Search

While there exist numerous methods to identify binary phenotypes (i.e. COPD) using electronic health record (EHR) data, few exist to ascertain the timings of phenotype events (i.e. COPD onset or exacerbations). Estimating event times could enable more powerful use of EHR data for longitudinal risk modeling, including survival analysis. Here we introduce Semi-supervised Adaptive Markov Gaussian Embedding Process (SAMGEP), a semi-supervised machine learning algorithm to estimate phenotype event times using EHR data with limited observed labels, which require resource-intensive chart review to obtain. SAMGEP models latent phenotype states as a binary Markov process, and it employs an adaptive weighting strategy to map timestamped EHR features to an embedding function that it models as a state-dependent Gaussian process. SAMGEP’s feature weighting achieves meaningful feature selection, and its predictions significantly improve AUCs and F1 scores over existing approaches in diverse simulations and real-world settings. It is particularly adept at predicting cumulative risk and event counting process functions, and is robust to diverse generative model parameters. Moreover, it achieves high accuracy with few (50–100) labels, efficiently leveraging unlabeled EHR data to maximize information gain from costly-to-obtain event time labels. SAMGEP can be used to estimate accurate phenotype state functions for risk modeling research.

OBJECTIVE Identifying pseudogout in large data sets is difficult due to its episodic nature and a lack of billing codes specific to this acute subtype of calcium pyrophosphate (CPP) deposition disease. The objective of this study was to evaluate a novel machine learning approach for classifying pseudogout using electronic health record (EHR) data. METHODS We created an EHR data mart of patients with ≥1 relevant billing code or ≥2 natural language processing (NLP) mentions of pseudogout or chondrocalcinosis, 1991-2017. We selected 900 subjects for gold standard chart review for definite pseudogout (synovitis + synovial fluid CPP crystals), probable pseudogout (synovitis + chondrocalcinosis), or not pseudogout. We applied a topic modeling approach to identify definite/probable pseudogout. A combined algorithm included topic modeling plus manually reviewed CPP crystal results. We compared algorithm performance and cohorts identified by billing codes, the presence of CPP crystals, topic modeling, and a combined algorithm. RESULTS Among 900 subjects, 123 (13.7%) had pseudogout by chart review (68 definite, 55 probable). Billing codes had a sensitivity of 65% and a positive predictive value (PPV) of 22% for pseudogout. The presence of CPP crystals had a sensitivity of 29% and a PPV of 92%. Without using CPP crystal results, topic modeling had a sensitivity of 29% and a PPV of 79%. The combined algorithm yielded a sensitivity of 42% and a PPV of 81%. The combined algorithm identified 50% more patients than the presence of CPP crystals; the latter captured a portion of definite pseudogout and missed probable pseudogout. CONCLUSION For pseudogout, an episodic disease with no specific billing code, combining NLP, machine learning methods, and synovial fluid laboratory results yielded an algorithm that significantly boosted the PPV compared to billing codes.

While there exist numerous methods to identify binary phenotypes (i.e. COPD) using electronic health record (EHR) data, few exist to ascertain the timings of phenotype events (i.e. COPD onset or exacerbations). Estimating event times could enable more powerful use of EHR data for longitudinal risk modeling, including survival analysis. Here we introduce Semi-supervised Adaptive Markov Gaussian Embedding Process (SAMGEP), a semi-supervised machine learning algorithm to estimate phenotype event times using EHR data with limited observed labels, which require resource-intensive chart review to obtain. SAMGEP models latent phenotype states as a binary Markov process, and it employs an adaptive weighting strategy to map timestamped EHR features to an embedding function that it models as a state-dependent Gaussian process. SAMGEP’s feature weighting achieves meaningful feature selection, and its predictions significantly improve AUCs and F1 scores over existing approaches in diverse simulations and real-world settings. It is particularly adept at predicting cumulative risk and event counting process functions, and is robust to diverse generative model parameters. Moreover, it achieves high accuracy with few (50-100) labels, efficiently leveraging unlabeled EHR data to maximize information gain from costly-to-obtain event time labels. SAMGEP can be used to estimate accurate phenotype state functions for risk modeling research.

Electronic Health Records (EHRs) contain rich clinical data collected at the point of the care, and their increasing adoption offers exciting opportunities for clinical informatics, disease risk prediction, and personalized treatment recommendation. However, effective use of EHR data for research and clinical decision support is often hampered by a lack of reliable disease labels. To compile gold-standard labels, researchers often rely on clinical experts to develop rule-based phenotyping algorithms from billing codes and other surrogate features. This process is tedious and error-prone due to recall and observer biases in how codes and measures are selected, and some phenotypes are incompletely captured by a handful of surrogate features. To address this challenge, we present a novel automatic phenotyping model called MixEHR-Guided (MixEHR-G), a multimodal hierarchical Bayesian topic model that efficiently models the EHR generative process by identifying latent phenotype structure in the data. Unlike existing topic modeling algorithms wherein the inferred topics are not identifiable, MixEHR-G uses prior information from informative surrogate features to align topics with known phenotypes. We applied MixEHR-G to an openly-available EHR dataset of 38,597 intensive care patients (MIMIC-III) in Boston, USA and to administrative claims data for a population-based cohort (PopHR) of 1.3 million people in Quebec, Canada. Qualitatively, we demonstrate that MixEHR-G learns interpretable phenotypes and yields meaningful insights about phenotype similarities, comorbidities, and epidemiological associations. Quantitatively, MixEHR-G outperforms existing unsupervised phenotyping methods on a phenotype label annotation task, and it can accurately estimate relative phenotype prevalence functions without gold-standard phenotype information. Altogether, MixEHR-G is an important step towards building an interpretable and automated phenotyping system using EHR data.

ObjectiveWhile there exist numerous methods to predict binary phenotypes using electronic health record (EHR) data, few exist for prediction of phenotype event times, or equivalently phenotype state progression. Estimating such quantities could enable more powerful use of EHR data for temporal analyses such as survival and disease progression. We propose Semi-supervised Adaptive Markov Gaussian Embedding Process (SAMGEP), a semi-supervised machine learning algorithm to predict phenotype event times using EHR data.MethodsSAMGEP broadly consists of four steps: (i) assemble time-evolving EHR features predictive of the target phenotype event, (ii) optimize weights for combining raw features and feature embeddings into dense patient-timepoint embeddings, (iii) fit supervised and semi-supervised Markov Gaussian Process models to this embedding progression to predict marginal phenotype probabilities at each timepoint, and (iv) take a weighted average of these supervised and semi-supervised predictions. SAMGEP models latent phenotype states as a binary Markov process, conditional on which patient-timepoint embeddings are assumed to follow a Gaussian Process.ResultsSAMGEP achieves significantly improved AUCs and F1 scores relative to common machine learning approaches in both simulations and a real-world task using EHR data to predict multiple sclerosis relapse. It is particularly adept at predicting a patient’s longitudinal phenotype course, which can be used to estimate population-level cumulative probability and count process estimators. Reassuringly, it is robust to a variety of generative model parameters.DiscussionSAMGEP’s event time predictions can be used to estimate accurate phenotype progression curves for use in downstream temporal analyses, such as a survival study for comparative effectiveness research.

Leveraging large-scale electronic health record (EHR) data to estimate survival curves for clinical events can enable more powerful risk estimation and comparative effectiveness research. However, use of EHR data is hindered by a lack of direct event times observations. Occurrence times of relevant diagnostic codes or target disease mentions in clinical notes are at best a good approximation of the true disease onset time. On the other hand, extracting precise information on the exact event time requires laborious manual chart review and is sometimes altogether infeasible due to a lack of detailed documentation. Current status labels – binary indicators of phenotype status during follow up – are significantly more efficient and feasible to compile, enabling more precise survival curve estimation given limited resources. Existing survival analysis methods using current status labels focus almost entirely on supervised estimation, and naive incorporation of unlabeled data into these methods may lead to biased results. In this paper we propose Semi-supervised Calibration of Risk with Noisy Event Times (SCORNET), which yields a consistent and efficient survival curve estimator by leveraging a small size of current status labels and a large size of imperfect surrogate features. In addition to providing theoretical justification of SCORNET, we demonstrate in both simulation and real-world EHR settings that SCORNET achieves efficiency akin to the parametric Weibull regression model, while also exhibiting non-parametric flexibility and relatively low empirical bias in a variety of generative settings.

Summary Leveraging large-scale electronic health record (EHR) data to estimate survival curves for clinical events can enable more powerful risk estimation and comparative effectiveness research. However, use of EHR data is hindered by a lack of direct event time observations. Occurrence times of relevant diagnostic codes or target disease mentions in clinical notes are at best a good approximation of the true disease onset time. On the other hand, extracting precise information on the exact event time requires laborious manual chart review and is sometimes altogether infeasible due to a lack of detailed documentation. Current status labels—binary indicators of phenotype status during follow-up—are significantly more efficient and feasible to compile, enabling more precise survival curve estimation given limited resources. Existing survival analysis methods using current status labels focus almost entirely on supervised estimation, and naive incorporation of unlabeled data into these methods may lead to biased estimates. In this article, we propose Semisupervised Calibration of Risk with Noisy Event Times (SCORNET), which yields a consistent and efficient survival function estimator by leveraging a small set of current status labels and a large set of informative features. In addition to providing theoretical justification of SCORNET, we demonstrate in both simulation and real-world EHR settings that SCORNET achieves efficiency akin to the parametric Weibull regression model, while also exhibiting semi-nonparametric flexibility and relatively low empirical bias in a variety of generative settings.

Objective A major bottleneck hindering utilization of electronic health record data for translational research is the lack of precise phenotype labels. Chart review as well as rule-based and supervised phenotyping approaches require laborious expert input, hampering applicability to studies that require many phenotypes to be defined and labeled de novo. Though International Classification of Diseases codes are often used as surrogates for true labels in this setting, these sometimes suffer from poor specificity. We propose a fully automated topic modeling algorithm to simultaneously annotate multiple phenotypes. Materials and Methods Surrogate-guided ensemble latent Dirichlet allocation (sureLDA) is a label-free multidimensional phenotyping method. It first uses the PheNorm algorithm to initialize probabilities based on 2 surrogate features for each target phenotype, and then leverages these probabilities to constrain the LDA topic model to generate phenotype-specific topics. Finally, it combines phenotype-feature counts with surrogates via clustering ensemble to yield final phenotype probabilities. Results sureLDA achieves reliably high accuracy and precision across a range of simulated and real-world phenotypes. Its performance is robust to phenotype prevalence and relative informativeness of surogate vs nonsurrogate features. It also exhibits powerful feature selection properties. Discussion sureLDA combines attractive properties of PheNorm and LDA to achieve high accuracy and precision robust to diverse phenotype characteristics. It offers particular improvement for phenotypes insufficiently captured by a few surrogate features. Moreover, sureLDA’s feature selection ability enables it to handle high feature dimensions and produce interpretable computational phenotypes. Conclusions sureLDA is well suited toward large-scale electronic health record phenotyping for highly multiphenotype applications such as phenome-wide association studies .

Objective Accurate coding is critical for medical billing and electronic medical record (EMR)-based research. Recent research has been focused on developing supervised methods to automatically assign International Classification of Diseases (ICD) codes from clinical notes. However, supervised approaches rely on ICD code data stored in the hospital EMR system and is subject to bias rising from the practice and coding behavior. Consequently, portability of trained supervised algorithms to external EMR systems may suffer. Method We developed an unsupervised knowledge integration (UNITE) algorithm to automatically assign ICD codes for a specific disease by analyzing clinical narrative notes via semantic relevance assessment. The algorithm was validated using coded ICD data for 6 diseases from Partners HealthCare (PHS) Biobank and Medical Information Mart for Intensive Care (MIMIC-III). We compared the performance of UNITE against penalized logistic regression (LR), topic modeling, and neural network models within each EMR system. We additionally evaluated the portability of UNITE by training at PHS Biobank and validating at MIMIC-III, and vice versa. Results UNITE achieved an averaged AUC of 0.91 at PHS and 0.92 at MIMIC over 6 diseases, comparable to LR and MLP. It had substantially better performance than topic models. In regards to portability, the performance of UNITE was consistent across different EMR systems, superior to LR, topic models and neural network models. Conclusion UNITE accurately assigns ICD code in EMR without requiring human labor, and has major advantages over commonly used machine learning approaches. In addition, the UNITE attained stable performance and high portability across EMRs in different institutions.

Background: Individuals with an interleukin 6 receptor (IL6R) genetic variant not on IL6R blocking therapy have biomarker profiles similar to those treated with IL6R blockers. Thus, studying whether the IL6R variant is protective for a phenotype can inform which diseases may benefit from treatment with IL6R blockade. To test this hypothesis, we performed a Phenome-Wide Association Study (PheWAS) to screen for associations between an IL6R variant and a broad range of phenotypes in the electronic health records (EHR). Methods: We studied veteran participants in the Veteran’s Affairs Million Veteran’s Project using genomic data linked to EHR. We extracted all diagnoses codes and mapped them to phenotype groups using published PheWAS methods. Routine laboratory measurements, e.g. liver function tests, were also extracted. A PheWAS was performed by constructing logistic regression models testing associations between the IL6R variant (Asp358Ala, rs2228145) and 1,866 phenotype groups; linear regression models were constructed to screen for associations between IL6R and 26 routine laboratory measurements. All models were adjusted for age, gender, and race. Significance was reported using false discovery rate ≤0.05 and Bonferroni correction. Results: We studied 342,529 participants; the minor allele frequency of the IL6R variant was 35.3%. IL6R was most strongly associated with a reduced risk of aortic aneurysm (OR 0.91-0.92, 95% CI 0.89, 0.94) (Figure 1). We observed the expected association between IL6R and reduced C-reactive protein. We also observed known side effects of IL6R blockade, elevated transaminases, as well as elevated triglycerides, an initially unexpected result in the early clinical trials. Conclusion: In this proof of concept study, we demonstrate the utility of PheWAS to inform drug effects using the largest US-based biobank study. The strong association with aortic aneurysm corresponded with the newest indication for IL6R blockade to prevent aortic aneurysms due to large vessel vasculitis.

Objective With its increasingly widespread adoption, electronic health records (EHR) have enabled phenotypic information extraction at an unprecedented granularity and scale. However, often a medical concept (e.g. diagnosis, prescription, symptom) is described in various synonyms across different EHR systems, hindering data integration for signal enhancement and complicating dimensionality reduction for knowledge discovery. Despite existing ontologies and hierarchies, tremendous human effort is needed for curation and maintenance – a process that is both unscalable and susceptible to subjective biases. This paper aims to develop a data-driven approach to automate grouping medical terms into clinically relevant concepts by combining multiple up-to-date data sources in an unbiased manner. Methods We present a novel data-driven grouping approach – multi-view banded spectral clustering (mvBSC) combining summary data from multiple healthcare systems. The proposed method consists of a banding step that leverages the prior knowledge from the existing coding hierarchy, and a combining step that performs spectral clustering on an optimally weighted matrix. Results We apply the proposed method to group ICD-9 and ICD-10-CM codes together by integrating data from two healthcare systems. We show grouping results and hierarchies for 13 representative disease categories. Individual grouping qualities were evaluated using normalized mutual information, adjusted Rand index, and F1-measure, and were found to consistently exhibit great similarity to the existing manual grouping counterpart. The resulting ICD groupings also enjoy comparable interpretability and are well aligned with the current ICD hierarchy. Conclusion The proposed approach, by systematically leveraging multiple data sources, is able to overcome bias while maximizing consensus to achieve generalizability. It has the advantage of being efficient, scalable, and adaptive to the evolving human knowledge reflected in the data, showing a significant step toward automating medical knowledge integration.

Over the past decade, there has been a marked increase in the number of primary and revision total hip and knee arthroplasties performed in the United States. Acute kidney injury (AKI) is a common complication of these procedures; however, little is known about its epidemiology in the elective arthroplasty population. We conducted a study to determine the incidence, risk factors, and outcomes of AKI after elective joint arthroplasty. Drawing on the Nationwide Inpatient Sample database, we found that the proportion of hospitalizations complicated by AKI increased rapidly from 0.5% in 2002 to 1.8% to 1.9% in 2012. Multivariate analysis revealed that the key risk factors for AKI were chronic kidney disease and the postoperative events of sepsis, acute myocardial infarction, and blood transfusion. Moreover, codiagnosis with chronic kidney disease increased the risk for AKI associated with all 3 postoperative events. After adjusting for confounders, we found an association between AKI and a significantly increased risk for in-hospital mortality and discharge to long-term facilities. AKI serves as an important quality indicator in elective hip and knee surgeries. With elective arthroplasties expected to rise, carefully planned approach to interdisciplinary perioperative care is essential to reduce both the risk and consequences of AKI.

Erythrodontia is the hallmark of human congenital erythropoietic porphyria (CEP), but is also a major phenotypic feature of acute intermittent porphyria (AIP) in cats. In this study, detailed biochemical and molecular analyses were performed on two unrelated cats with autosomal dominant AIP that presented with erythrodontia, yellow-brown urine and mild changes in erythrocytes. The cats had elevated concentrations of urinary 5-aminolevulinic acid and porphobilinogen, and half normal erythrocytic hydroxymethylbilane synthase (HMBS) activity. Two novel HMBS mutations were detected; one cat had a deletion (c.107_110delACAG) and one cat had a splicing alteration (c.826-1G>A), both leading to premature stop codons and truncated proteins (p.D36Vfs*6 and p.L276Efs*6, respectively). These studies highlight the importance of appropriate biochemical and molecular genetic analyses for the accurate diagnoses of porphyrias in cats and extend the molecular genetic heterogeneity of feline AIP. Thus, although erythrodontia is a classic sign of congenital erythropoietic porphyria in human beings, cats with erythrodontia may have acute intermittent porphyria, a hepatic porphyria.