Predictive Modeling of Hospital Readmission Rates Using Electronic Medical Record-Wide Machine Learning: A Case-Study Using Mount Sinai Heart Failure Cohort.

Abstract Reduction of preventable hospital readmissions that result from chronic or acute conditions like stroke, heart failure, myocardial infarction and pneumonia remains a significant challenge for improving the outcomes and decreasing the cost of healthcare delivery in the United States. Patient readmission rates are relatively high for conditions like heart failure (HF) despite the implementation of high-quality healthcare delivery operation guidelines created by regulatory authorities. Multiple predictive models are currently available to evaluate potential 30-day readmission rates of patients. Most of these models are hypothesis driven and repetitively assess the predictive abilities of the same set of biomarkers as predictive features. In this manuscript, we discuss our attempt to develop a data-driven, electronic-medical record-wide (EMR-wide) feature selection approach and subsequent machine learning to predict readmission probabilities. We have assessed a large repertoire of variables from electronic medical records of heart failure patients in a single center. The cohort included 1,068 patients with 178 patients were readmitted within a 30-day interval (16.66% readmission rate). A total of 4,205 variables were extracted from EMR including diagnosis codes (n=1,763), medications (n=1,028), laboratory measurements (n=846), surgical procedures (n=564) and vital signs (n=4). We designed a multistep modeling strategy using the Naïve Bayes algorithm. In the first step, we created individual models to classify the cases (readmitted) and controls (non-readmitted). In the second step, features contributing to predictive risk from independent models were combined into a composite model using a correlation-based feature selection (CFS) method. All models were trained and tested using a 5-fold cross-validation method, with 70% of the cohort used for training and the remaining 30% for testing. Compared to existing predictive models for HF readmission rates (AUCs in the range of 0.6-0.7), results from our EMR-wide predictive model (AUC=0.78; Accuracy=83.19%) and phenome-wide feature selection strategies are encouraging and reveal the utility of such datadriven machine learning. Fine tuning of the model, replication using multi-center cohorts and prospective clinical trial to evaluate the clinical utility would help the adoption of the model as a clinical decision system for evaluating readmission status. Full PDF...

Read More

Exploring The Reproducibility of Probabilistic Causal Molecular Network Models.

Abstract Network reconstruction algorithms are increasingly being employed in biomedical and life sciences research to integrate large-scale, high-dimensional data informing on living systems. One particular class of probabilistic causal networks being applied to model the complexity and causal structure of biological data is Bayesian networks (BNs). BNs provide an elegant mathematical framework for not only inferring causal relationships among many different molecular and higher order phenotypes, but also for incorporating highly diverse priors that provide an efficient path for incorporating existing knowledge. While significant methodological developments have broadly enabled the application of BNs to generate and validate meaningful biological hypotheses, the reproducibility of BNs in this context has not been systematically explored. In this study, we aim to determine the criteria for generating reproducible BNs in the context of transcription-based regulatory networks. We utilize two unique tissues from independent datasets, whole blood from the GTEx Consortium and liver from the Stockholm-Tartu Atherosclerosis Reverse Network Engineering Team (STARNET) study. We evaluated the reproducibility of the BNs by creating networks on data subsampled at different levels from each cohort and comparing these networks to the BNs constructed using the complete data. To help validate our results, we used simulated networks at varying sample sizes. Our study indicates that reproducibility of BNs in biological research is an issue worthy of further consideration, especially in light of the many publications that now employ findings from such constructs without appropriate attention paid to reproducibility. We find that while edge-to-edge reproducibility is strongly dependent on sample size, identification of more highly connected key driver nodes in BNs can be carried out with high confidence across a range of sample sizes. Full PDF...

Read More

Machine-Learning Algorithms to Automate Morphological and Functional Assessments in 2D Echocardiography.

Abstract   BACKGROUND: Machine-learning models may aid cardiac phenotypic recognition by using features of cardiac tissue deformation. OBJECTIVES: This study investigated the diagnostic value of a machine-learning framework that incorporates speckle-tracking echocardiographic data for automated discrimination of hypertrophic cardiomyopathy (HCM) from physiological hypertrophy seen in athletes (ATH). METHODS: Expert-annotated speckle-tracking echocardiographic datasets obtained from 77 ATH and 62 HCM patients were used for developing an automated system. An ensemble machine-learning model with 3 different machine-learning algorithms (support vector machines, random forests, and artificial neural networks) was developed and a majority voting method was used for conclusive predictions with further K-fold cross-validation. RESULTS: Feature selection using an information gain (IG) algorithm revealed that volume was the best predictor for differentiating between HCM ands. ATH (IG = 0.24) followed by mid-left ventricular segmental (IG = 0.134) and average longitudinal strain (IG = 0.131). The ensemble machine-learning model showed increased sensitivity and specificity compared with early-to-late diastolic transmitral velocity ratio (p < 0.01), average early diastolic tissue velocity (e’) (p < 0.01), and strain (p = 0.04). Because ATH were younger, adjusted analysis was undertaken in younger HCM patients and compared with ATH with left ventricular wall thickness >13 mm. In this subgroup analysis, the automated model continued to show equal sensitivity, but increased specificity relative to early-to-late diastolic transmitral velocity ratio, e’, and strain. CONCLUSIONS: Our results suggested that machine-learning algorithms can assist in the discrimination of physiological versus pathological patterns of hypertrophic remodeling. This effort represents a step toward the development of a real-time, machine-learning-based system for automated interpretation of echocardiographic images, which may help novice readers with limited...

Read More

Transcriptional Profiling of Egg Allergy and Relationship to Disease Phenotype.

Abstract BACKGROUND: Egg allergy is one of the most common food allergies of childhood. There is a lack of information on the immunologic basis of egg allergy beyond the role of IgE. OBJECTIVE: To use transcriptional profiling as a novel approach to uncover immunologic processes associated with different phenotypes of egg allergy. METHODS: Peripheral blood mononuclear cells (PBMCs) were obtained from egg-allergic children who were defined as reactive (BER) or tolerant (BET) to baked egg, and from food allergic controls (AC) who were egg non-allergic. PBMCs were stimulated with egg white protein. Gene transcription was measured by microarray after 24 h, and cytokine secretion by multiplex assay after 5 days. RESULTS: The transcriptional response of PBMCs to egg protein differed between BER and BET versus AC subjects. Compared to the AC group, the BER group displayed increased expression of genes associated with allergic inflammation as well as corresponding increased secretion of IL-5, IL-9 and TNF-α. A similar pattern was observed for the BET group. Further similarities in gene expression patterns between BER and BET groups, as well as some important differences, were revealed using a novel Immune Annotation resource developed for this project. This approach identified several novel processes not previously associated with egg allergy, including positive associations with TLR4-stimulated myeloid cells and activated NK cells, and negative associations with an induced Treg signature. Further pathway analysis of differentially expressed genes comparing BER to BET subjects showed significant enrichment of IFN-α and IFN-γ response genes, as well as genes associated with virally-infected DCs. CONCLUSIONS: Transcriptional profiling identified several novel pathways and processes that differed when comparing the response to egg allergen in BET, BER, and AC groups. We conclude that this approach is a useful hypothesis-generating mechanism to identify novel immune processes associated with allergy and tolerance to forms of egg. Full PDF...

Read More

e-GRASP: an integrated evolutionary and GRASP resource for exploring disease associations

Abstract BACKGROUND: Genome-wide association studies (GWAS) have become a mainstay of biological research concerned with discovering genetic variation linked to phenotypic traits and diseases. Both discrete and continuous traits can be analyzed in GWAS to discover associations between single nucleotide polymorphisms (SNPs) and traits of interest. Associations are typically determined by estimating the significance of the statistical relationship between genetic loci and the given trait. However, the prioritization of bona fide, reproducible genetic associations from GWAS results remains a central challenge in identifying genomic loci underlying common complex diseases. Evolutionary-aware meta-analysis of the growing GWAS literature is one way to address this challenge and to advance from association to causation in the discovery of genotype-phenotype relationships. DESCRIPTION: We have created an evolutionary GWAS resource to enable in-depth query and exploration of published GWAS results. This resource uses the publically available GWAS results annotated in the GRASP2 database. The GRASP2 database includes results from 2082 studies, 177 broad phenotype categories, and ~8.87 million SNP-phenotype associations. For each SNP in e-GRASP, we present information from the GRASP2 database for convenience as well as evolutionary information (e.g., rate and timespan). Users can, therefore, identify not only SNPs with highly significant phenotype-association P-values, but also SNPs that are highly replicated and/or occur at evolutionarily conserved sites that are likely to be functionally important. Additionally, we provide an evolutionary-adjusted SNP association ranking (E-rank) that uses cross-species evolutionary conservation scores and population allele frequencies to transform P-values in an effort to enhance the discovery of SNPs with a greater probability of biologically meaningful disease associations. CONCLUSION: By adding an evolutionary dimension to the GWAS results available in the GRASP2 database, our e-GRASP resource will enable a more effective exploration of SNPs not only by the statistical significance of trait associations, but also by the number of studies in which associations have been replicated, and the evolutionary context of the associated mutations. Therefore, e-GRASP will be a valuable resource for aiding researchers in the identification of bona fide, reproducible genetic associations from GWAS results. This resource is freely available at http://www.mypeg.info/egrasp . Full PDF...

Read More