recent years, increasing evidence has shown that microRNA (miRNA) can control the expression of targeted human messenger RNA (mRNA) by reducing their abundance or translation, acting as oncogenes or tumor suppressors in various cancers, including CRC. Due to the significant up-regulation of oncogenic miRNAs in CRC, elucidating the underlying mechanism and identifying dysregulated miRNA targets may provide a basis for improving current therapeutic interventions. In this paper, we proposed Gra-CRC-miRTar, a pre-trained nucleotide-to-graph neural network framework, for identifying potential miRNA targets in CRC. Different from previous studies, we constructed two pre-trained models to encode RNA sequences and transformed them into de Bruijn graphs. We employed different graph neural networks to learn the latent representations. The embeddings generated from de Bruijn graphs were then fed into a Multilayer Perceptron (MLP) to perform the prediction tasks. Our extensive experiments show that Gra-CRC-miRTar achieves better performance than other deep learning algorithms and existing predictors. In addition, our analyses also successfully revealed 172 out of 201 functional interactions through experimentally validated miRNA-mRNA pairs in CRC. Collectively, our effort provides an accurate and efficient framework to identify potential miRNA targets in CRC, which can also be used to reveal miRNA target interactions in other malignancies, facilitating the development of novel therapeutics. The Gra-CRC-miRTar web server can be found at: http://gra-crc-mirtar.com/.

Abstract Background SARS-CoV-2-infected patients may develop new conditions in the period after the acute infection. These conditions, the post-acute sequelae of SARS-CoV-2 infection (PASC, or Long COVID), involve a diverse set of organ systems. Limited studies have investigated the predictability of Long COVID development and its associated risk factors. Methods In this retrospective cohort study, we used electronic healthcare records from two large-scale PCORnet clinical research networks, INSIGHT (~1.4 million patients from New York) and OneFlorida+ (~0.7 million patients from Florida), to identify factors associated with having Long COVID, and to develop machine learning-based models for predicting Long COVID development. Both SARS-CoV-2-infected and non-infected adults were analysed during the period of March 2020 to November 2021. Factors associated with Long COVID risk were identified by removing background associations and correcting for multiple tests. Results We observed complex association patterns between baseline factors and a variety of Long COVID conditions, and we highlight that severe acute SARS-CoV-2 infection, being underweight, and having baseline comorbidities (e.g., cancer and cirrhosis) are likely associated with increased risk of developing Long COVID. Several Long COVID conditions, e.g., dementia, malnutrition, chronic obstructive pulmonary disease, heart failure, PASC diagnosis U099, and acute kidney failure are well predicted (C-index > 0.8). Moderately predictable conditions include atelectasis, pulmonary embolism, diabetes, pulmonary fibrosis, and thromboembolic disease (C-index 0.7–0.8). Less predictable conditions include fatigue, anxiety, sleep disorders, and depression (C-index around 0.6). Conclusions This observational study suggests that association patterns between investigated factors and Long COVID are complex, and the predictability of different Long COVID conditions varies. However, machine learning-based predictive models can help in identifying patients who are at risk of developing a variety of Long COVID conditions.

Abstract Parkinson’s disease (PD) is a serious neurodegenerative disorder marked by significant clinical and progression heterogeneity. This study aimed at addressing heterogeneity of PD through integrative analysis of various data modalities. We analyzed clinical progression data (≥5 years) of individuals with de novo PD using machine learning and deep learning, to characterize individuals’ phenotypic progression trajectories for PD subtyping. We discovered three pace subtypes of PD exhibiting distinct progression patterns: the Inching Pace subtype (PD-I) with mild baseline severity and mild progression speed; the Moderate Pace subtype (PD-M) with mild baseline severity but advancing at a moderate progression rate; and the Rapid Pace subtype (PD-R) with the most rapid symptom progression rate. We found cerebrospinal fluid P-tau/α-synuclein ratio and atrophy in certain brain regions as potential markers of these subtypes. Analyses of genetic and transcriptomic profiles with network-based approaches identified molecular modules associated with each subtype. For instance, the PD-R-specific module suggested STAT3, FYN, BECN1, APOA1, NEDD4, and GATA2 as potential driver genes of PD-R. It also suggested neuroinflammation, oxidative stress, metabolism, PI3K/AKT, and angiogenesis pathways as potential drivers for rapid PD progression (i.e., PD-R). Moreover, we identified repurposable drug candidates by targeting these subtype-specific molecular modules using network-based approach and cell line drug-gene signature data. We further estimated their treatment effects using two large-scale real-world patient databases; the real-world evidence we gained highlighted the potential of metformin in ameliorating PD progression. In conclusion, this work helps better understand clinical and pathophysiological complexity of PD progression and accelerate precision medicine.

Abstract Alzheimer’s Disease (AD) pathology has been increasingly explored through single-cell and single-nucleus RNA-sequencing (scRNA-seq & snRNA-seq) and spatial transcriptomics (ST). However, the surge in data demands a comprehensive, user-friendly repository. Addressing this, we introduce a single-cell and spatial RNA-seq database for Alzheimer’s disease (ssREAD). It offers a broader spectrum of AD-related datasets, an optimized analytical pipeline, and improved usability. The database encompasses 1,053 samples (277 integrated datasets) from 67 AD-related scRNA-seq & snRNA-seq studies, totaling 7,332,202 cells. Additionally, it archives 381 ST datasets from 18 human and mouse brain studies. Each dataset is annotated with details such as species, gender, brain region, disease/control status, age, and AD Braak stages. ssREAD also provides an analysis suite for cell clustering, identification of differentially expressed and spatially variable genes, cell-type-specific marker genes and regulons, and spot deconvolution for integrative analysis. ssREAD is freely available at https://bmblx.bmi.osumc.edu/ssread/ .

At the Worldwide Developers Conference in June 2023, Apple introduced the Vision Pro. The Apple Vision Pro (AVP) is a mixed reality headset; more specifically, it is a virtual reality device with an additional video see-through capability. The video see-through capability turns the AVP into an augmented reality (AR) device. The AR feature is enabled by streaming the real world via cameras on the (virtual reality) screens in front of the user’s eyes. This is, of course, not unique and is similar to other devices, such as the Varjo XR-3 (Varjo Technologies Oy). Nevertheless, the AVP has some interesting features, such as an inside-out screen that can show the headset wearer’s eyes to “outsiders,” and a button on the top, called the “digital crown,” that allows a seamless blend of digital content with the user’s physical space by turning it. In addition, it is untethered, except for the cable to the battery, which makes the headset more agile, compared to the Varjo XR-3. This could actually come closer to “The Ultimate Display,” which Ivan Sutherland had already sketched in 1965. After a great response from the media and social networks to the release, we were able to test and review the new AVP ourselves in March 2024. Including an expert survey with 13 of our colleagues after testing the AVP in our institute, this Viewpoint explores whether the AVP can overcome clinical challenges that AR especially still faces in the medical domain; we also go beyond this and discuss whether the AVP could support clinicians in essential tasks to allow them to spend more time with their patients.

BackgroundAstrocytic activation in the spinal dorsal horn contributes to the central sensitization of neuropathic pain. Bone morphogenetic protein (BMP) 10, one of the BMPs highly expressed in the central nervous system, has been demonstrated to have an accelerated effect on astrocytic activation. This study aimed to investigate the functional effects of BMP10 on the activation of astrocytes in the spinal dorsal horn of animal model of neuropathic pain and to explore potential mechanisms involved in this process.MethodsA neuropathic pain mice model was established using the spared nerve injury (SNI). Western blot analysis was performed to detect the expressional levels of BMP10, activin receptor-like receptor 2 (ALK2), Smad1/5/8, phosphorylated Smad1/5/8, and glial fibrillary acidic protein (GFAP). Immunofluorescence staining was used to detect BMP10, ALK2, and GFAP distribution and expression. The behavioral changes in mice were evaluated using paw withdrawal threshold (PWT), thermal withdrawal latency (TWL), and open field test (OFT). The BMP10 siRNA, Smad1 siRNA, BMP10 peptide, and ALK2-IN-2 (ALK2 inhibitor) were intrathecally administrated to mice. A model of lipopolysaccharide (LPS)-stimulated astrocytes was established to investigate the effect of Smad1. The transfection efficiency of siRNAs was detected by western blot and qRT-PCR analysis.ResultsBMP10 levels were increased in the L4-6 ipsilateral spinal dorsal horn of SNI mice and particularly elevated in astrocytes. Consistently, GFAP and phosphorylated Smad1/5/8 were upregulated in the L4-6 ipsilateral spinal dorsal horn after SNI, indicating the activation of astrocytes and Smad1/5/8 signaling. An intrathecal injection of BMP10 siRNA abrogated pain hypersensitivity and astrocytic activation in SNI mice. In addition, intrathecal administration of BMP10 peptide evoked pain hypersensitivity and astrocytic activation in normal mice, and this action was reversed by inhibiting the ALK2. Furthermore, targeting Smad1 in vitro with the help of siRNA inhibited the activation of astrocytes induced by LPS. Finally, targeting Smad1 abrogated BMP10-induced hypersensitivity and activation of astrocytes.ConclusionThese findings indicate that the BMP10/ALK2/Smad1/5/8 axis plays a key role in pain hypersensitivity after peripheral nerve injury, which indicates its stimulative ability toward astrocytes.

Abstract Background Osteoarthritis (OA) is a common degenerative joint disease characterized by persistent articular cartilage degeneration and synovitis. Oxymatrine (OMT) is a quinzolazine alkaloid extracted from the traditional Chinese medicine, matrine, and possesses anti-inflammatory properties that may help regulate the pathogenesis of OA; however, its mechanism has not been elucidated. This study aimed to investigate the effects of OMT on interleukin-1β (IL-1β)-induced damage and the potential mechanisms of action. Methods Chondrocytes were isolated from Sprague-Dawley rats. Toluidine blue and Collagen II immunofluorescence staining were used to determine the purity of the chondrocytes. Thereafter, the chondrocytes were subjected to IL-1β stimulation, both in the presence and absence of OMT, or the autophagy inhibitor 3-methyladenine (3-MA). Cell viability was assessed using the MTT assay and SYTOX Green staining. Additionally, flow cytometry was used to determine cell apoptosis rate and reactive oxygen species (ROS) levels. The protein levels of AKT, mTOR, LC3, P62, matrix metalloproteinase-13, and collagen II were quantitatively analyzed using western blotting. Immunofluorescence was used to assess LC3 expression. Results OMT alleviated IL-1β-induced damage in chondrocytes, by increasing the survival rate, reducing the apoptosis rates of chondrocytes, and preventing the degradation of the cartilage matrix. In addition, OMT decreased the ROS levels and inhibited the AKT/mTOR signaling pathway while promoting autophagy in IL-1β treated chondrocytes. However, the effectiveness of OMT in improving chondrocyte viability under IL-1β treatment was limited when autophagy was inhibited by 3-MA. Conclusions OMT decreases oxidative stress and inhibits the AKT/mTOR signaling pathway to enhance autophagy, thus inhibiting IL-1β-induced damage. Therefore, OMT may be a novel and effective therapeutic agent for the clinical treatment of OA.

Abstract INTRODUCTION Alzheimer's disease (AD) is often misclassified in electronic health records (EHRs) when relying solely on diagnosis codes. This study aimed to develop a more accurate, computable phenotype (CP) for identifying AD patients using structured and unstructured EHR data. METHODS We used EHRs from the University of Florida Health (UFHealth) system and created rule‐based CPs iteratively through manual chart reviews. The CPs were then validated using data from the University of Texas Health Science Center at Houston (UTHealth) and the University of Minnesota (UMN). RESULTS Our best‐performing CP was “patient has at least 2 AD diagnoses and AD‐related keywords in AD encounters,” with an F1‐score of 0.817 at UF, 0.961 at UTHealth, and 0.623 at UMN, respectively. DISCUSSION We developed and validated rule‐based CPs for AD identification with good performance, which will be crucial for studies that aim to use real‐world data like EHRs. Highlights Developed a computable phenotype (CP) to identify Alzheimer's disease (AD) patients using EHR data. Utilized both structured and unstructured EHR data to enhance CP accuracy. Achieved a high F1‐score of 0.817 at UFHealth, and 0.961 and 0.623 at UTHealth and UMN. Validated the CP across different demographics, ensuring robustness and fairness.

ObjectivesTo investigate the risk of atrial fibrillation (AF) with sodium-glucose cotransporter-2 inhibitors (SGLT2is) compared to dipeptidyl peptidase-4 inhibitor (DPP4i) use in older US adults and across diverse subgroups.MethodsWe conducted a retrospective cohort analysis using claims data from 15% random samples of Medicare fee-for-service beneficiaries. Patients were adults with type 2 diabetes (T2D), no preexisting AF, and were newly initiated on SGLT2i or DPP4i. The outcome was the first incident AF. Inverse probability treatment weighting (IPTW) was used to balance the baseline covariates between the treatment groups including sociodemographics, comorbidities, and co-medications. Cox regression models were used to assess the effect of SGLT2i compared to DPP4i on incident AF.ResultsOf the 97,436 eligible individuals (mean age 71.2 ± 9.8 years, 54.6% women), 1.01% (n = 983) had incident AF over a median follow-up of 361 days. The adjusted incidence rate was 8.39 (95% CI: 6.67–9.99) and 11.70 (95% CI: 10.9–12.55) per 1,000 person-years in the SGLT2i and DPP4i groups, respectively. SGLT2is were associated with a significantly lower risk of incident AF (HR 0.73; 95% CI, 0.57 to 0.91; p = 0.01) than DPP4is. The risk reduction of incident AF was significant in non-Hispanic White individuals and subgroups with existing atherosclerotic cardiovascular diseases and chronic kidney disease.ConclusionCompared to the use of DPP4i, that of SGLT2i was associated with a lower risk of AF in patients with T2D. Our findings contribute to the real-world evidence regarding the effectiveness of SGLT2i in preventing AF and support a tailored therapeutic approach to optimize treatment selection based on individual characteristics.

During the COVID-19 pandemic, forecasting COVID-19 trends to support planning and response was a priority for scientists and decision makers alike. In the United States, COVID-19 forecasting was coordinated by a large group of universities, companies, and government entities led by the Centers for Disease Control and Prevention and the US COVID-19 Forecast Hub (https://covid19forecasthub.org). We evaluated approximately 9.7 million forecasts of weekly state-level COVID-19 cases for predictions 1-4 weeks into the future submitted by 24 teams from August 2020 to December 2021. We assessed coverage of central prediction intervals and weighted interval scores (WIS), adjusting for missing forecasts relative to a baseline forecast, and used a Gaussian generalized estimating equation (GEE) model to evaluate differences in skill across epidemic phases that were defined by the effective reproduction number. Overall, we found high variation in skill across individual models, with ensemble-based forecasts outperforming other approaches. Forecast skill relative to the baseline was generally higher for larger jurisdictions (e.g., states compared to counties). Over time, forecasts generally performed worst in periods of rapid changes in reported cases (either in increasing or decreasing epidemic phases) with 95% prediction interval coverage dropping below 50% during the growth phases of the winter 2020, Delta, and Omicron waves. Ideally, case forecasts could serve as a leading indicator of changes in transmission dynamics. However, while most COVID-19 case forecasts outperformed a naïve baseline model, even the most accurate case forecasts were unreliable in key phases. Further research could improve forecasts of leading indicators, like COVID-19 cases, by leveraging additional real-time data, addressing performance across phases, improving the characterization of forecast confidence, and ensuring that forecasts were coherent across spatial scales. In the meantime, it is critical for forecast users to appreciate current limitations and use a broad set of indicators to inform pandemic-related decision making.

Abstract Target trial emulation is the process of mimicking target randomized trials using real-world data, where effective confounding control for unbiased treatment effect estimation remains a main challenge. Although various approaches have been proposed for this challenge, a systematic evaluation is still lacking. Here we emulated trials for thousands of medications from two large-scale real-world data warehouses, covering over 10 years of clinical records for over 170 million patients, aiming to identify new indications of approved drugs for Alzheimer’s disease. We assessed different propensity score models under the inverse probability of treatment weighting framework and suggested a model selection strategy for improved baseline covariate balancing. We also found that the deep learning-based propensity score model did not necessarily outperform logistic regression-based methods in covariate balancing. Finally, we highlighted five top-ranked drugs (pantoprazole, gabapentin, atorvastatin, fluticasone, and omeprazole) originally intended for other indications with potential benefits for Alzheimer’s patients.

Abstract There are enormous enthusiasm and concerns in applying large language models (LLMs) to healthcare. Yet current assumptions are based on general-purpose LLMs such as ChatGPT, which are not developed for medical use. This study develops a generative clinical LLM, GatorTronGPT, using 277 billion words of text including (1) 82 billion words of clinical text from 126 clinical departments and approximately 2 million patients at the University of Florida Health and (2) 195 billion words of diverse general English text. We train GatorTronGPT using a GPT-3 architecture with up to 20 billion parameters and evaluate its utility for biomedical natural language processing (NLP) and healthcare text generation. GatorTronGPT improves biomedical natural language processing. We apply GatorTronGPT to generate 20 billion words of synthetic text. Synthetic NLP models trained using synthetic text generated by GatorTronGPT outperform models trained using real-world clinical text. Physicians’ Turing test using 1 (worst) to 9 (best) scale shows that there are no significant differences in linguistic readability (p = 0.22; 6.57 of GatorTronGPT compared with 6.93 of human) and clinical relevance (p = 0.91; 7.0 of GatorTronGPT compared with 6.97 of human) and that physicians cannot differentiate them (p

Accurate estimation of gross primary production (GPP) of paddy rice fields is essential for understanding cropland carbon cycles, yet remains challenging due to spatial heterogeneity. In this study, we integrated high-resolution unmanned aerial vehicle (UAV) imagery into a leaf biochemical properties-based model for improving GPP estimation. The key parameter, maximum carboxylation rate at the top of the canopy (Vcmax,025), was quantified using various spatial information representation methods, including mean (μref) and standard deviation (σref) of reflectance, gray-level co-occurrence matrix (GLCM)-based features, local binary pattern histogram (LBPH), and convolutional neural networks (CNNs). Our models were evaluated using a two-year eddy covariance (EC) system and UAV measurements. The result shows that incorporating spatial information can vastly improve the accuracy of Vcmax,025 and GPP estimation. CNN methods achieved the best Vcmax,025 estimation, with an R of 0.94, an RMSE of 19.44 μmol m−2 s−1, and an MdAPE of 11%, and further produced highly accurate GPP estimates, with an R of 0.92, an RMSE of 6.5 μmol m−2 s−1, and an MdAPE of 23%. The μref-GLCM texture features and μref-LBPH joint-driven models also gave promising results. However, σref contributed less to Vcmax,025 estimation. The Shapley value analysis revealed that the contribution of input features varied considerably across different models. The CNN model focused on nir and red-edge bands and paid much attention to the subregion with high spatial heterogeneity. The μref-LBPH joint-driven model mainly prioritized reflectance information. The μref-GLCM-based features joint-driven model emphasized the role of GLCM texture indices. As the first study to leverage the spatial information from high-resolution UAV imagery for GPP estimation, our work underscores the critical role of spatial information and provides new insight into monitoring the carbon cycle.

The escalating concern over global warming has garnered significant international attention, with carbon emission intensity emerging as a crucial barrier to sustainable economic development across various regions. While previous studies have largely focused on annual scales, this study introduces a novel examination of Tianjin’s quarterly carbon emission intensity and its influencing factors from 2012 to 2022 using quarterly data and the Logarithmic Mean Divisia Index (LMDI) model. The analysis considers the carbon emission effects of thermal power generation, the power supply structure, power intensity effects, and economic activity intensity. The results indicate a general decline in Tianjin’s carbon emission intensity from 2012 to 2020, followed by an increase in 2021 and 2022. This trend, exhibiting significant seasonal fluctuations, revealed the highest carbon emission intensity in the first quarter (an average of 1.4093) and the lowest in the second quarter (an average of 1.0019). Economic activity intensity emerged as the predominant factor influencing carbon emission intensity changes, particularly notable in the second quarter (an average of −0.0374). Thermal power generation and electricity intensity effects were significant in specific seasons, while the power supply structure’s impact remained relatively minor yet stable. These findings provide essential insights for formulating targeted carbon reduction strategies, underscoring the need to optimize energy structures, enhance energy efficiency, and account for the seasonal impacts of economic activity patterns on carbon emissions.

Partitioning of evapotranspiration (ET) into soil evaporation (E) and plant transpiration (T) poses a significant challenge. This study proposed a novel approach that combines eddy covariance, micro-lysimeter, and high-resolution unmanned aerial vehicle (UAV) images within the flux footprint to optimize the canopy and soil resistances of the Shuttleworth-Wallace model (S-W) using a multi-objective optimization scheme. A two-year experiment was conducted at the Yucheng and Luancheng sites to investigate the effectiveness of the proposed method for the summer maize-wheat winter system in the North China Plain. Our results showed that the S-W model with multi-objective optimization by using high-resolution UAV images within the flux footprint significantly improved the estimation accuracy of ET components compared with the traditional one- and two-objective optimization schemes. The estimation of ET, T, and E of our method had an average root mean square error of 0.67 mm day−1, 0.66 mm day−1, and 0.28 mm day−1, respectively. The optimized S-W model using high-resolution UAV data within the flux footprint produced better ET partitioning than that optimized using Moderate Resolution Imaging Spectroradiometer (MODIS) data. We then applied the proposed method to estimate ET and ET components in the North China Plain. This study highlighted the importance of utilizing multi-source data (especially high-resolution UAV images) and multi-objective optimization for accurate ET partitioning.

Abstract Durable superhydrophobic anti‐erosion/anticorrosion coatings are highly demanded across various applications. However, achieving coatings with exceptional superhydrophobicity, mechanical strength, and corrosion resistance remains a grand challenge. Herein, a robust microstructure coating, inspired by the cylindrical structures situated on the surface of conch shell, for mitigating erosion and corrosion damages in gas transportation pipelines is reported. Specifically, citric acid monohydrate as a pore‐forming agent is leveraged to create a porous structure between layers, effectively buffering the impact on the surface. As a result, the coating demonstrates remarkable wear resistance and water repellency. Importantly, even after abrasion by sandpaper and an erosion loop test, the resulting superhydrophobic surfaces retain the water repellency. The design strategy offers a promising route to manufacturing multifunctional materials with desired features and structural complexities, thereby enabling effective self‐cleaning and antifouling abilities in harsh operating environments for an array of applications, including self‐cleaning windows, antifouling coatings for medical devices, and anti‐erosion/anticorrosion protection, among other areas.

The use of a supersonic separator presents a novel approach for separating ammonia (NH3) from hydrogen (H2). Even trace amounts of water (H2O) can significantly influence the condensation characteristics of NH3. This study presents mathematical models for homogeneous NH3 and H2O condensation and heterogeneous NH3–H2O condensation. Through computational analysis, we revealed that the droplet radius and liquid mass fraction of a single condensable gas condensation increased accordingly with increasing inlet condensable gas content. Interactions between the condensation processes of NH3 and H2O were also captured in the simulation. Notably, at 100 ppm H2O content, H2O spontaneously condensed before NH3, serving as a condensation nucleus for NH3, and the heterogeneous condensation of NH3 decreased the rate of its spontaneous condensation. At an inlet H2O content of 50 ppm, the onset of spontaneous NH3 condensation coincided with that of H2O, with H2O having a negligible effect on NH3 condensation. The impact of H2O on NH3 condensation transitioned from minimal to inhibitory and then to facilitative as the content of H2O increased from 25 ppm to 250 ppm. At the nozzle outlet, the liquid mass fraction of NH3 initially decreased from 0.0414 to 0.0352 and then increased to 0.0446.

Background The rapid evolution of artificial intelligence (AI) in conjunction with recent updates in dual antiplatelet therapy (DAPT) management guidelines emphasizes the necessity for innovative models to predict ischemic or bleeding events after drug‐eluting stent implantation. Leveraging AI for dynamic prediction has the potential to revolutionize risk stratification and provide personalized decision support for DAPT management. Methods and Results We developed and validated a new AI‐based pipeline using retrospective data of drug‐eluting stent‐treated patients, sourced from the Cerner Health Facts data set (n=98 236) and Optum's de‐identified Clinformatics Data Mart Database (n=9978). The 36 months following drug‐eluting stent implantation were designated as our primary forecasting interval, further segmented into 6 sequential prediction windows. We evaluated 5 distinct AI algorithms for their precision in predicting ischemic and bleeding risks. Model discriminative accuracy was assessed using the area under the receiver operating characteristic curve, among other metrics. The weighted light gradient boosting machine stood out as the preeminent model, thus earning its place as our AI‐DAPT model. The AI‐DAPT demonstrated peak accuracy in the 30 to 36 months window, charting an area under the receiver operating characteristic curve of 90% [95% CI, 88%–92%] for ischemia and 84% [95% CI, 82%–87%] for bleeding predictions. Conclusions Our AI‐DAPT excels in formulating iterative, refined dynamic predictions by assimilating ongoing updates from patients' clinical profiles, holding value as a novel smart clinical tool to facilitate optimal DAPT duration management with high accuracy and adaptability.

Abstract Background Prognostic models of hospital-induced delirium, that include potential predisposing and precipitating factors, may be used to identify vulnerable patients and inform the implementation of tailored preventive interventions. It is recommended that, in prediction model development studies, candidate predictors are selected on the basis of existing knowledge, including knowledge from clinical practice. The purpose of this article is to describe the process of identifying and operationalizing candidate predictors of hospital-induced delirium for application in a prediction model development study using a practice-based approach. Methods This study is part of a larger, retrospective cohort study that is developing prognostic models of hospital-induced delirium for medical-surgical older adult patients using structured data from administrative and electronic health records. First, we conducted a review of the literature to identify clinical concepts that had been used as candidate predictors in prognostic model development-and-validation studies of hospital-induced delirium. Then, we consulted a multidisciplinary task force of nine members who independently judged whether each clinical concept was associated with hospital-induced delirium. Finally, we mapped the clinical concepts to the administrative and electronic health records and operationalized our candidate predictors. Results In the review of 34 studies, we identified 504 unique clinical concepts. Two-thirds of the clinical concepts (337/504) were used as candidate predictors only once. The most common clinical concepts included age (31/34), sex (29/34), and alcohol use (22/34). 96% of the clinical concepts (484/504) were judged to be associated with the development of hospital-induced delirium by at least two members of the task force. All of the task force members agreed that 47 or 9% of the 504 clinical concepts were associated with hospital-induced delirium. Conclusions Heterogeneity among candidate predictors of hospital-induced delirium in the literature suggests a still evolving list of factors that contribute to the development of this complex phenomenon. We demonstrated a practice-based approach to variable selection for our model development study of hospital-induced delirium. Expert judgement of variables enabled us to categorize the variables based on the amount of agreement among the experts and plan for the development of different models, including an expert-model and data-driven model.

Objective: To establish, apply, and evaluate a computable phenotype for the recruitment of individuals with successful cognitive aging. Participants and Methods: Interviews with 10 aging experts identified electronic health record (EHR)-available variables representing successful aging among individuals aged 85 years and older. On the basis of the identified variables, we developed a rule-based computable phenotype algorithm composed of 17 eligibility criteria. Starting September 1, 2019, we applied the computable phenotype algorithm to all living persons aged 85 years and older at the University of Florida Health, which identified 24,024 individuals. This sample was comprised of 13,841 (58%) women, 13,906 (58%) Whites, and 16,557 (69%) non-Hispanics. A priori permission to be contacted for research had been obtained for 11,898 individuals, of whom 470 responded to study announcements and 333 consented to evaluation. Then, we contacted those who consented to evaluate whether their cognitive and functional status clinically met out successful cognitive aging criteria of a modified Telephone Interview for Cognitive Status score of more than 27 and Geriatric Depression Scale of less than 6. The study was completed on December 31, 2022. Results: Of the 45% of living persons aged 85 years and older included in the University of Florida Health EHR database identified by the computable phenotype as successfully aged, approximately 4% of these responded to study announcements and 333 consented, of which 218 (65%) met successful cognitive aging criteria through direct evaluation. Conclusion: The study evaluated a computable phenotype algorithm for the recruitment of individuals for a successful aging study using large-scale EHRs. Our study provides proof of concept of using big data and informatics as aids for the recruitment of individuals for prospective cohort studies.

The demand for freshwater is increasing with population growth and rapid socio-economic development. It is more and more important for refined irrigation water management to conduct research on crop evapotranspiration (ET) data with a high spatiotemporal resolution in agricultural regions. We propose the unmixing–weight ET image fusion model (UWET), which integrates the advantages of the unmixing method in spatial downscaling and the weight-based method in temporal prediction to produce daily ET maps with a high spatial resolution. The Landsat-ET and MODIS-ET datasets for the UWET fusion data are retrieved from Landsat and MODIS images based on the surface energy balance model. The UWET model considers the effects of crop phenology, precipitation, and land cover in the process of the ET image fusion. The precision evaluation is conducted on the UWET results, and the measured ET values are monitored by eddy covariance at the Luancheng station, with average MAE values of 0.57 mm/day. The image results of UWET show fine spatial details and capture the dynamic ET changes. The seasonal ET values of winter wheat from the ET map mainly range from 350 to 660 mm in 2019–2020 and from 300 to 620 mm in 2020–2021. The average seasonal ET in 2019–2020 is 499.89 mm, and in 2020–2021, it is 459.44 mm. The performance of UWET is compared with two other fusion models: the Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM) and the Spatial and Temporal Reflectance Unmixing Model (STRUM). UWET performs better in the spatial details than the STARFM and is better in the temporal characteristics than the STRUM. The results indicate that UWET is suitable for generating ET products with a high spatial–temporal resolution in agricultural regions.

Abnormal aggregation and accumulation of pathological amyloid proteins such as amyloid-β, Tau, and α-synuclein play key pathological roles and serve as histological hallmarks in different neurodegenerative diseases (NDs) such as Alzheimer's disease (AD) and Parkinson's disease (PD). In addition, various post-translational modifications (PTMs) have been identified on pathological amyloid proteins and are subjected to change during disease progression. Given the central role of amyloid proteins in NDs, tremendous efforts have been made to develop amyloid-targeting strategies for clinical diagnosis and molecular classification of NDs. In this review, we summarize two major strategies for targeting amyloid aggregates, with a focus on the trials in AD diagnosis. The first strategy is a positron emission tomography (PET) scan of protein aggregation in the brain. We mainly focus on introducing the development of small-molecule PET tracers for specifically recognizing pathological amyloid fibrils. The second strategy is the detection of PTM biomarkers on amyloid proteins in cerebrospinal fluid and plasma. We discuss the pathological roles of different PTMs in diseases and how we can use the PTM profile of amyloid proteins for clinical diagnosis. Finally, we point out the potential technical challenges of these two strategies, and outline other potential strategies, as well as a combination of multiple strategies, for molecular diagnosis of NDs.

Abstract Background Allostatic load (AL) has been linked to an increased risk of death in various populations. However, to date, there is no research specifically investigating the effect of AL on mortality in older cancer survivors. Aims To investigate the association between AL and mortality in older cancer survivors. Method A total of 1291 adults aged 60 years or older who survived for ≥1 year since cancer diagnoses were identified from the 1999 to 2010 National Health and Nutrition Examination Survey. AL was the exposure of interest incorporating nine clinical measures/biomarkers; one point was added to AL if any of the measures/biomarkers exceeded the normal level. The sum of points was categorized as an ordinal variable to reflect low, moderate, and high ALs. Our outcomes of interest were all‐cause, cancer‐specific, and cardiovascular disease–specific mortality. Death was identified by linkage to the National Death Index. Multivariable Cox proportional hazards models were used to estimate adjusted hazard ratio (aHR) and 95% confidence intervals (CI) of mortality by AL category. Results Overall, 53.6% of participants were male and 78.4% were white. The mean age of study participants at interview was 72.8 years (standard deviation = 7.1). A total of 546 participants died during the follow‐up (median follow‐up time: 8.0 years). Among them, 158 died of cancer, and 106 died of cardiovascular events. Results from multivariable Cox proportional hazards models showed that higher ALS was positively associated with higher all‐cause mortality (ALS = 4–9 vs. ALS = 0–1: aHR = 1.52, 95% CI = 1.17–1.98, p‐trend

Abstract Background Prior studies on the association between asthma and cancer show inconsistent results. This study aimed to generate additional evidence on the association between asthma and cancer, both overall, and by cancer type, in the United States. Method We conducted a retrospective cohort study using 2012–2020 electronic health records and claims data in the OneFlorida+ clinical research network. Our study population included a cohort of adult patients with asthma (n = 90,021) and a matching cohort of adult patients without asthma (n = 270,063). We built Cox proportional hazards models to examine the association between asthma diagnosis and subsequent cancer risk. Results Our results showed that asthma patients were more likely to develop cancer compared to patients without asthma in multivariable analysis (hazard ratio [HR] = 1.36, 99% confidence interval [CI] = 1.29–1.44). Elevated cancer risk was observed in asthma patients without (HR = 1.60; 99% CI: 1.50–1.71) or with (HR = 1.11; 99% CI: 1.03–1.21) inhaled steroid use. However, in analyses of specific cancer types, cancer risk was elevated for nine of 13 cancers in asthma patients without inhaled steroid use but only for two of 13 cancers in asthma patients with inhaled steroid use, suggesting a protective effect of inhaled steroid use on cancer. Conclusion This is the first study to report a positive association between asthma and overall cancer risk in the US population. More in‐depth studies using real‐word data are needed to further explore the causal mechanisms of asthma on cancer risk.

BackgroundPrior studies have shown disparities in the uptake of cardioprotective newer glucose-lowering drugs (GLDs), including sodium-glucose cotranwsporter-2 inhibitors (SGLT2i) and glucagon-like peptide-1 receptor agonists (GLP1a). This study aimed to characterize geographic variation in the initiation of newer GLDs and the geographic variation in the disparities in initiating these medications.MethodsUsing 2017-2018 claims data from a 15% random nationwide sample of Medicare Part D beneficiaries, we identified individuals diagnosed with type 2 diabetes (T2D), who had ≥1 GLD prescriptions, and did not use SGLT2i or GLP1a in the year prior to the index date,1/1/2018. Patients were followed up for a year. The cohort was spatiotemporally linked to Dartmouth hospital-referral regions (HRRs), with each patient assigned to 1 of 306 HRRs. We performed multivariable Poisson regression to estimate adjusted initiation rates, and multivariable logistic regression to assess racial disparities in each HRR.ResultsAmong 795,469 individuals with T2D included in the analyses, the mean (SD) age was 73 (10) y, 53.3% were women, 12.2% were non-Hispanic Black, and 7.2% initiated a newer GLD in the follow-up year. In the adjusted model including clinical factors, compared to non-Hispanic White patients, non-Hispanic Black (initiation rate ratio, IRR [95% CI]: 0.66 [0.64-0.68]), American Indian/Alaska Native (0.74 [0.66-0.82]), Hispanic (0.85 [0.82-0.87]), and Asian/Pacific islander (0.94 [0.89-0.98]) patients were less likely to initiate newer GLDs. Significant geographic variation was observed across HRRs, with an initiation rate spanning 2.7%-13.6%.ConclusionsThis study uncovered substantial geographic variation and the racial disparities in initiating newer GLDs.

Introduction Traditional survey-based surveillance is costly, limited in its ability to distinguish diabetes types and time-consuming, resulting in reporting delays. The Diabetes in Children, Adolescents and Young Adults (DiCAYA) Network seeks to advance diabetes surveillance efforts in youth and young adults through the use of large-volume electronic health record (EHR) data. The network has two primary aims, namely: (1) to refine and validate EHR-based computable phenotype algorithms for accurate identification of type 1 and type 2 diabetes among youth and young adults and (2) to estimate the incidence and prevalence of type 1 and type 2 diabetes among youth and young adults and trends therein. The network aims to augment diabetes surveillance capacity in the USA and assess performance of EHR-based surveillance. This paper describes the DiCAYA Network and how these aims will be achieved.Methods and analysis The DiCAYA Network is spread across eight geographically diverse US-based centres and a coordinating centre. Three centres conduct diabetes surveillance in youth aged 0–17 years only (component A), three centres conduct surveillance in young adults aged 18–44 years only (component B) and two centres conduct surveillance in components A and B. The network will assess the validity of computable phenotype definitions to determine diabetes status and type based on sensitivity, specificity, positive predictive value and negative predictive value of the phenotypes against the gold standard of manually abstracted medical charts. Prevalence and incidence rates will be presented as unadjusted estimates and as race/ethnicity, sex and age-adjusted estimates using Poisson regression.Ethics and dissemination The DiCAYA Network is well positioned to advance diabetes surveillance methods. The network will disseminate EHR-based surveillance methodology that can be broadly adopted and will report diabetes prevalence and incidence for key demographic subgroups of youth and young adults in a large set of regions across the USA.

In this study, the authors characterise post-acute sequelae of SARS-CoV-2 (PASC) in two large cohorts based on electronic health records from the USA. They identify a broad range of PASC-related conditions which were only partially replicated across the two cohorts, indicating possible heterogeneity between populations.

This paper explores a novel structure aimed at enhancing its blast resistance performance by adding a layer of polyurea coating to the steel-PVC foam-steel sandwich panel. The response of 13 different arrangements of sandwich panels under explosive loading was studied using numerical simulation. The response process can be divided into three deformation stages: (1) Fluid-structure interaction; (2) Compression of the sandwich panel; (3) Dynamic structural response. The dynamic responses of the various sandwich panels to close-range air blast loading were analyzed based on the deformation characteristics, deflection, effective plastic strain, energy absorption, and pressure of the shock wave. The study draws the following conclusions: Reasonably adding a layer of polyurea to the traditional PVC foam sandwich panel can enhance its resistance to shock wave absorption, with a maximum increase of 29.8%; the optimal arrangement for explosion resistance is steel plate-PVC foam-polyurea-steel plate when the polyurea is coated on the back; and the best quality ratio between polyurea and PVC foam is 1:7 when the polyurea is coated on the front.

In order to mitigate the risk of roof-dominated coal burst in underground coal mining, horizontal long borehole staged hydraulic fracturing technology has been prevailingly employed to facilitate the weakening treatment of the hard roof in advance. Such weakening effect, however, can hardly be evaluated, which leads to a lack of a basis in which to design the schemes and parameters of hydraulic fracturing. In this study, a combined underground–ground integrated microseismic monitoring and transient electromagnetic detection method was utilized to carry out simultaneous evaluations of the seismic responses to each staged fracturing and the apparent resistivity changes before and after all finished fracturing. On this basis, the comparable and applicable fracturing effects on coal burst prevention were evaluated and validated by the distribution of microseismic events and their energy magnitude during the mining process. Results show that the observed mining-induced seismic events are consistent with the evaluation results obtained from the combined seismic-electromagnetic detection method. However, there is a limited reduction effect on resistivity near the fractured section that induces far-field seismic events. Mining-induced seismic events are concentrated primarily within specific areas, while microseismic events in the fractured area exhibit high frequency but low energy overall. This study validates the rationality of combined seismic-electromagnetic detection results and provides valuable insights for optimizing fracturing construction schemes as well as comprehensively evaluating outcomes associated with underground directional long borehole staged hydraulic fracturing.

The parameters of the GR4J-CemaNeige coupling model (GR4neige) are typically treated as constants. However, the maximum capacity of the production store (parX1) exhibits time-varying characteristics due to climate variability and vegetation coverage change. This study employed differentiable parameter learning (dPL) to identify the time-varying parX1 in the GR4neige across 671 catchments within the United States. We built two types of dPL, including static and dynamic parameter networks, to assess the advantages of the time-varying parameter. In the dynamic parameter network, we evaluated the impact of potential evapotranspiration (PET), precipitation (P), temperature (T), soil moisture (SM), and normalized difference vegetation index (NDVI) datasets on the performance of dPL. We then compared dPL with the empirical functional method (fm). The results demonstrated that the dynamic parameter network outperformed the static parameter network in streamflow estimation. There were differences in streamflow estimation among the dynamic parameter network driven by various input features. In humid catchments, simultaneously incorporating all five factors, including PET, P, T, SM, and the NDVI, achieved optimal streamflow simulation accuracy. In arid catchments, it was preferable to introduce PET, T, and the NDVI separately for improved performance. dPL significantly outperformed the empirical fm in estimating streamflow and uncalibrated intermediate variables, like evapotranspiration (ET). Both the derived parX1 from dPL and the empirical fm exhibited significant spatiotemporal variation across 671 catchments. Notably, compared to parX1 obtained through the empirical fm, parX1 derived from dPL exhibited a distinct spatial clustering pattern. This study highlights the potential of dPL in enhancing model accuracy and contributes to understanding the spatiotemporal variation characteristics of parX1 under the influence of climate factors, soil conditions, and vegetation change.

Although lung cancer is a leading cause of death among people living with HIV (PLWH), limited research exists characterizing real-world lung cancer screening adherence among PLWH. Our objective was to compare low-dose computed tomography (LDCT) adherence among PLWH to those without HIV treated at one integrated health system. Using the University of Florida’s Health Integrated Data Repository (01/01/2012–10/31/2021), we identified PLWH with at least one LDCT procedure, using Current Procedural Terminology codes(S8032/G0297/71271). Lung cancer screening adherence was defined as a second LDCT based on the Lung Imaging Reporting and Data System (Lung-RADS®). Lung-RADS categories were extracted from radiology reports using a natural language processing system. PLWH were matched with 4 randomly selected HIV-negative patients based on (+/- 1 year) age, Lung-RADS category, and calendar year. Seventy-three PLWH and 292 matched HIV-negative adults with at least one LDCT were identified. PLWH were more likely to be male (66% vs.52%,p

Abstract Overly restrictive eligibility criteria for clinical trials may limit the generalizability of the trial results to their target real-world patient populations. We developed a novel machine learning approach using large collections of real-world data (RWD) to better inform clinical trial eligibility criteria design. We extracted patients’ clinical events from electronic health records (EHRs), which include demographics, diagnoses, and drugs, and assumed certain compositions of these clinical events within an individual’s EHRs can determine the subphenotypes—homogeneous clusters of patients, where patients within each subgroup share similar clinical characteristics. We introduced an outcome-guided probabilistic model to identify those subphenotypes, such that the patients within the same subgroup not only share similar clinical characteristics but also at similar risk levels of encountering severe adverse events (SAEs). We evaluated our algorithm on two previously conducted clinical trials with EHRs from the OneFlorida+ Clinical Research Consortium. Our model can clearly identify the patient subgroups who are more likely to suffer or not suffer from SAEs as subphenotypes in a transparent and interpretable way. Our approach identified a set of clinical topics and derived novel patient representations based on them. Each clinical topic represents a certain clinical event composition pattern learned from the patient EHRs. Tested on both trials, patient subgroup (#SAE=0) and patient subgroup (#SAE>0) can be well-separated by k-means clustering using the inferred topics. The inferred topics characterized as likely to align with the patient subgroup (#SAE>0) revealed meaningful combinations of clinical features and can provide data-driven recommendations for refining the exclusion criteria of clinical trials. The proposed supervised topic modeling approach can infer the clinical topics from the subphenotypes with or without SAEs. The potential rules for describing the patient subgroups with SAEs can be further derived to inform the design of clinical trial eligibility criteria.

Abstract There is an increasing interest in developing artificial intelligence (AI) systems to process and interpret electronic health records (EHRs). Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. However, there are few clinical language models, the largest of which trained in the clinical domain is comparatively small at 110 million parameters (compared with billions of parameters in the general domain). It is not clear how large clinical language models with billions of parameters can help medical AI systems utilize unstructured EHRs. In this study, we develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text) and systematically evaluate it on five clinical NLP tasks including clinical concept extraction, medical relation extraction, semantic textual similarity, natural language inference (NLI), and medical question answering (MQA). We examine how (1) scaling up the number of parameters and (2) scaling up the size of the training data could benefit these NLP tasks. GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks (e.g., 9.6% and 9.5% improvement in accuracy for NLI and MQA), which can be applied to medical AI systems to improve healthcare delivery. The GatorTron models are publicly available at: https://catalog.ngc.nvidia.com/orgs/nvidia/teams/clara/models/gatortron_og .

The extensive usage of fossil fuels has caused significant environmental pollution, climate change and energy crises. The significant advantages of hydrogen, such as cleanliness, high efficiency, and a wide range of sources, make it quite promising. Hydrogen is prone to material damage, which may lead to leakage. High-pressure leaking hydrogen is highly susceptible to spontaneous combustion due to its combustion characteristics, which may cause jet fire or explosion accidents, resulting in serious casualties and property damage. This paper presents a detailed review of the research progress on hydrogen leak diffusion characteristics, leak spontaneous combustion mechanisms and material hydrogen damage mechanisms from the perspectives of theoretical analysis, experiments and numerical simulations. This review points out that although a large number of research results have been obtained on the safety characteristics of hydrogen, there are still some deficiencies and limitations. Further research topics are clarified, such as further optimizing the kinetic mechanism of the high-pressure hydrogen leakage reaction and turbulence model, exploring the expansion and dilution law of hydrogen clouds after liquid hydrogen flooding, further studying the spontaneous combustion mechanism of leaked hydrogen and the interaction between mechanisms, and investigating the synergistic damage effect of hydrogen and other components on materials. The leakage spontaneous combustion process in open space, the development process of the bidirectional effect of hydrogen jet fuel and crack growth under the impact of high-pressure hydrogen jet fuel on the material may need to be explored next.

Abstract Objective The aim of this study was to determine whether a secure, privacy-preserving record linkage (PPRL) methodology can be implemented in a scalable manner for use in a large national clinical research network. Results We established the governance and technical capacity to support the use of PPRL across the National Patient-Centered Clinical Research Network (PCORnet®). As a pilot, four sites used the Datavant software to transform patient personally identifiable information (PII) into de-identified tokens. We queried the sites for patients with a clinical encounter in 2018 or 2019 and matched their tokens to determine whether overlap existed. We described patient overlap among the sites and generated a “deduplicated” table of patient demographic characteristics. Overlapping patients were found in 3 of the 6 site-pairs. Following deduplication, the total patient count was 3,108,515 (0.11% reduction), with the largest reduction in count for patients with an “Other/Missing” value for Sex; from 198 to 163 (17.6% reduction). The PPRL solution successfully links patients across data sources using distributed queries without directly accessing patient PII. The overlap queries and analysis performed in this pilot is being replicated across the full network to provide additional insight into patient linkages among a distributed research network.

The back pressure at the outlet of supersonic cyclone separator usually causes aerodynamic shock wave in Laval nozzle, and the interaction of supersonic flow, shock wave and phase transition is very complicated. In order to explore the supersonic flow properties of wet gas under the action of shock wave, study was performed on the methane-water vapor mixtures. Then, a mathematical model for the supersonic condensation and flow of wet natural gas was established. Meanwhile, the effect of the outlet back pressure and inlet pressure on the supersonic condensation property of wet natural gas was analyzed to evaluate the performance of the supersonic cyclone separator under the action of shock waves. The simulation results show that the shock waves can cause the rapid increase in pressure and temperature of water vapor in a short time, leading to the condensed droplets re-evaporate and the liquid mass fraction at outlet reduces sharply. Besides, as the inlet pressure increases, the location of nucleation approaches the throat of Laval nozzle gradually, and the limit nucleation rate and the droplet number decrease accordingly. However, the liquid mass fraction increases due to the increasing of growth rate and sizes of droplets. Moreover, the nozzle loses its function of liquefaction when the back pressure is higher than 25 kPa (i.e., the pressure ratio is higher than 0.58), and then the nozzle cannot operate normally. Therefore, it is necessary to control the reasonable pressure ratio to ensure the dehydration efficiency of the supersonic separator with back pressure.

Background Hispanic populations are more likely to develop diabetes and its related diseases than non‐Hispanic White populations. Little evidence exists to support whether the cardiovascular and renal benefits of sodium‐glucose cotransporter 2 inhibitors and glucagon‐like peptide‐1 receptor agonists are generalizable to the Hispanic populations. Methods and Results We included the cardiovascular and renal outcome trials (up to March 2021) that reported the major adverse cardiovascular events (MACEs), cardiovascular death/hospitalization for heart failure, and composite renal outcomes by ethnicity in individuals with type 2 diabetes (T2D), calculated pooled hazard ratios (HRs) with 95% CIs using fixed‐effects models, and tested the differences between Hispanic and non‐Hispanic populations (P for interaction [Pinteraction]). In 3 sodium‐glucose cotransporter 2 inhibitor trials, there was a statistically significant difference between Hispanic (HR, 0.70 [95% CI, 0.54–0.91]) and non‐Hispanic (HR, 0.96 [95% CI, 0.86–1.07]) groups in treatment effects on MACE risk (Pinteraction=0.03), except for risks of cardiovascular death/hospitalization for heart failure (Pinteraction=0.46) and composite renal outcome (Pinteraction=0.31). In 5 glucagon‐like peptide‐1 receptor agonist trials, there was no statistically significant difference in treatment effect on MACE risk between Hispanic (HR, 0.82 [95% CI, 0.70–0.96]) and non‐Hispanic (HR, 0.92 [95% CI, 0.84–1.00]) populations (Pinteraction=0.22). In 3 dipeptidyl peptidase‐4 inhibitor trials, the HR for MACE risk appeared greater in Hispanic (HR, 1.15 [95% CI, 0.98–1.35]) than non‐Hispanic (HR, 0.96 [95% CI, 0.88–1.04]) populations (Pinteraction=0.045). Conclusions Compared with non‐Hispanic individuals, Hispanic individuals with T2D appeared to obtain a greater benefit of lowered MACE risk with sodium‐glucose cotransporter 2 inhibitors.

It is prudent to take a unified approach to exploring how contextual social determinants of health (SDoH) relate to COVID-19 occurrence and outcomes. Poor geographically represented data and a small number of contextual SDoH examined in most previous research studies have left a knowledge gap in the relationships between contextual SDoH and COVID-19 outcomes. In this study, we linked 199 contextual SDoH factors covering 11 domains of social and built environments with electronic health records (EHRs) from a large clinical research network (CRN) in the National Patient-Centered Clinical Research Network (PCORnet) to explore the relation between contextual SDoH and COVID-19 occurrence and hospitalization. We identified 15,890 COVID-19 patients and 63,560 matched non-COVID-19 patients in Florida between January 2020 and May 2021. We adopted a two-phase multiple linear regression approach modified from that in the exposome-wide association (ExWAS) study. After removing the highly correlated SDoH variables, 86 contextual SDoH variables were included in the data analysis. Adjusting for race, ethnicity, and comorbidities, we found six contextual SDoH variables (i.e., hospital available beds and utilization, percent of vacant property, number of golf courses, and percent of minority) related to the occurrence of COVID-19, and three variables (i.e., farmers market, low access, and religion) related to the hospitalization of COVID-19. To our best knowledge, this is the first study to explore the relationship between contextual SDoH and COVID-19 occurrence and hospitalization using EHRs in a major PCORnet CRN. As an exploratory study, the causal effect of SDoH on COVID-19 outcomes will be evaluated in future studies.

Abstract Background Dietary supplements (DS) have been widely used by consumers, but the information around the efficacy and safety of DS is disparate or incomplete, thus creating barriers for consumers to find information effectively. Conversational agent (CA) systems have been applied to healthcare domain, but there is no such system to answer consumers regarding DS use, although widespread use of DS. In this study, we develop the first CA system for DS use. Methods Our CA system for DS use developed on the MindMeld framework, consists of three components: question understanding, DS knowledge base, and answer generation. We collected and annotated 1509 questions to develop a natural language understanding module (e.g., question type classifier, named entity recognizer) which was then integrated into MindMeld framework. CA then queries the DS knowledge base (i.e., iDISK) and generates answers using rule-based slot filling techniques. We evaluated the algorithms of each component and the CA system as a whole. Results CNN is the best question classifier with an F1 score of 0.81, and CRF is the best named entity recognizer with an F1 score of 0.87. The system achieves an overall accuracy of 81% and an average score of 1.82 with succ@3 + score of 76.2% and succ@2 + of 66% approximately. Conclusion This study develops the first CA system for DS use using the MindMeld framework and iDISK domain knowledge base.

Abstract Background The US continues to face public health crises related to both chronic pain and opioid overdoses. Thirty percent of Americans suffer from chronic noncancer pain at an estimated yearly cost of over $600 billion. Most patients with chronic pain turn to primary care clinicians who must choose from myriad treatment options based on relative risks and benefits, patient history, available resources, symptoms, and goals. Recently, with attention to opioid-related risks, prescribing has declined. However, clinical experts have countered with concerns that some patients for whom opioid-related benefits outweigh risks may be inappropriately discontinued from opioids. Unfortunately, primary care clinicians lack usable tools to help them partner with their patients in choosing pain treatment options that best balance risks and benefits in the context of patient history, resources, symptoms, and goals. Thus, primary care clinicians and patients would benefit from patient-centered clinical decision support (CDS) for this shared decision-making process. Methods The objective of this 3-year project is to study the adaptation and implementation of an existing interoperable CDS tool for pain treatment shared decision making, with tailored implementation support, in new clinical settings in the OneFlorida Clinical Research Consortium. Our central hypothesis is that tailored implementation support will increase CDS adoption and shared decision making. We further hypothesize that increases in shared decision making will lead to improved patient outcomes, specifically pain and physical function. The CDS implementation will be guided by the Exploration, Preparation, Implementation, Sustainment (EPIS) framework. The evaluation will be organized by the Reach, Effectiveness, Adoption, Implementation, and Maintenance (RE-AIM) framework. We will adapt and tailor PainManager, an open source interoperable CDS tool, for implementation in primary care clinics affiliated with the OneFlorida Clinical Research Consortium. We will evaluate the effect of tailored implementation support on PainManager’s adoption for pain treatment shared decision making. This evaluation will establish the feasibility and obtain preliminary data in preparation for a multi-site pragmatic trial targeting the effectiveness of PainManager and tailored implementation support on shared decision making and patient-reported pain and physical function. Discussion This research will generate evidence on strategies for implementing interoperable CDS in new clinical settings across different types of electronic health records (EHRs). The study will also inform tailored implementation strategies to be further tested in a subsequent hybrid effectiveness-implementation trial. Together, these efforts will lead to important new technology and evidence that patients, clinicians, and health systems can use to improve care for millions of Americans who suffer from pain and other chronic conditions. Trial registration ClinicalTrials.gov, NCT05256394 , Registered 25 February 2022.

Abstract Integrating data across institutions can improve learning efficiency. To integrate data efficiently while protecting privacy, we propose A one-shot, summary-statistics-based, Distributed Algorithm for fitting Penalized (ADAP) regression models across multiple datasets. ADAP utilizes patient-level data from a lead site and incorporates the first-order (ADAP1) and second-order gradients (ADAP2) of the objective function from collaborating sites to construct a surrogate objective function at the lead site, where model fitting is then completed with proper regularizations applied. We evaluate the performance of the proposed method using both simulation and a real-world application to study risk factors for opioid use disorder (OUD) using 15,000 patient data from the OneFlorida Clinical Research Consortium. Our results show that ADAP performs nearly the same as the pooled estimator but achieves higher estimation accuracy and better variable selection than the local and average estimators. Moreover, ADAP2 successfully handles heterogeneity in covariate distributions.

Summary: The abundance of biomedical knowledge gained from biological experiments and clinical practices is an invaluable resource for biomedicine. The emerging biomedical knowledge graphs (BKGs) provide an efficient and effective way to manage the abundant knowledge in biomedical and life science. In this study, we created a comprehensive BKG called the integrative Biomedical Knowledge Hub (iBKH) by harmonizing and integrating information from diverse biomedical resources. To make iBKH easily accessible for biomedical research, we developed a web-based, user-friendly graphical portal that allows fast and interactive knowledge retrieval. Additionally, we also implemented an efficient and scalable graph learning pipeline for discovering novel biomedical knowledge in iBKH. As a proof of concept, we performed our iBKH-based method for computational in-silico drug repurposing for Alzheimer’s disease. The iBKH is publicly available.

Post-acute sequelae of SARS-CoV-2 infection (PASC) affects a wide range of organ systems among a large proportion of patients with SARS-CoV-2 infection. Although studies have identified a broad set of patient-level risk factors for PASC, little is known about the association between “exposome”—the totality of environmental exposures and the risk of PASC. Using electronic health data of patients with COVID-19 from two large clinical research networks in New York City and Florida, we identified environmental risk factors for 23 PASC symptoms and conditions from nearly 200 exposome factors. The three domains of exposome include natural environment, built environment, and social environment. We conducted a two-phase environment-wide association study. In Phase 1, we ran a mixed effects logistic regression with 5-digit ZIP Code tabulation area (ZCTA5) random intercepts for each PASC outcome and each exposome factor, adjusting for a comprehensive set of patient-level confounders. In Phase 2, we ran a mixed effects logistic regression for each PASC outcome including all significant (false positive discovery adjusted p-value < 0.05) exposome characteristics identified from Phase I and adjusting for confounders. We identified air toxicants (e.g., methyl methacrylate), particulate matter (PM2.5) compositions (e.g., ammonium), neighborhood deprivation, and built environment (e.g., food access) that were associated with increased risk of PASC conditions related to nervous, blood, circulatory, endocrine, and other organ systems. Specific environmental risk factors for each PASC condition and symptom were different across the New York City area and Florida. Future research is warranted to extend the analyses to other regions and examine more granular exposome characteristics to inform public health efforts to help patients recover from SARS-CoV-2 infection.