PheMIME: an interactive web app and knowledge base for phenome-wide, multi-institutional multimorbidity analysis.

OBJECTIVES: To address the need for interactive visualization tools and databases in characterizing multimorbidity patterns across different populations, we developed the Phenome-wide Multi-Institutional Multimorbidity Explorer (PheMIME). This tool leverages three large-scale EHR systems to facilitate efficient analysis and visualization of disease multimorbidity, aiming to reveal both robust and novel disease associations that are consistent across different systems and to provide insight for enhancing personalized healthcare strategies. MATERIALS AND METHODS: PheMIME integrates summary statistics from phenome-wide analyses of disease multimorbidities, utilizing data from Vanderbilt University Medical Center, Mass General Brigham, and the UK Biobank. It offers interactive and multifaceted visualizations for exploring multimorbidity. Incorporating an enhanced version of associationSubgraphs, PheMIME also enables dynamic analysis and inference of disease clusters, promoting the discovery of complex multimorbidity patterns. A case study on schizophrenia demonstrates its capability for generating interactive visualizations of multimorbidity networks within and across multiple systems. Additionally, PheMIME supports diverse multimorbidity-based discoveries, detailed further in online case studies. RESULTS: The PheMIME is accessible at https://prod.tbilab.org/PheMIME/. A comprehensive tutorial and multiple case studies for demonstration are available at https://prod.tbilab.org/PheMIME_supplementary_materials/. The source code can be downloaded from https://github.com/tbilab/PheMIME. DISCUSSION: PheMIME represents a significant advancement in medical informatics, offering an efficient solution for accessing, analyzing, and interpreting the complex and noisy real-world patient data in electronic health records. CONCLUSION: PheMIME provides an extensive multimorbidity knowledge base that consolidates data from three EHR systems, and it is a novel interactive tool designed to analyze and visualize multimorbidities across multiple EHR datasets. It stands out as the first of its kind to offer extensive multimorbidity knowledge integration with substantial support for efficient online analysis and interactive visualization.

Model-based estimation of individual-level social determinants of health and its applications in All of Us.

OBJECTIVES: We introduce a widely applicable model-based approach for estimating individual-level Social Determinants of Health (SDoH) and evaluate its effectiveness using the All of Us Research Program. MATERIALS AND METHODS: Our approach utilizes aggregated SDoH datasets to estimate individual-level SDoH, demonstrated with examples of no high school diploma (NOHSDP) and no health insurance (UNINSUR) variables. Models are estimated using American Community Survey data and applied to derive individual-level estimates for All of Us participants. We assess concordance between model-based SDoH estimates and self-reported SDoHs in All of Us and examine associations with undiagnosed hypertension and diabetes. RESULTS: Compared to self-reported SDoHs, the area under the curve for NOHSDP is 0.727 (95% CI, 0.724-0.730) and for UNINSUR is 0.730 (95% CI, 0.727-0.733) among the 329 074 All of Us participants, both significantly higher than aggregated SDoHs. The association between model-based NOHSDP and undiagnosed hypertension is concordant with those estimated using self-reported NOHSDP, with a correlation coefficient of 0.649. Similarly, the association between model-based NOHSDP and undiagnosed diabetes is concordant with those estimated using self-reported NOHSDP, with a correlation coefficient of 0.900. DISCUSSION AND CONCLUSION: The model-based SDoH estimation method offers a scalable and easily standardized approach for estimating individual-level SDoHs. Using the All of Us dataset, we demonstrate reasonable concordance between model-based SDoH estimates and self-reported SDoHs, along with consistent associations with health outcomes. Our findings also underscore the critical role of geographic contexts in SDoH estimation and in evaluating the association between SDoHs and health outcomes.

Early diagnosis of HIV cases by means of text mining and machine learning models on clinical notes.

Undiagnosed and untreated human immunodeficiency virus (HIV) infection increases morbidity in the HIV-positive person and allows onward transmission of the virus. Minimizing missed opportunities for HIV diagnosis when a patient visits a healthcare facility is essential in restraining the epidemic and working toward its eventual elimination. Most state-of-the-art proposals employ machine learning (ML) methods and structured data to enhance HIV diagnoses, however, there is a dearth of recent proposals utilizing unstructured textual data from Electronic Health Records (EHRs). In this work, we propose to use only the unstructured text of the clinical notes as evidence for the classification of patients as suspected or not suspected. For this purpose, we first compile a dataset of real clinical notes from a hospital with patients classified as suspects and non-suspects of having HIV. Then, we evaluate the effectiveness of two types of classification models to identify patients suspected of being infected with the virus: classical ML algorithms and two Large Language Models (LLMs) from the biomedical domain in Spanish. The results show that both LLMs outperform classical ML algorithms in the two settings we explore: one dataset version is balanced, containing an equal number of suspicious and non-suspicious patients, while the other reflects the real distribution of patients in the hospital, being unbalanced. We obtain F1 score figures of 94.7 with both LLMs in the unbalanced setting, while in the balance one, RoBERTaBio model outperforms the other one with a F1 score of 95.7. The findings indicate that leveraging unstructured text with LLMs in the biomedical domain yields promising outcomes in diminishing missed opportunities for HIV diagnosis. A tool based on our system could assist a doctor in deciding whether a patient in consultation should undergo a serological test.

TEE4EHR: Transformer event encoder for better representation learning in electronic health records.

Irregular sampling of time series in electronic health records (EHRs) is one of the main challenges for developing machine learning models. Additionally, the pattern of missing values in certain clinical variables is not at random but depends on the decisions of clinicians and the state of the patient. Point process is a mathematical framework for analyzing event sequence data consistent with irregular sampling patterns. Our model, TEE4EHR, is a transformer event encoder (TEE) with point process loss that encodes the pattern of laboratory tests in EHRs. The utility of our TEE has been investigated in various benchmark event sequence datasets. Additionally, we conduct experiments on two real-world EHR databases to provide a more comprehensive evaluation of our model. Firstly, in a self-supervised learning approach, the TEE is jointly learned with an existing attention-based deep neural network, which gives superior performance in negative log-likelihood and future event prediction. Besides, we propose an algorithm for aggregating attention weights to reveal the events' interactions. Secondly, we transfer and freeze the learned TEE to the downstream task for the outcome prediction, where it outperforms state-of-the-art models for handling irregularly sampled time series. Furthermore, our results demonstrate that our approach can improve representation learning in EHRs and be useful for clinical prediction tasks.

Forecasting acute kidney injury and resource utilization in ICU patients using longitudinal, multimodal models.

BACKGROUND: Advances in artificial intelligence (AI) have realized the potential of revolutionizing healthcare, such as predicting disease progression via longitudinal inspection of Electronic Health Records (EHRs) and lab tests from patients admitted to Intensive Care Units (ICU). Although substantial literature exists addressing broad subjects, including the prediction of mortality, length-of-stay, and readmission, studies focusing on forecasting Acute Kidney Injury (AKI), specifically dialysis anticipation like Continuous Renal Replacement Therapy (CRRT) are scarce. The technicality of how to implement AI remains elusive. OBJECTIVE: This study aims to elucidate the important factors and methods that are required to develop effective predictive models of AKI and CRRT for patients admitted to ICU, using EHRs in the Medical Information Mart for Intensive Care (MIMIC) database. METHODS: We conducted a comprehensive comparative analysis of established predictive models, considering both time-series measurements and clinical notes from MIMIC-IV databases. Subsequently, we proposed a novel multi-modal model which integrates embeddings of top-performing unimodal models, including Long Short-Term Memory (LSTM) and BioMedBERT, and leverages both unstructured clinical notes and structured time series measurements derived from EHRs to enable the early prediction of AKI and CRRT. RESULTS: Our multimodal model achieved a lead time of at least 12 h ahead of clinical manifestation, with an Area Under the Receiver Operating Characteristic Curve (AUROC) of 0.888 for AKI and 0.997 for CRRT, as well as an Area Under the Precision Recall Curve (AUPRC) of 0.727 for AKI and 0.840 for CRRT, respectively, which significantly outperformed the baseline models. Additionally, we performed a SHapley Additive exPlanation (SHAP) analysis using the expected gradients algorithm, which highlighted important, previously underappreciated predictive features for AKI and CRRT. CONCLUSION: Our study revealed the importance and the technicality of applying longitudinal, multimodal modeling to improve early prediction of AKI and CRRT, offering insights for timely interventions. The performance and interpretability of our model indicate its potential for further assessment towards clinical applications, to ultimately optimize AKI management and enhance patient outcomes.

Inherent Bias in Electronic Health Records: A Scoping Review of Sources of Bias.

Objectives: 1.1Biases inherent in electronic health records (EHRs), and therefore in medical artificial intelligence (AI) models may significantly exacerbate health inequities and challenge the adoption of ethical and responsible AI in healthcare. Biases arise from multiple sources, some of which are not as documented in the literature. Biases are encoded in how the data has been collected and labeled, by implicit and unconscious biases of clinicians, or by the tools used for data processing. These biases and their encoding in healthcare records undermine the reliability of such data and bias clinical judgments and medical outcomes. Moreover, when healthcare records are used to build data-driven solutions, the biases are further exacerbated, resulting in systems that perpetuate biases and induce healthcare disparities. This literature scoping review aims to categorize the main sources of biases inherent in EHRs. Methods: 1.2We queried PubMed and Web of Science on January 19th, 2023, for peer-reviewed sources in English, published between 2016 and 2023, using the PRISMA approach to stepwise scoping of the literature. To select the papers that empirically analyze bias in EHR, from the initial yield of 430 papers, 27 duplicates were removed, and 403 studies were screened for eligibility. 196 articles were removed after the title and abstract screening, and 96 articles were excluded after the full-text review resulting in a final selection of 116 articles. Results: 1.3Systematic categorizations of diverse sources of bias are scarce in the literature, while the effects of separate studies are often convoluted and methodologically contestable. Our categorization of published empirical evidence identified the six main sources of bias: a) bias arising from past clinical trials; b) data-related biases arising from missing, incomplete information or poor labeling of data; human-related bias induced by c) implicit clinician bias, d) referral and admission bias; e) diagnosis or risk disparities bias and finally, (f) biases in machinery and algorithms. Conclusions: 1.4Machine learning and data-driven solutions can potentially transform healthcare delivery, but not without limitations. The core inputs in the systems (data and human factors) currently contain several sources of bias that are poorly documented and analyzed for remedies. The current evidence heavily focuses on data-related biases, while other sources are less often analyzed or anecdotal. However, these different sources of biases add to one another exponentially. Therefore, to understand the issues holistically we need to explore these diverse sources of bias. While racial biases in EHR have been often documented, other sources of biases have been less frequently investigated and documented (e.g. gender-related biases, sexual orientation discrimination, socially induced biases, and implicit, often unconscious, human-related cognitive biases). Moreover, some existing studies lack causal evidence, illustrating the different prevalences of disease across groups, which does not per se prove the causality. Our review shows that data-, human- and machine biases are prevalent in healthcare and they significantly impact healthcare outcomes and judgments and exacerbate disparities and differential treatment. Understanding how diverse biases affect AI systems and recommendations is critical. We suggest that researchers and medical personnel should develop safeguards and adopt data-driven solutions with a "bias-in-mind" approach. More empirical evidence is needed to tease out the effects of different sources of bias on health outcomes.

User Personas for a "Better Design" of Nation-Wide EHRs Based on Thorough Expert Evaluation and Field Analysis: Modeling Users as Individuals Plus Family Members for an Enhanced Mapping of Healthcare Situations.

Electronic Health Records (EHRs) are pivotal in prevention, therapy, and care. Their design necessitates the representation of users, activities, context, and technology. Among various participative and ethnographic design methods, user personas are an effective tool for encapsulating users in the design process. Goals and methods: This research focused on the creation of user personas for the design of a nation-wide EHR, specifically the German "elektronische Patientenakte" (ePA). We employed qualitative methods, such as field analysis and expert workshops, to generate, assess, and refine a set of user personas that can cover the complexity of real-life familial care environments. We used an innovative bottom-up approach applying a whole new process for persona generation especially in the context of family management. Results: The research yielded an initial set of five personas that accurately represent fictional user types. Importantly, at least two of these personas encapsulate the unique challenges inherent in family care work. These results provide a foundation for future work, which can utilize these personas for EHR design, as well as for further evaluation and refinement.

Forecasting Acute Kidney Injury and Resource Utilization in ICU patients using longitudinal, multimodal models.

Background: Advances in artificial intelligence (AI) have realized the potential of revolutionizing healthcare, such as predicting disease progression via longitudinal inspection of Electronic Health Records (EHRs) and lab tests from patients admitted to Intensive Care Units (ICU). Although substantial literature exists addressing broad subjects, including the prediction of mortality, length-of-stay, and readmission, studies focusing on forecasting Acute Kidney Injury (AKI), specifically dialysis anticipation like Continuous Renal Replacement Therapy (CRRT) are scarce. The technicality of how to implement AI remains elusive. Objective: This study aims to elucidate the important factors and methods that are required to develop effective predictive models of AKI and CRRT for patients admitted to ICU, using EHRs in the Medical Information Mart for Intensive Care (MIMIC) database. Methods: We conducted a comprehensive comparative analysis of established predictive models, considering both time-series measurements and clinical notes from MIMIC-IV databases. Subsequently, we proposed a novel multi-modal model which integrates embeddings of top-performing unimodal models, including Long Short-Term Memory (LSTM) and BioMedBERT, and leverages both unstructured clinical notes and structured time series measurements derived from EHRs to enable the early prediction of AKI and CRRT. Results: Our multimodal model achieved a lead time of at least 12 hours ahead of clinical manifestation, with an Area Under the Receiver Operating Characteristic Curve (AUROC) of 0.888 for AKI and 0.997 for CRRT, as well as an Area Under the Precision Recall Curve (AUPRC) of 0.727 for AKI and 0.840 for CRRT, respectively, which significantly outperformed the baseline models. Additionally, we performed a SHapley Additive exPlanation (SHAP) analysis using the expected gradients algorithm, which highlighted important, previously underappreciated predictive features for AKI and CRRT. Conclusion: Our study revealed the importance and the technicality of applying longitudinal, multimodal modeling to improve early prediction of AKI and CRRT, offering insights for timely interventions. The performance and interpretability of our model indicate its potential for further assessment towards clinical applications, to ultimately optimize AKI management and enhance patient outcomes.

Hybrid artificial intelligence outcome prediction using features extraction from stress perfusion cardiac magnetic resonance images and electronic health records.

Background: Prediction of clinical outcomes in coronary artery disease (CAD) has been conventionally achieved using clinical risk factors. The relationship between imaging features and outcome is still not well understood. This study aims to use artificial intelligence to link image features with mortality outcome. Methods: A retrospective study was performed on patients who had stress perfusion cardiac magnetic resonance (SP-CMR) between 2011 and 2021. The endpoint was all-cause mortality. Convolutional neural network (CNN) was used to extract features from stress perfusion images, and multilayer perceptron (MLP) to extract features from electronic health records (EHRs), both networks were concatenated in a hybrid neural network (HNN) to predict study endpoint. Image CNN was trained to predict study endpoint directly from images. HNN and image CNN were compared with a linear clinical model using area under the curve (AUC), F1 scores, and McNemar's test. Results: Total of 1,286 cases were identified, with 201 death events (16%). The clinical model had good performance (AUC =80%, F1 score =37%). Best Image CNN model showed AUC =72% and F1 score =38%. HNN outperformed the other two models (AUC =82%, F1 score =43%). McNemar's test showed statistical difference between image CNN and both clinical model (P<0.01) and HNN (P<0.01). There was no significant difference between HNN and clinical model (P=0.15). Conclusions: Death in patients with suspected or known CAD can be predicted directly from stress perfusion images without clinical knowledge. Prediction can be improved by HNN that combines clinical and SP-CMR images.

Making commercial bracelet smarter with a biochemical button module.

Despite the rapid development of mobile health based on wearable devices in recent years, lack of access to biochemical detection remains a vital challenge for most commercial wearable devices, which hinders the provision of effective electronic health records (EHRs) for disease control strategies, and further constraining the development of personalized precision medicine. Herein, we propose a strategy to graft biochemical detection function onto commercial bracelet. Different from the conventional development process of designing a completely new wearable biochemical device, we prefer to upgrade existing commercial wearable device to achieve simpler, faster, and more effective research and commercialization processes. An affordable and user-friendly biochemical button module has been designed that enables to integrate sensitive, specific, and rapid biochemical detection function into the idle space on the strap of the bracelet without increasing the size of the main body. This "Smart Bracelet Plus" shows the ability to simultaneously monitor physical and biochemical signals, and will serve as a reliable and systematic personal diagnostics and monitoring platform for providing real-time EHRs for disease control strategies and improving the efficiency of the healthcare system.

Developing a computable phenotype for glioblastoma.

BACKGROUND: Glioblastoma is the most common malignant brain tumor, and thus it is important to be able to identify patients with this diagnosis for population studies. However, this can be challenging as diagnostic codes are nonspecific. The aim of this study was to create a computable phenotype (CP) for glioblastoma multiforme (GBM) from structured and unstructured data to identify patients with this condition in a large electronic health record (EHR). METHODS: We used the University of Florida (UF) Health Integrated Data Repository, a centralized clinical data warehouse that stores clinical and research data from various sources within the UF Health system, including the EHR system. We performed multiple iterations to refine the GBM-relevant diagnosis codes, procedure codes, medication codes, and keywords through manual chart review of patient data. We then evaluated the performances of various possible proposed CPs constructed from the relevant codes and keywords. RESULTS: We underwent six rounds of manual chart reviews to refine the CP elements. The final CP algorithm for identifying GBM patients was selected based on the best F1-score. Overall, the CP rule "if the patient had at least 1 relevant diagnosis code and at least 1 relevant keyword" demonstrated the highest F1-score using both structured and unstructured data. Thus, it was selected as the best-performing CP rule. CONCLUSIONS: We developed and validated a CP algorithm for identifying patients with GBM using both structured and unstructured EHR data from a large tertiary care center. The final algorithm achieved an F1-score of 0.817, indicating a high performance, which minimizes possible biases from misclassification errors.

Evolving availability and standardization of patient attributes for matching.

Variation in availability, format, and standardization of patient attributes across health care organizations impacts patient-matching performance. We report on the changing nature of patient-matching features available from 2010-2020 across diverse care settings. We asked 38 health care provider organizations about their current patient attribute data-collection practices. All sites collected name, date of birth (DOB), address, and phone number. Name, DOB, current address, social security number (SSN), sex, and phone number were most commonly used for cross-provider patient matching. Electronic health record queries for a subset of 20 participating sites revealed that DOB, first name, last name, city, and postal codes were highly available (>90%) across health care organizations and time. SSN declined slightly in the last years of the study period. Birth sex, gender identity, language, country full name, country abbreviation, health insurance number, ethnicity, cell phone number, email address, and weight increased over 50% from 2010 to 2020. Understanding the wide variation in available patient attributes across care settings in the United States can guide selection and standardization efforts for improved patient matching in the United States.

Developing Clinical Prediction Models Using Primary Care Electronic Health Record Data: The Impact of Data Preparation Choices on Model Performance.

Objective: To quantify prediction model performance in relation to data preparation choices when using electronic health records (EHR). Study Design and Setting: Cox proportional hazards models were developed for predicting the first-ever main adverse cardiovascular events using Dutch primary care EHR data. The reference model was based on a 1-year run-in period, cardiovascular events were defined based on both EHR diagnosis and medication codes, and missing values were multiply imputed. We compared data preparation choices based on (i) length of the run-in period (2- or 3-year run-in); (ii) outcome definition (EHR diagnosis codes or medication codes only); and (iii) methods addressing missing values (mean imputation or complete case analysis) by making variations on the derivation set and testing their impact in a validation set. Results: We included 89,491 patients in whom 6,736 first-ever main adverse cardiovascular events occurred during a median follow-up of 8 years. Outcome definition based only on diagnosis codes led to a systematic underestimation of risk (calibration curve intercept: 0.84; 95% CI: 0.83-0.84), while complete case analysis led to overestimation (calibration curve intercept: -0.52; 95% CI: -0.53 to -0.51). Differences in the length of the run-in period showed no relevant impact on calibration and discrimination. Conclusion: Data preparation choices regarding outcome definition or methods to address missing values can have a substantial impact on the calibration of predictions, hampering reliable clinical decision support. This study further illustrates the urgency of transparent reporting of modeling choices in an EHR data setting.