Search

Abstract Increased intracranial pressure (ICP) ≥15 mmHg is associated with adverse neurological outcomes, but needs invasive intracranial monitoring. Using the publicly available MIMIC-III Waveform Database (2000–2013) from Boston, we developed an artificial intelligence-derived biomarker for elevated ICP (aICP) for adult patients. aICP uses routinely collected extracranial waveform data as input, reducing the need for invasive monitoring. We externally validated aICP with an independent dataset from the Mount Sinai Hospital (2020–2022) in New York City. The AUROC, accuracy, sensitivity, and specificity on the external validation dataset were 0.80 (95% CI, 0.80–0.80), 73.8% (95% CI, 72.0–75.6%), 73.5% (95% CI 72.5–74.5%), and 73.0% (95% CI, 72.0–74.0%), respectively. We also present an exploratory analysis showing aICP predictions are associated with clinical phenotypes. A ten-percentile increment was associated with brain malignancy (OR = 1.68; 95% CI, 1.09-2.60), intracerebral hemorrhage (OR = 1.18; 95% CI, 1.07–1.32), and craniotomy (OR = 1.43; 95% CI, 1.12–1.84; P

Abstract This study was designed to assess how different prompt engineering techniques, specifically direct prompts, Chain of Thought (CoT), and a modified CoT approach, influence the ability of GPT-3.5 to answer clinical and calculation-based medical questions, particularly those styled like the USMLE Step 1 exams. To achieve this, we analyzed the responses of GPT-3.5 to two distinct sets of questions: a batch of 1000 questions generated by GPT-4, and another set comprising 95 real USMLE Step 1 questions. These questions spanned a range of medical calculations and clinical scenarios across various fields and difficulty levels. Our analysis revealed that there were no significant differences in the accuracy of GPT-3.5's responses when using direct prompts, CoT, or modified CoT methods. For instance, in the USMLE sample, the success rates were 61.7% for direct prompts, 62.8% for CoT, and 57.4% for modified CoT, with a p-value of 0.734. Similar trends were observed in the responses to GPT-4 generated questions, both clinical and calculation-based, with p-values above 0.05 indicating no significant difference between the prompt types. The conclusion drawn from this study is that the use of CoT prompt engineering does not significantly alter GPT-3.5's effectiveness in handling medical calculations or clinical scenario questions styled like those in USMLE exams. This finding is crucial as it suggests that performance of ChatGPT remains consistent regardless of whether a CoT technique is used instead of direct prompts. This consistency could be instrumental in simplifying the integration of AI tools like ChatGPT into medical education, enabling healthcare professionals to utilize these tools with ease, without the necessity for complex prompt engineering.

Abstract Adenosine-to-inosine (A-to-I) editing is a prevalent post-transcriptional RNA modification within the brain. Yet, most research has relied on postmortem samples, assuming it is an accurate representation of RNA biology in the living brain. We challenge this assumption by comparing A-to-I editing between postmortem and living prefrontal cortical tissues. Major differences were found, with over 70,000 A-to-I sites showing higher editing levels in postmortem tissues. Increased A-to-I editing in postmortem tissues is linked to higher ADAR and ADARB1 expression, is more pronounced in non-neuronal cells, and indicative of postmortem activation of inflammation and hypoxia. Higher A-to-I editing in living tissues marks sites that are evolutionarily preserved, synaptic, developmentally timed, and disrupted in neurological conditions. Common genetic variants were also found to differentially affect A-to-I editing levels in living versus postmortem tissues. Collectively, these discoveries offer more nuanced and accurate insights into the regulatory mechanisms of RNA editing in the human brain.

Abstract Malnutrition is a frequently underdiagnosed condition leading to increased morbidity, mortality, and healthcare costs. The Mount Sinai Health System (MSHS) deployed a machine learning model (MUST-Plus) to detect malnutrition upon hospital admission. However, in diverse patient groups, a poorly calibrated model may lead to misdiagnosis, exacerbating health care disparities. We explored the model’s calibration across different variables and methods to improve calibration. Data from adult patients admitted to five MSHS hospitals from January 1, 2021 - December 31, 2022, were analyzed. We compared MUST-Plus prediction to the registered dietitian’s formal assessment. Hierarchical calibration was assessed and compared between the recalibration sample (N = 49,562) of patients admitted between January 1, 2021 - December 31, 2022, and the hold-out sample (N = 17,278) of patients admitted between January 1, 2023 - September 30, 2023. Statistical differences in calibration metrics were tested using bootstrapping with replacement. Before recalibration, the overall model calibration intercept was −1.17 (95% CI: −1.20, −1.14), slope was 1.37 (95% CI: 1.34, 1.40), and Brier score was 0.26 (95% CI: 0.25, 0.26). Both weak and moderate measures of calibration were significantly different between White and Black patients and between male and female patients. Logistic recalibration significantly improved calibration of the model across race and gender in the hold-out sample. The original MUST-Plus model showed significant differences in calibration between White vs. Black patients. It also overestimated malnutrition in females compared to males. Logistic recalibration effectively reduced miscalibration across all patient subgroups. Continual monitoring and timely recalibration can improve model accuracy.

BackgroundWith their unmatched ability to interpret and engage with human language and context, large language models (LLMs) hint at the potential to bridge AI and human cognitive processes. This review explores the current application of LLMs, such as ChatGPT, in the field of psychiatry.MethodsWe followed PRISMA guidelines and searched through PubMed, Embase, Web of Science, and Scopus, up until March 2024.ResultsFrom 771 retrieved articles, we included 16 that directly examine LLMs’ use in psychiatry. LLMs, particularly ChatGPT and GPT-4, showed diverse applications in clinical reasoning, social media, and education within psychiatry. They can assist in diagnosing mental health issues, managing depression, evaluating suicide risk, and supporting education in the field. However, our review also points out their limitations, such as difficulties with complex cases and potential underestimation of suicide risks.ConclusionEarly research in psychiatry reveals LLMs’ versatile applications, from diagnostic support to educational roles. Given the rapid pace of advancement, future investigations are poised to explore the extent to which these models might redefine traditional roles in mental health care.

Abstract Introduction End-stage kidney disease (ESKD) is associated with increased morbidity and mortality. Identifying patients with stage 4 CKD (CKD4) at risk of rapid progression to ESKD remains challenging. Accurate prediction of CKD4 progression can improve patient outcomes by improving advanced care planning and optimizing healthcare resource allocation. Methods We obtained electronic health record data from patients with CKD4 in a large health system between January 1, 2006, and December 31, 2016. We developed and validated four models, including Least Absolute Shrinkage and Selection Operator (LASSO) regression, random forest, eXtreme Gradient Boosting (XGBoost), and artificial neural network (ANN), to predict ESKD at 3 years. We utilized area under the receiver operating characteristic curve (AUROC) to evaluate model performances and utilized Shapley additive explanation (SHAP) values and plots to define feature dependence of the best performance model. Results We included 3,160 patients with CKD4. ESKD was observed in 538 patients (21%). All approaches had similar AUROCs; ANN yielded the highest AUROC (0.77; 95%CI 0.75 to 0.79) and LASSO regression (0.77; 95%CI 0.75 to 0.79), followed by random forest (0.76; 95% CI 0.74 to 0.79), and XGBoost (0.76; 95% CI 0.74 to 0.78). Conclusions We developed and validated several models for near-term prediction of kidney failure in CKD4. ANN, random forest, and XGBoost demonstrated similar predictive performances. Using this suite of models, interventions can be customized based on risk, and population health and resources appropriately allocated.

BackgroundArea-level social determinants of health (SDOH) based on patients' ZIP codes or census tracts have been commonly used in research instead of individual SDOHs. To our knowledge, whether machine learning (ML) could be used to derive individual SDOH measures, specifically individual educational attainment, is unknown.MethodsThis is a retrospective study using data from the Mount Sinai BioMe Biobank. We included participants that completed a validated questionnaire on educational attainment and had home addresses in New York City. ZIP code-level education was derived from the American Community Survey matched for the participant's gender and race/ethnicity. We tested several algorithms to predict individual educational attainment from routinely collected clinical and demographic data. To evaluate how using different measures of educational attainment will impact model performance, we developed three distinct models for predicting cardiovascular (CVD) hospitalization. Educational attainment was imputed into models as either survey-derived, ZIP code-derived, or ML-predicted educational attainment.ResultsA total of 20,805 participants met inclusion criteria. Concordance between survey and ZIP code-derived education was 47%, while the concordance between survey and ML model-predicted education was 67%. A total of 13,715 patients from the cohort were included into our CVD hospitalization prediction models, of which 1,538 (11.2%) had a history of CVD hospitalization. The AUROC of the model predicting CVD hospitalization using survey-derived education was significantly higher than the model using ZIP code-level education (0.77 versus 0.72; p < 0.001) and the model using ML model-predicted education (0.77 versus 0.75; p < 0.001). The AUROC for the model using ML model-predicted education was also significantly higher than that using ZIP code-level education (p = 0.003).ConclusionThe concordance of survey and ZIP code-level educational attainment in NYC was low. As expected, the model utilizing survey-derived education achieved the highest performance. The model incorporating our ML model-predicted education outperformed the model relying on ZIP code-derived education. Implementing ML techniques can improve the accuracy of SDOH data and consequently increase the predictive performance of outcome models.

Introduction/Objective: The KidneyIntelX is a multiplex, bioprognostic, immunoassay consisting of 3 plasma biomarkers and clinical variables that uses machine learning to predict a patient’s risk for a progressive decline in kidney function over 5 years. We report the 1-year pre- and post-test clinical impact on care management, eGFR slope, and A1C along with engagement of population health clinical pharmacists and patient coordinators to promote a program of sustainable kidney, metabolic, and cardiac health. Methods: The KidneyIntelX in vitro prognostic test was previously validated for patients with type 2 diabetes and diabetic kidney disease (DKD) to predict kidney function decline within 5 years was introduced into the RWE study (NCT04802395) across the Health System as part of a population health chronic disease management program from [November 2020 to April 2023]. Pre- and post-test patients with a minimum of 12 months of follow-up post KidneyIntelX were assessed across all aspects of the program. Results: A total of 5348 patients with DKD had a KidneyIntelX assay. The median age was 68 years old, 52% were female, 27% self-identified as Black, and 89% had hypertension. The median baseline eGFR was 62 ml/min/1.73 m 2 , urine albumin-creatinine ratio was 54 mg/g, and A1C was 7.3%. The KidneyIntelX risk level was low in 49%, intermediate in 40%, and high in 11% of cases. New prescriptions for SGLT2i, GLP-1 RA, or referral to a specialist were noted in 19%, 33%, and 43% among low-, intermediate-, and high-risk patients, respectively. The median A1C decreased from 8.2% pre-test to 7.5% post-test in the high-risk group ( P

Background Right ventricular ejection fraction (RVEF) and end‐diastolic volume (RVEDV) are not readily assessed through traditional modalities. Deep learning–enabled ECG analysis for estimation of right ventricular (RV) size or function is unexplored. Methods and Results We trained a deep learning–ECG model to predict RV dilation (RVEDV >120 mL/m2), RV dysfunction (RVEF ≤40%), and numerical RVEDV and RVEF from a 12‐lead ECG paired with reference‐standard cardiac magnetic resonance imaging volumetric measurements in UK Biobank (UKBB; n=42 938). We fine‐tuned in a multicenter health system (MSHoriginal [Mount Sinai Hospital]; n=3019) with prospective validation over 4 months (MSHvalidation; n=115). We evaluated performance with area under the receiver operating characteristic curve for categorical and mean absolute error for continuous measures overall and in key subgroups. We assessed the association of RVEF prediction with transplant‐free survival with Cox proportional hazards models. The prevalence of RV dysfunction for UKBB/MSHoriginal/MSHvalidation cohorts was 1.0%/18.0%/15.7%, respectively. RV dysfunction model area under the receiver operating characteristic curve for UKBB/MSHoriginal/MSHvalidation cohorts was 0.86/0.81/0.77, respectively. The prevalence of RV dilation for UKBB/MSHoriginal/MSHvalidation cohorts was 1.6%/10.6%/4.3%. RV dilation model area under the receiver operating characteristic curve for UKBB/MSHoriginal/MSHvalidation cohorts was 0.91/0.81/0.92, respectively. MSHoriginal mean absolute error was RVEF=7.8% and RVEDV=17.6 mL/m2. The performance of the RVEF model was similar in key subgroups including with and without left ventricular dysfunction. Over a median follow‐up of 2.3 years, predicted RVEF was associated with adjusted transplant‐free survival (hazard ratio, 1.40 for each 10% decrease; P=0.031). Conclusions Deep learning–ECG analysis can identify significant cardiac magnetic resonance imaging RV dysfunction and dilation with good performance. Predicted RVEF is associated with clinical outcome.

Abstract Background Decentralized, digital health studies can provide real-world evidence of the lasting effects of COVID-19 on physical, socioeconomic, psychological, and social determinant factors of health in India. Existing research cohorts, however, are small and were not designed for longitudinal collection of comprehensive data from India’s diverse population. Data4Life is a nationwide, digitally enabled, health research initiative to examine the post-acute sequelae of COVID-19 across individuals, communities, and regions. Data4Life seeks to build an ethnically and geographically diverse population of at least 100,000 participants in India. Methods Here we discuss the feasibility of developing a completely decentralized COVID-19 cohort in India through qualitative analysis of data collection procedures, participant characteristics, participant perspectives on recruitment and reported study motivation. Results As of June 13th, 2022, more than 6,000 participants from 17 Indian states completed baseline surveys. Friend and family referral were identified as the most common recruitment method (64.8%) across all demographic groups. Helping family and friends was the primary reason reported for joining the study (61.5%). Conclusions Preliminary findings support the use of digital technology for rapid enrollment and data collection to develop large health research cohorts in India. This demonstrates the potential for expansion of digitally enabled health research in India. These findings also outline the value of person-to-person recruitment strategies when conducting digital health research in modern-day India. Qualitative analysis reveals opportunities to increase diversity and retention in real time. It also informs strategies for improving participant experiences in the current Data4Life initiative and future studies.

Abstract Background Acute kidney injury (AKI) is a known complication of COVID-19 and is associated with an increased risk of in-hospital mortality. Unbiased proteomics using biological specimens can lead to improved risk stratification and discover pathophysiological mechanisms. Methods Using measurements of ~4000 plasma proteins in two cohorts of patients hospitalized with COVID-19, we discovered and validated markers of COVID-associated AKI (stage 2 or 3) and long-term kidney dysfunction. In the discovery cohort (N = 437), we identified 413 higher plasma abundances of protein targets and 30 lower plasma abundances of protein targets associated with COVID-AKI (adjusted p

Abstract The electrocardiogram (ECG) is a ubiquitous diagnostic modality. Convolutional neural networks (CNNs) applied towards ECG analysis require large sample sizes, and transfer learning approaches for biomedical problems may result in suboptimal performance when pre-training is done on natural images. We leveraged masked image modeling to create a vision-based transformer model, HeartBEiT, for electrocardiogram waveform analysis. We pre-trained this model on 8.5 million ECGs and then compared performance vs. standard CNN architectures for diagnosis of hypertrophic cardiomyopathy, low left ventricular ejection fraction and ST elevation myocardial infarction using differing training sample sizes and independent validation datasets. We find that HeartBEiT has significantly higher performance at lower sample sizes compared to other models. We also find that HeartBEiT improves explainability of diagnosis by highlighting biologically relevant regions of the EKG vs. standard CNNs. Domain specific pre-trained transformer models may exceed the classification performance of models trained on natural images especially in very low data regimes. The combination of the architecture and such pre-training allows for more accurate, granular explainability of model predictions.

Abstract Systemic autoimmune rheumatic diseases (SARDs) can lead to irreversible damage if left untreated, yet these patients often endure long diagnostic journeys before being diagnosed and treated. Machine learning may help overcome the challenges of diagnosing SARDs and inform clinical decision-making. Here, we developed and tested a machine learning model to identify patients who should receive rheumatological evaluation for SARDs using longitudinal electronic health records of 161,584 individuals from two institutions. The model demonstrated high performance for predicting cases of autoantibody-tested individuals in a validation set, an external test set, and an independent cohort with a broader case definition. This approach identified more individuals for autoantibody testing compared with current clinical standards and a greater proportion of autoantibody carriers among those tested. Diagnoses of SARDs and other autoimmune conditions increased with higher model probabilities. The model detected a need for autoantibody testing and rheumatology encounters up to five years before the test date and assessment date, respectively. Altogether, these findings illustrate that the clinical manifestations of a diverse array of autoimmune conditions are detectable in electronic health records using machine learning, which may help systematize and accelerate autoimmune testing.

Background: The integration of artificial intelligence (AI) into healthcare has opened new avenues for enhancing patient care and clinical research. In gastroenterology, the potential of AI tools, specifically large language models like ChatGPT, is being explored to understand their utility and effectiveness. Objectives: The primary goal of this systematic review is to assess the various applications, ascertain the benefits, and identify the limitations of utilizing ChatGPT within the realm of gastroenterology. Design: Through a systematic approach, this review aggregates findings from multiple studies to evaluate the impact of ChatGPT on the field. Data sources and methods: The review was based on a detailed literature search of the PubMed database, targeting research that delves into the use of ChatGPT for gastroenterological purposes. It incorporated six selected studies, which were meticulously evaluated for quality using the Joanna Briggs Institute critical appraisal instruments. The data were then synthesized narratively to encapsulate the roles, advantages, and drawbacks of ChatGPT in gastroenterology. Results: The investigation unearthed various roles of ChatGPT, including its use in patient education, diagnostic self-assessment, disease management, and the formulation of research queries. Notable benefits were its capability to provide pertinent recommendations, enhance communication between patients and physicians, and prompt valuable research inquiries. Nonetheless, it encountered obstacles in decoding intricate medical questions, yielded inconsistent responses at times, and exhibited limitations in generating novel content. The review also considered ethical implications. Conclusion: ChatGPT has demonstrated significant potential in the field of gastroenterology, especially in facilitating patient–physician interactions and managing diseases. Despite these advancements, the review underscores the necessity for ongoing refinement, customization, and ethical regulation of AI tools. These findings serve to enrich the dialog concerning AI’s role in healthcare, with a specific focus on ChatGPT’s application in gastroenterology.

Abstract BackgroundMachine learning clustering offers an unbiased approach to better understand the interactions of complex social and clinical variables via integrative subphenotypes, an approach not studied in out-of-hospital cardiac arrest (OHCA). ObjectiveWe conducted a cluster analysis for a cohort of OHCA survivors to examine the association of clinical and social factors for mortality at 1 year. MethodsWe used a retrospective observational OHCA cohort identified from Medicare claims data, including area-level social determinants of health (SDOH) features and hospital-level data sets. We applied k-means clustering algorithms to identify subphenotypes of beneficiaries who had survived an OHCA and examined associations of outcomes by subphenotype. ResultsWe identified 27,028 unique beneficiaries who survived to discharge after OHCA. We derived 4 distinct subphenotypes. Subphenotype 1 included a distribution of more urban, female, and Black beneficiaries with the least robust area-level SDOH measures and the highest 1-year mortality (2375/4417, 53.8%). Subphenotype 2 was characterized by a greater distribution of male, White beneficiaries and had the strongest zip code–level SDOH measures, with 1-year mortality at 49.9% (4577/9165). Subphenotype 3 had the highest rates of cardiac catheterization at 34.7% (1342/3866) and the greatest distribution with a driving distance to the index OHCA hospital from their primary residence >16.1 km at 85.4% (8179/9580); more were also discharged to a skilled nursing facility after index hospitalization. Subphenotype 4 had moderate median household income at US $51,659.50 (IQR US $41,295 to $67,081) and moderate to high median unemployment at 5.5% (IQR 4.2%-7.1%), with the lowest 1-year mortality (1207/3866, 31.2%). Joint modeling of these features demonstrated an increased hazard of death for subphenotypes 1 to 3 but not for subphenotype 4 when compared to reference. ConclusionsWe identified 4 distinct subphenotypes with differences in outcomes by clinical and area-level SDOH features for OHCA. Further work is needed to determine if individual or other SDOH domains are specifically tied to long-term survival after OHCA.

BackgroundThe increasing use of smartphones, wearables, and connected devices has enabled the increasing application of digital technologies for research. Remote digital study platforms comprise a patient-interfacing digital application that enables multimodal data collection from a mobile app and connected sources. They offer an opportunity to recruit at scale, acquire data longitudinally at a high frequency, and engage study participants at any time of the day in any place. Few published descriptions of centralized digital research platforms provide a framework for their development. ObjectiveThis study aims to serve as a road map for those seeking to develop a centralized digital research platform. We describe the technical and functional aspects of the ehive app, the centralized digital research platform of the Hasso Plattner Institute for Digital Health at Mount Sinai Hospital, New York, New York. We then provide information about ongoing studies hosted on ehive, including usership statistics and data infrastructure. Finally, we discuss our experience with ehive in the broader context of the current landscape of digital health research platforms. MethodsThe ehive app is a multifaceted and patient-facing central digital research platform that permits the collection of e-consent for digital health studies. An overview of its development, its e-consent process, and the tools it uses for participant recruitment and retention are provided. Data integration with the platform and the infrastructure supporting its operations are discussed; furthermore, a description of its participant- and researcher-facing dashboard interfaces and the e-consent architecture is provided. ResultsThe ehive platform was launched in 2020 and has successfully hosted 8 studies, namely 6 observational studies and 2 clinical trials. Approximately 1484 participants downloaded the app across 36 states in the United States. The use of recruitment methods such as bulk messaging through the EPIC electronic health records and standard email portals enables broad recruitment. Light-touch engagement methods, used in an automated fashion through the platform, maintain high degrees of engagement and retention. The ehive platform demonstrates the successful deployment of a central digital research platform that can be modified across study designs. ConclusionsCentralized digital research platforms such as ehive provide a novel tool that allows investigators to expand their research beyond their institution, engage in large-scale longitudinal studies, and combine multimodal data streams. The ehive platform serves as a model for groups seeking to develop similar digital health research programs. International Registered Report Identifier (IRRID)DERR1-10.2196/49204

Vaid et al performed a multi-center retrospective cohort study using electrocardiograms from patients with mitral regurgitation to train a deep learning model to detect valvular disease and validated it in externally. They demonstrate the model could potentially enable earlier disease detection and improve overall prognosis.

Abstract The electronic Medical Records and Genomics (eMERGE) Network assessed the feasibility of deploying portable phenotype rule-based algorithms with natural language processing (NLP) components added to improve performance of existing algorithms using electronic health records (EHRs). Based on scientific merit and predicted difficulty, eMERGE selected six existing phenotypes to enhance with NLP. We assessed performance, portability, and ease of use. We summarized lessons learned by: (1) challenges; (2) best practices to address challenges based on existing evidence and/or eMERGE experience; and (3) opportunities for future research. Adding NLP resulted in improved, or the same, precision and/or recall for all but one algorithm. Portability, phenotyping workflow/process, and technology were major themes. With NLP, development and validation took longer. Besides portability of NLP technology and algorithm replicability, factors to ensure success include privacy protection, technical infrastructure setup, intellectual property agreement, and efficient communication. Workflow improvements can improve communication and reduce implementation time. NLP performance varied mainly due to clinical document heterogeneity; therefore, we suggest using semi-structured notes, comprehensive documentation, and customization options. NLP portability is possible with improved phenotype algorithm performance, but careful planning and architecture of the algorithms is essential to support local customizations.

Abstract Sample size estimation is a crucial step in experimental design but is understudied in the context of deep learning. Currently, estimating the quantity of labeled data needed to train a classifier to a desired performance, is largely based on prior experience with similar models and problems or on untested heuristics. In many supervised machine learning applications, data labeling can be expensive and time-consuming and would benefit from a more rigorous means of estimating labeling requirements. Here, we study the problem of estimating the minimum sample size of labeled training data necessary for training computer vision models as an exemplar for other deep learning problems. We consider the problem of identifying the minimal number of labeled data points to achieve a generalizable representation of the data, a minimum converging sample (MCS). We use autoencoder loss to estimate the MCS for fully connected neural network classifiers. At sample sizes smaller than the MCS estimate, fully connected networks fail to distinguish classes, and at sample sizes above the MCS estimate, generalizability strongly correlates with the loss function of the autoencoder. We provide an easily accessible, code-free, and dataset-agnostic tool to estimate sample sizes for fully connected networks. Taken together, our findings suggest that MCS and convergence estimation are promising methods to guide sample size estimates for data collection and labeling prior to training deep learning models in computer vision.

Heart failure is a major cause of cardiovascular morbidity and mortality. Here, the authors report results of a genome-wide association study meta-analysis, characterizing the role of common genetic variants in heart failure, finding overlap with common cardiovascular risk factors and imaging measures of cardiac structure/function.

Background: Electrocardiogram (ECG) deep learning (DL) has promise to improve the outcomes of patients with cardiovascular abnormalities. In ECG DL, researchers often use convolutional neural networks (CNNs) and traditionally use the full duration of raw ECG waveforms that create redundancies in feature learning and result in inaccurate predictions with large uncertainties. Objective: For enhancing these predictions, we introduced a sub-waveform representation that leverages the rhythmic pattern of ECG waveforms (data-centric approach) rather than changing the CNN architecture (model-centric approach). Results: We applied the proposed representation to a population with 92,446 patients to identify left ventricular dysfunction. We found that the sub-waveform representation increases the performance metrics compared to the full-waveform representation. We observed a 2% increase for area under the receiver operating characteristic curve and 10% increase for area under the precision-recall curve. We also carefully examined three reliability components of explainability, interpretability, and fairness. We provided an explanation for enhancements obtained by heartbeat alignment mechanism. By developing a new scoring system, we interpreted the clinical relevance of ECG features and showed that sub-waveform representation further pushes the scores towards clinical predictions. Finally, we showed that the new representation significantly reduces prediction uncertainties within subgroups that contributes to individual fairness. Conclusion: We expect that this added control over the granularity of ECG data will improve the DL modeling for new artificial intelligence technologies in the cardiovascular space.

Background: Gastric cancer (GC), a significant health burden worldwide, is typically diagnosed in the advanced stages due to its non-specific symptoms and complex morphological features. Deep learning (DL) has shown potential for improving and standardizing early GC detection. This systematic review aims to evaluate the current status of DL in pre-malignant, early-stage, and gastric neoplasia analysis. Methods: A comprehensive literature search was conducted in PubMed/MEDLINE for original studies implementing DL algorithms for gastric neoplasia detection using endoscopic images. We adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The focus was on studies providing quantitative diagnostic performance measures and those comparing AI performance with human endoscopists. Results: Our review encompasses 42 studies that utilize a variety of DL techniques. The findings demonstrate the utility of DL in GC classification, detection, tumor invasion depth assessment, cancer margin delineation, lesion segmentation, and detection of early-stage and pre-malignant lesions. Notably, DL models frequently matched or outperformed human endoscopists in diagnostic accuracy. However, heterogeneity in DL algorithms, imaging techniques, and study designs precluded a definitive conclusion about the best algorithmic approach. Conclusions: The promise of artificial intelligence in improving and standardizing gastric neoplasia detection, diagnosis, and segmentation is significant. This review is limited by predominantly single-center studies and undisclosed datasets used in AI training, impacting generalizability and demographic representation. Further, retrospective algorithm training may not reflect actual clinical performance, and a lack of model details hinders replication efforts. More research is needed to substantiate these findings, including larger-scale multi-center studies, prospective clinical trials, and comprehensive technical reporting of DL algorithms and datasets, particularly regarding the heterogeneity in DL algorithms and study designs.

A large-scale GWAS provides insight on diabetes-dependent genetic effects on the glomerular filtration rate, a common metric to monitor kidney health in disease.

Rationale & Objective: The association between cannabis use and chronic kidney disease (CKD) is controversial. We aimed to assess association of CKD with cannabis use in a large cohort study and then assess causality using Mendelian randomization with a genome-wide association study (GWAS). Study Design: Retrospective cohort study and genome-wide association study. Setting & Participants: The retrospective study was conducted on the All of Us cohort (N=223,354). Genetic instruments for cannabis use disorder were identified from 3 GWAS: the Psychiatric Genomics Consortium Substance Use Disorders, iPSYCH, and deCODE (N=384,032). Association between genetic instruments and CKD was investigated in the CKDGen GWAS (N > 1.2 million). Exposure: Cannabis consumption. Outcomes: CKD outcomes included: cystatin-C and creatinine-based kidney function, proteinuria, and blood urea nitrogen. Analytical Approach: We conducted association analyses to test for frequency of cannabis use and CKD. To evaluate causality, we performed a 2-sample Mendelian randomization. Results: In the retrospective study, compared to former users, less than monthly (OR, 1.01; 95% CI, 0.87-1.18; P = 0.87) and monthly cannabis users (OR, 1.15; 95% CI, 0.86-1.52; P = 0.33) did not have higher CKD odds. Conversely, weekly (OR, 1.28; 95% CI, 1.01-1.60; P = 0.04) and daily use (OR, 1.25; 95% CI, 1.04-1.50; P = 0.02) was significantly associated with CKD, adjusted for multiple confounders. In Mendelian randomization, genetic liability to cannabis use disorder was not associated with increased odds for CKD (OR, 1.00; 95% CI, 0.99-1.01; P = 0.96). These results were robust across different Mendelian randomization techniques and multiple kidney traits. Limitations: Likely underreporting of cannabis use. In Mendelian randomization, genetic instruments were identified in the GWAS that included individuals primarily of European ancestry. Conclusions: Despite the epidemiological association between cannabis use and CKD, there was no evidence of a causal effect, indicating confounding in observational studies.

ImportanceSARS-CoV-2 infection can result in ongoing, relapsing, or new symptoms or other health effects after the acute phase of infection; termed post-acute sequelae of SARS-CoV-2 infection (PASC), or long COVID. The characteristics, prevalence, trajectory and mechanisms of PASC are ill-defined. The objectives of the Researching COVID to Enhance Recovery (RECOVER) Multi-site Observational Study of PASC in Adults (RECOVER-Adult) are to: (1) characterize PASC prevalence; (2) characterize the symptoms, organ dysfunction, natural history, and distinct phenotypes of PASC; (3) identify demographic, social and clinical risk factors for PASC onset and recovery; and (4) define the biological mechanisms underlying PASC pathogenesis.MethodsRECOVER-Adult is a combined prospective/retrospective cohort currently planned to enroll 14,880 adults aged ≥18 years. Eligible participants either must meet WHO criteria for suspected, probable, or confirmed infection; or must have evidence of no prior infection. Recruitment occurs at 86 sites in 33 U.S. states, Washington, DC and Puerto Rico, via facility- and community-based outreach. Participants complete quarterly questionnaires about symptoms, social determinants, vaccination status, and interim SARS-CoV-2 infections. In addition, participants contribute biospecimens and undergo physical and laboratory examinations at approximately 0, 90 and 180 days from infection or negative test date, and yearly thereafter. Some participants undergo additional testing based on specific criteria or random sampling. Patient representatives provide input on all study processes. The primary study outcome is onset of PASC, measured by signs and symptoms. A paradigm for identifying PASC cases will be defined and updated using supervised and unsupervised learning approaches with cross-validation. Logistic regression and proportional hazards regression will be conducted to investigate associations between risk factors, onset, and resolution of PASC symptoms.DiscussionRECOVER-Adult is the first national, prospective, longitudinal cohort of PASC among US adults. Results of this study are intended to inform public health, spur clinical trials, and expand treatment options.RegistrationNCT05172024.

Introduction and Objective: The lack of precision to identify patients with early-stage diabetic kidney disease (DKD) at near-term risk for progressive decline in kidney function results in poor disease management often leading to kidney failure requiring unplanned dialysis. The KidneyIntelX is a multiplex, bioprognostic, immunoassay consisting of 3 plasma biomarkers and clinical variables that uses machine learning to generate a risk score for progressive decline in kidney function over 5-year in adults with early-stage DKD. Our objective was to assess the impact of KidneyIntelX on management and outcomes in a Health System in the real-world evidence (RWE) study. Methods: KidneyIntelX was introduced into a large metropolitan Health System via a population health-defined approved care pathway for patients with stages 1 to 3 DKD between [November 2020 to March 2022]. Decision impact on visit frequency, medication management, specialist referral, and selected lab values was assessed. We performed an interim analysis in patients through 6-months post-test date to evaluate the impact of risk level with clinical decision-making and outcomes. Results: A total of 1686 patients were enrolled in the RWE study and underwent KidneyIntelX testing and subsequent care pathway management. The median age was 68 years, 52% were female, 26% self-identified as Black, and 94% had hypertension. The median baseline eGFR was 59 ml/minute/1.73 m 2 , urine albumin-creatinine ratio was 69 mg/g, and HbA1c was 7.7%. After testing, a clinical encounter in the first month occurred in 13%, 43%, and 53% of low-risk, intermediate-risk, and high-risk patients, respectively and 46%, 61%, and 71% had at least 1 action taken within the first 6 months. High-risk patients were more likely to be placed on SGLT2 inhibitors (OR = 4.56; 95% CI 3.00-6.91 vs low-risk), and more likely to be referred to a specialist such as a nephrologist, endocrinologist, or dietician (OR = 2.49; 95% CI 1.53-4.01) compared to low-risk patients. Conclusions: The combination of KidneyIntelX, clinical guidelines and educational support resulted in changes in clinical management by clinicians. After testing, there was an increase in visit frequency, referrals for disease management, and introduction to guideline-recommended medications. These differed by risk category, indicating an impact of KidneyIntelX risk stratification on clinical care.

OBJECTIVES:. Older age is a key risk factor for adverse outcomes in critically ill patients with COVID-19. However, few studies have investigated whether preexisting comorbidities and acute physiologic ICU factors modify the association between age and death. DESIGN:. Multicenter cohort study. SETTING:. ICUs at 68 hospitals across the United States. PATIENTS:. A total of 5,037 critically ill adults with COVID-19 admitted to ICUs between March 1, 2020, and July 1, 2020. INTERVENTIONS:. None. MEASUREMENTS AND MAIN RESULTS:. The primary exposure was age, modeled as a continuous variable. The primary outcome was 28-day inhospital mortality. Multivariable logistic regression tested the association between age and death. Effect modification by the number of risk factors was assessed through a multiplicative interaction term in the logistic regression model. Among the 5,037 patients included (mean age, 60.9 yr [± 14.7], 3,179 [63.1%] male), 1,786 (35.4%) died within 28 days. Age had a nonlinear association with 28-day mortality (p for nonlinearity

BackgroundSome but not all African-Americans (AA) who carry APOL1 nephropathy risk variants (APOL1) develop kidney failure (end-stage kidney disease, ESKD). To identify genetic modifiers, we assessed gene–gene interactions in a large prospective cohort of the REasons for Geographic and Racial Differences in Stroke (REGARDS) study.MethodsGenotypes from 8,074 AA participants were obtained from Illumina Infinium Multi-Ethnic AMR/AFR Extended BeadChip. We compared 388 incident ESKD cases with 7,686 non-ESKD controls, using a two-locus interaction approach. Logistic regression was used to examine the effect of APOL1 risk status (using recessive and additive models), single nucleotide polymorphism (SNP), and APOL1*SNP interaction on incident ESKD, adjusting for age, sex, and ancestry. APOL1*SNP interactions that met the threshold of 1.0 × 10−5 were replicated in the Genetics of Hypertension Associated Treatment (GenHAT) study (626 ESKD cases and 6,165 controls). In a sensitivity analysis, models were additionally adjusted for diabetes status. We conducted additional replication in the BioVU study.ResultsTwo APOL1 risk alleles prevalence (recessive model) was similar in the REGARDS and GenHAT studies. Only one APOL1–SNP interaction, for rs7067944 on chromosome 10, ~10 KB from the PCAT5 gene met the genome-wide statistical threshold (Pinteraction = 3.4 × 10−8), but this interaction was not replicated in the GenHAT study. Among other relevant top findings (with Pinteraction < 1.0 × 10−5), a variant (rs2181251) near SMOC2 on chromosome six interacted with APOL1 risk status (additive) on ESKD outcomes (REGARDS study, Pinteraction =5.3 × 10−6) but the association was not replicated (GenHAT study, Pinteraction = 0.07, BioVU study, Pinteraction = 0.53). The association with the locus near SMOC2 persisted further in stratified analyses. Among those who inherited ≥1 alternate allele of rs2181251, APOL1 was associated with an increased risk of incident ESKD (OR [95%CI] = 2.27[1.53, 3.37]) but APOL1 was not associated with ESKD in the absence of the alternate allele (OR [95%CI] = 1.34[0.96, 1.85]) in the REGARDS study. The associations were consistent after adjusting for diabetes.ConclusionIn a large genome-wide association study of AAs, a locus SMOC2 exhibited a significant interaction with the APOL1 locus. SMOC2 contributes to the progression of fibrosis after kidney injury and the interaction with APOL1 variants may contribute to an explanation for why only some APOLI high-risk individuals develop ESKD.

N-of-1 trials are the gold standard study design to evaluate individual treatment effects and derive personalized treatment strategies. Digital tools have the potential to initiate a new era of N-of-1 trials in terms of scale and scope, but fully functional platforms are not yet available. Here, we present the open source StudyU platform, which includes the StudyU Designer and StudyU app. With the StudyU Designer, scientists are given a collaborative web application to digitally specify, publish, and conduct N-of-1 trials. The StudyU app is a smartphone app with innovative user-centric elements for participants to partake in trials published through the StudyU Designer to assess the effects of different interventions on their health. Thereby, the StudyU platform allows clinicians and researchers worldwide to easily design and conduct digital N-of-1 trials in a safe manner. We envision that StudyU can change the landscape of personalized treatments both for patients and healthy individuals, democratize and personalize evidence generation for self-optimization and medicine, and can be integrated in clinical practice.

BackgroundFluid management during continuous renal replacement therapy (CRRT) requires accuracy in the prescription of desired patient fluid balance (FBGoal) and precision in the attainable patient fluid balance (FBAchieved). Herein, we examined the association of the gap between prescribed vs. achieved patient fluid balance during CRRT (%FBGap) with hospital mortality in critically ill patients.MethodsCohort study of critically ill adults with acute kidney injury (AKI) requiring CRRT and a prescription of negative fluid balance (mean patient fluid balance goal of negative ≥0.5 liters per day). Fluid management parameters included: 1) NUF (net ultrafiltration rate); 2) FBGoal; 3) FBAchieved; and 4) FBGap (% gap of fluid balance achieved vs. goal), all adjusted by patient's weight (kg) and duration of CRRT (hours).ResultsData from 653 patients (median of 102.2 patient-hours of CRRT) were analyzed. Mean (SD) age was 56.7 (14.6) years and 61.9% were male. Hospital mortality rate was 64%. Despite FBGoal was similar in patients who died vs. survived, survivors achieved greater negative fluid balance during CRRT than non-survivors: median FBAchieved -0.25 [-0.52 to -0.05] vs. 0.06 [-0.26 to 0.62] ml/kg/h, pConclusionsHigher %FBGap was independently associated with an increased risk of hospital mortality in critically ill adults with AKI on CRRT in whom clinicians prescribed negative fluid balance via CRRT. %FBGap represents a novel quality indicator of CRRT delivery that could assist with operationalizing fluid management interventions during CRRT.

On the electrocardiogram, the PR interval reflects conduction from the atria to ventricles and also serves as risk indicator of cardiovascular morbidity and mortality. Here, the authors perform genome-wide meta-analyses for PR interval in multiple ancestries and identify 141 previously unreported genetic loci.

Background Despite advances in cardiovascular disease and risk factor management, mortality from ischemic heart failure (HF) in patients with coronary artery disease (CAD) remains high. Given the partial role of genetics in HF and lack of reliable risk stratification tools, we developed and validated a polygenic risk score for HF in patients with CAD, which we term HF‐PRS. Methods and Results Using summary statistics from a recent genome‐wide association study for HF, we developed candidate PRSs in the Mount Sinai BioMe CAD patient cohort (N=6274) by using the pruning and thresholding method and LDPred. We validated the best score in the Penn Medicine BioBank (N=7250) and performed a subgroup analysis in a high‐risk cohort who had undergone coronary catheterization. We observed a significant association between HF‐PRS score and ischemic HF even after adjusting for evidence of obstructive CAD in patients of European ancestry in both BioMe (odds ratio [OR], 1.14 per SD; 95% CI, 1.05–1.24; P=0.003) and Penn Medicine BioBank (OR, 1.07 per SD; 95% CI, 1.01–1.13; P=0.016). In European patients with CAD in Penn Medicine BioBank who had undergone coronary catheterization, individuals in the top 10th percentile of PRS had a 2‐fold increased odds of ischemic HF (OR, 2.0; 95% CI, 1.1–3.7; P=0.02) compared with the bottom 10th percentile. Conclusions A PRS for HF enables risk stratification in patients with CAD. Future prospective studies aimed at demonstrating clinical utility are warranted for adoption in the patient setting.

BackgroundThe COVID-19 pandemic has resulted in a high degree of psychological distress among health care workers (HCWs). There is a need to characterize which HCWs are at an increased risk of developing psychological effects from the pandemic. Given the differences in the response of individuals to stress, an analysis of both the perceived and physiological consequences of stressors can provide a comprehensive evaluation of its impact. ObjectiveThis study aimed to determine characteristics associated with longitudinal perceived stress in HCWs and to assess whether changes in heart rate variability (HRV), a marker of autonomic nervous system function, are associated with features protective against longitudinal stress. MethodsHCWs across 7 hospitals in New York City, NY, were prospectively followed in an ongoing observational digital study using the custom Warrior Watch Study app. Participants wore an Apple Watch for the duration of the study to measure HRV throughout the follow-up period. Surveys measuring perceived stress, resilience, emotional support, quality of life, and optimism were collected at baseline and longitudinally. ResultsA total of 361 participants (mean age 36.8, SD 10.1 years; female: n=246, 69.3%) were enrolled. Multivariate analysis found New York City’s COVID-19 case count to be associated with increased longitudinal stress (P=.008). Baseline emotional support, quality of life, and resilience were associated with decreased longitudinal stress (P

Summary: Many US states published crisis standards of care (CSC) guidelines for allocating scarce critical care resources during the COVID-19 pandemic. However, the performance of these guidelines in maximizing their population benefit has not been well tested. In 2,272 adults with COVID-19 requiring mechanical ventilation drawn from the Study of the Treatment and Outcomes in Critically Ill Patients with COVID-19 (STOP-COVID) multicenter cohort, we test the following three approaches to CSC algorithms: Sequential Organ Failure Assessment (SOFA) scores grouped into ranges, SOFA score ranges plus comorbidities, and a hypothetical approach using raw SOFA scores not grouped into ranges. We find that area under receiver operating characteristic (AUROC) curves for all three algorithms demonstrate only modest discrimination for 28-day mortality. Adding comorbidity scoring modestly improves algorithm performance over SOFA scores alone. The algorithm incorporating comorbidities has modestly worse predictive performance for Black compared to white patients. CSC algorithms should be empirically examined to refine approaches to the allocation of scarce resources during pandemics and to avoid potential exacerbation of racial inequities.

OBJECTIVES:. Critically ill patients with coronavirus disease 2019 have variable mortality. Risk scores could improve care and be used for prognostic enrichment in trials. We aimed to compare machine learning algorithms and develop a simple tool for predicting 28-day mortality in ICU patients with coronavirus disease 2019. DESIGN:. This was an observational study of adult patients with coronavirus disease 2019. The primary outcome was 28-day inhospital mortality. Machine learning models and a simple tool were derived using variables from the first 48 hours of ICU admission and validated externally in independent sites and temporally with more recent admissions. Models were compared with a modified Sequential Organ Failure Assessment score, National Early Warning Score, and CURB-65 using the area under the receiver operating characteristic curve and calibration. SETTING:. Sixty-eight U.S. ICUs. PATIENTS:. Adults with coronavirus disease 2019 admitted to 68 ICUs in the United States between March 4, 2020, and June 29, 2020. INTERVENTIONS:. None. MEASUREMENTS AND MAIN RESULTS:. The study included 5,075 patients, 1,846 (36.4%) of whom died by day 28. eXtreme Gradient Boosting had the highest area under the receiver operating characteristic curve in external validation (0.81) and was well-calibrated, while k-nearest neighbors were the lowest performing machine learning algorithm (area under the receiver operating characteristic curve 0.69). Findings were similar with temporal validation. The simple tool, which was created using the most important features from the eXtreme Gradient Boosting model, had a significantly higher area under the receiver operating characteristic curve in external validation (0.78) than the Sequential Organ Failure Assessment score (0.69), National Early Warning Score (0.60), and CURB-65 (0.65; p < 0.05 for all comparisons). Age, number of ICU beds, creatinine, lactate, arterial pH, and Pao2/Fio2 ratio were the most important predictors in the eXtreme Gradient Boosting model. CONCLUSIONS:. eXtreme Gradient Boosting had the highest discrimination overall, and our simple tool had higher discrimination than a modified Sequential Organ Failure Assessment score, National Early Warning Score, and CURB-65 on external validation. These models could be used to improve triage decisions and clinical trial enrichment.