OBJECTIVES: To improve prediction of chronic kidney disease (CKD) progression to end-stage renal disease (ESRD) using machine learning (ML) and deep learning (DL) models applied to integrated clinical and claims data with varying observation windows, supported by explainable artificial intelligence (AI) to enhance interpretability and reduce bias. MATERIALS AND METHODS: We utilized data from 10 326 CKD patients, combining clinical and claims information from 2009 to 2018. After preprocessing, cohort identification, and feature engineering, we evaluated multiple statistical, ML and DL models using 5 distinct observation windows. Feature importance and SHapley Additive exPlanations (SHAP) analysis were employed to understand key predictors. Models were tested for robustness, clinical relevance, misclassification patterns, and bias. RESULTS: Integrated data models outperformed single data source models, with long short-term memory achieving the highest area under the receiver operating characteristic curve (AUROC) (0.93) and F1 score (0.65). A 24-month observation window optimally balanced early detection and prediction accuracy. The 2021 estimated glomerular filtration rate (eGFR) equation improved prediction accuracy and reduced racial bias, particularly for African American patients. DISCUSSION: Improved prediction accuracy, interpretability, and bias mitigation strategies have the potential to enhance CKD management, support targeted interventions, and reduce health-care disparities. CONCLUSION: This study presents a robust framework for predicting ESRD outcomes, improving clinical decision-making through integrated multisourced data and advanced analytics. Future research will expand data integration and extend this framework to other chronic diseases.

In this study, we use data from a novel 11-week cluster randomized controlled trial that we conducted in a school near Chennai (India) to evaluate the impact of regular exposure to an artificial intelligence(AI)-enabled, educational mobile health game – a low risk, non-invasive, digital vaccine candidate with neurocognitive training and implicit learning components – on students’ nutritional knowledge. This paper presents preliminary results from our analyses examining gameplay patterns of students’ who were exposed to the mHealth game, and to quantify association with their nutritional knowledge, captured from student surveys over the course of the intervention.

BACKGROUND: Studies confirm that significant biases exist in online recommendation platforms, exacerbating pre-existing disparities and leading to less-than-optimal outcomes for underrepresented demographics. We study issues of bias in inclusion and representativeness in the context of healthcare information disseminated via videos on the YouTube social media platform, a widely used online channel for multi-media rich information. With one in three US adults using the Internet to learn about a health concern, it is critical to assess inclusivity and representativeness regarding how health information is disseminated by digital platforms such as YouTube. METHODS: Leveraging methods from fair machine learning (ML), natural language processing and voice and facial recognition methods, we examine inclusivity and representativeness of video content presenters using a large corpus of videos and their metadata on a chronic condition (diabetes) extracted from the YouTube platform. Regression models are used to determine whether presenter demographics impact video popularity, measured by the video's average daily view count. A video that generates a higher view count is considered to be more popular. RESULTS: The voice and facial recognition methods predicted the gender and race of the presenter with reasonable success. Gender is predicted through voice recognition (accuracy = 78%, AUC = 76%), while the gender and race predictions use facial recognition (accuracy = 93%, AUC = 92% and accuracy = 82%, AUC = 80%, respectively). The gender of the presenter is more significant for video views only when the face of the presenter is not visible while videos with male presenters with no face visibility have a positive relationship with view counts. Furthermore, videos with white and male presenters have a positive influence on view counts while videos with female and non - white group have high view counts. CONCLUSION: Presenters' demographics do have an influence on average daily view count of videos viewed on social media platforms as shown by advanced voice and facial recognition algorithms used for assessing inclusion and representativeness of the video content. Future research can explore short videos and those at the channel level because popularity of the channel name and the number of videos associated with that channel do have an influence on view counts.

Early identification of patients at risk for postoperative complications can facilitate timely workups and treatments and improve health outcomes. Currently, a widely-used surgical risk calculator online web system developed by the American College of Surgeons (ACS) uses patients’ static features, e.g. gender, age, to assess the risk of postoperative complications. However, the most crucial signals that reflect the actual postoperative physical conditions of patients are usually real-time dynamic signals, including the vital signs of patients (e.g., heart rate, blood pressure) collected from postoperative monitoring. In this paper, we develop a dynamic postoperative complication risk scoring framework (DyCRS) to detect the “at-risk” patients in a real-time way based on postoperative sequential vital signs and static features. DyCRS is based on adaptations of the Hidden Markov Model (HMM) that captures hidden states as well as observable states to generate a real-time, probabilistic, complication risk score. Evaluating our model using electronic health record (EHR) on elective Colectomy surgery from a major health system, we show that DyCRS significantly outperforms the state-of-the-art ACS calculator and real-time predictors with 50.16% area under precision-recall curve (AUCPRC) gain on average in terms of detection effectiveness. In terms of earliness, our DyCRS can predict 15hrs55mins earlier on average than clinician’s diagnosis with the recall of 60% and precision of 55%. Furthermore, Our DyCRS can extract interpretable patients’ stages, which are consistent with previous medical postoperative complication studies. We believe that our contributions demonstrate significant promise for developing a more accurate, robust and interpretable postoperative complication risk scoring system, which can benefit more than 50 million annual surgeries in the US by substantially lowering adverse events and healthcare costs.

OBJECTIVE: This integrative review identifies and analyzes the extant literature to examine the integration of social determinants of health (SDoH) domains into electronic health records (EHRs), their impact on risk prediction, and the specific outcomes and SDoH domains that have been tracked. MATERIALS AND METHODS: In accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, we conducted a literature search in the PubMed, CINAHL, Cochrane, EMBASE, and PsycINFO databases for English language studies published until March 2020 that examined SDoH domains in the context of EHRs. RESULTS: Our search strategy identified 71 unique studies that are directly related to the research questions. 75% of the included studies were published since 2017, and 68% were U.S.-based. 79% of the reviewed articles integrated SDoH information from external data sources into EHRs, and the rest of them extracted SDoH information from unstructured clinical notes in the EHRs. We found that all but 1 study using external area-level SDoH data reported minimum contribution to performance improvement in the predictive models. In contrast, studies that incorporated individual-level SDoH data reported improved predictive performance of various outcomes such as service referrals, medication adherence, and risk of 30-day readmission. We also found little consensus on the SDoH measures used in the literature and current screening tools. CONCLUSIONS: The literature provides early and rapidly growing evidence that integrating individual-level SDoH into EHRs can assist in risk assessment and predicting healthcare utilization and health outcomes, which further motivates efforts to collect and standardize patient-level SDoH information.

Left ventricular assist devices (LVADs) are an increasingly common therapy for patients with advanced heart failure. However, implantation of the LVAD increases the risk of stroke, infection, bleeding, and other serious adverse events (AEs). Most post-LVAD AEs studies have focused on individual AEs in isolation, neglecting the possible interrelation, or causality between AEs. This study is the first to conduct an exploratory analysis to discover common sequential chains of AEs following LVAD implantation that are correlated with important clinical outcomes. This analysis was derived from 58,575 recorded AEs for 13,192 patients in International Registry for Mechanical Circulatory Support (INTERMACS) who received a continuous-flow LVAD between 2006 and 2015. The pattern mining procedure involved three main steps: (1) creating a bank of AE sequences by converting the AEs for each patient into a single, chronologically sequenced record, (2) grouping patients with similar AE sequences using hierarchical clustering, and (3) extracting temporal chains of AEs for each group of patients using Markov modeling. The mined results indicate the existence of seven groups of sequential chains of AEs, characterized by common types of AEs that occurred in a unique order. The groups were identified as: GRP1: Recurrent bleeding, GRP2: Trajectory of device malfunction & explant, GRP3: Infection, GRP4: Trajectories to transplant, GRP5: Cardiac arrhythmia, GRP6: Trajectory of neurological dysfunction & death, and GRP7: Trajectory of respiratory failure, renal dysfunction & death. These patterns of sequential post-LVAD AEs disclose potential interdependence between AEs and may aid prediction, and prevention, of subsequent AEs in future studies.

Almost 15% of global deaths in 2008 were attributed to Coronary Heart Disease (CHD). While major risk factors for CHD are widely known, lack of comprehensive data on relevant risk factors has limited the ability to predict risk of developing CHD in large populations. In this study, we explore the application of the Framingham Risk Model to predict CHD risk using a limited set of attributes present in a health risk assessment (HRA) dataset from a digital health company. HRAs often fail to capture all the needed attributes of the Framingham Model, such as LDL and HDL cholesterol values that significantly affect CHD risk. Hence, we enhance our analysis with the National Health and Nutrition Examination Survey (NHANES) data from the Centers for Disease Control (CDC), the United States public health agency. Our preliminary findings indicate that HRA data can be successfully used as input for the Framingham Risk Model in predicting risk of CHD utilizing NHANES data to predict missing attributes, thus extending the use of HRAs for disease risk prediction.

Recent research has highlighted the need for upstream planning in healthcare service delivery systems, patient scheduling, and resource allocation in the hospital inpatient setting. This study examines the value of upstream planning within hospital-wide resource allocation decisions based on machine learning (ML) and mixed-integer programming (MIP), focusing on prediction of diagnosis-related groups (DRGs) and the use of these predictions for allocating scarce hospital resources. DRGs are a payment scheme employed at patients’ discharge, where the DRG and length of stay determine the revenue that the hospital obtains. We show that early and accurate DRG classification using ML methods, incorporated into an MIP-based resource allocation model, can increase the hospital’s contribution margin, the number of admitted patients, and the utilization of resources such as operating rooms and beds. We test these methods on hospital data containing more than 16,000 inpatient records and demonstrate improved DRG classification accuracy as compared to the hospital’s current approach. The largest improvements were observed at and before admission, when information such as procedures and diagnoses is typically incomplete, but performance was improved even after a substantial portion of the patient’s length of stay, and under multiple scenarios making different assumptions about the available information. Using the improved DRG predictions within our resource allocation model improves contribution margin by 2.9% and the utilization of scarce resources such as operating rooms and beds from 66.3% to 67.3% and from 70.7% to 71.7%, respectively. This enables 9.0% more nonurgent elective patients to be admitted as compared to the baseline.

OBJECTIVE: Evidence suggests that the medication lists of patients are often incomplete and could negatively affect patient outcomes. In this article, the authors propose the application of collaborative filtering methods to the medication reconciliation task. Given a current medication list for a patient, the authors employ collaborative filtering approaches to predict drugs the patient could be taking but are missing from their observed list. DESIGN: The collaborative filtering approach presented in this paper emerges from the insight that an omission in a medication list is analogous to an item a consumer might purchase from a product list. Online retailers use collaborative filtering to recommend relevant products using retrospective purchase data. In this article, the authors argue that patient information in electronic medical records, combined with artificial intelligence methods, can enhance medication reconciliation. The authors formulate the detection of omissions in medication lists as a collaborative filtering problem. Detection of omissions is accomplished using several machine-learning approaches. The effectiveness of these approaches is evaluated using medication data from three long-term care centers. The authors also propose several decision-theoretic extensions to the methodology for incorporating medical knowledge into recommendations. RESULTS: Results show that collaborative filtering identifies the missing drug in the top-10 list about 40-50% of the time and the therapeutic class of the missing drug 50%-65% of the time at the three clinics in this study. CONCLUSION: Results suggest that collaborative filtering can be a valuable tool for reconciling medication lists, complementing currently recommended process-driven approaches. However, a one-size-fits-all approach is not optimal, and consideration should be given to context (eg, types of patients and drug regimens) and consequence (eg, the impact of omission on outcomes).

Unpredictable yet frequently occurring exception situations pervade clinical care. Handling them properly often requires aberrant actions temporarily departing from normal practice. In this study, the authors investigated several exception-handling procedures provided in an electronic health records system for facilitating clinical documentation, which the authors refer to as 'data entry exit strategies.' Through a longitudinal analysis of computer-recorded usage data, the authors found that (1) utilization of the exit strategies was not affected by postimplementation system maturity or patient visit volume, suggesting clinicians' needs to 'exit' unwanted situations are persistent; and (2) clinician type and gender are strong predictors of exit-strategy usage. Drilldown analyses further revealed that the exit strategies were judiciously used and enabled actions that would be otherwise difficult or impossible. However, many data entries recorded via them could have been 'properly' documented, yet were not, and a considerable proportion containing temporary or incomplete information was never subsequently amended. These findings may have significant implications for the design of safer and more user-friendly point-of-care information systems for healthcare.

OBJECTIVE: To study how social interactions influence physician adoption of an electronic health records (EHR) system. DESIGN: A social network survey was used to delineate the structure of social interactions among 40 residents and 15 attending physicians in an ambulatory primary care practice. Social network analysis was then applied to relate the interaction structures to individual physicians' utilization rates of an EHR system. MEASUREMENTS: The social network survey assessed three distinct types of interaction structures: professional network based on consultation on patient care-related matters; friendship network based on personal intimacy; and perceived influence network based on a person's perception of how other people have affected her intention to adopt the EHR system. EHR utilization rates were measured as the proportion of patient visits in which sentinel use events consisting of patient data documentation or retrieval activities were recorded. The usage data were collected over a time period of 14 months from computer-recorded audit trail logs. RESULTS: Neither the professional nor the perceived influence network is correlated with EHR usage. The structure of the friendship network significantly influenced individual physicians' adoption of the EHR system. Residents who occupied similar social positions in the friendship network shared similar EHR utilization rates (p<0.05). In other words, residents who had personal friends in common tended to develop comparable levels of EHR adoption. This effect is particularly prominent when the mutual personal friends of these 'socially similar' residents were attending physicians (p<0.001). CONCLUSIONS: Social influence affecting physician adoption of EHR seems to be predominantly conveyed through interactions with personal friends rather than interactions in professional settings.

OBJECTIVES: This study sought to investigate user interactions with an electronic health records (EHR) system by uncovering hidden navigational patterns in the EHR usage data automatically recorded as clinicians navigated through the system's software user interface (UI) to perform different clinical tasks. DESIGN: A homegrown EHR was adapted to allow real-time capture of comprehensive UI interaction events. These events, constituting time-stamped event sequences, were used to replay how the EHR was used in actual patient care settings. The study site is an ambulatory primary care clinic at an urban teaching hospital. Internal medicine residents were the primary EHR users. MEASUREMENTS: Computer-recorded event sequences reflecting the order in which different EHR features were sequentially accessed. METHODS: We apply sequential pattern analysis (SPA) and a first-order Markov chain model to uncover recurring UI navigational patterns. RESULTS: Of 17 main EHR features provided in the system, SPA identified 3 bundled features: "Assessment and Plan" and "Diagnosis," "Order" and "Medication," and "Order" and "Laboratory Test." Clinicians often accessed these paired features in a bundle together in a continuous sequence. The Markov chain analysis revealed a global navigational pathway, suggesting an overall sequential order of EHR feature accesses. "History of Present Illness" followed by "Social History" and then "Assessment and Plan" was identified as an example of such global navigational pathways commonly traversed by the EHR users. CONCLUSION: Users showed consistent UI navigational patterns, some of which were not anticipated by system designers or the clinic management. Awareness of such unanticipated patterns may help identify undesirable user behavior as well as reengineering opportunities for improving the system's usability.

Data sets with many discrete variables and relatively few cases arise in health care, e-commerce, information security, text mining, and many other domains. Learning effective and efficient prediction models from such data sets is a challenging task. In this paper, we propose a tabu search-enhanced Markov blanket (TS/MB) algorithm to learn a graphical Markov blanket model for classification of high-dimensional data sets. The TS/MB algorithm makes use of Markov blanket neighborhoods: restricted neighborhoods in a general Bayesian network based on the Markov condition. Computational results from real-world data sets drawn from several domains indicate that the TS/MB algorithm, when used as a feature selection method, is able to find a parsimonious model with substantially fewer predictor variables than is present in the full data set. The algorithm also provides good prediction performance when used as a graphical classifier compared with several machine-learning methods.

A version of cobweb/classit is proposed to incrementally cluster text documents into cluster hierarchies. The modification to classit consists of changes to the underlying distributional assumption of the original algorithm that are suggested by text document data. Both the algorithms are evaluated using standard text document datasets. We show that the modified algorithm performs better than the original Classit when presented with Reuters newswire articles in temporal order, i.e., the order in which they are going to be presented in real life situation. It also performs better than the original Classit on the larger of eleven standard text clustering datasets we used.

An important factor affecting the performance of distributed simulations running on parallel-processing computers is the allocation of logical processes to the available physical processors. An inefficient allocation can result in excessive communication times and unfavorable load conditions. This leads to long run times, possibly giving performance worse than that with a uniprocessor sequential event-list implementation. But the efficiency of any allocation strategy is dependent on the metric, or measure, it uses to characterize the load in the distributed system. This paper presents a simple and intuitive way of measuring and reallocating the load when the objective is to minimize simulation run time. The metric, based on estimating measures of message utilization at each processor, has been used in an adaptive scheme for load allocation, and experiments on an iPSC/2 Hypercube indicate that it successfully characterizes the load for purposes of reducing simulation run time. INFORMS Journal on Computing, ISSN 1091-9856, was published as ORSA Journal on Computing from 1989 to 1995 under ISSN 0899-1499.