An explainable artificial intelligence approach for predicting cardiovascular outcomes using electronic health records.

Understanding the conditionally-dependent clinical variables that drive cardiovascular health outcomes is a major challenge for precision medicine. Here, we deploy a recently developed massively scalable comorbidity discovery method called Poisson Binomial based Comorbidity discovery (PBC), to analyze Electronic Health Records (EHRs) from the University of Utah and Primary Children's Hospital (over 1.6 million patients and 77 million visits) for comorbid diagnoses, procedures, and medications. Using explainable Artificial Intelligence (AI) methodologies, we then tease apart the intertwined, conditionally-dependent impacts of comorbid conditions and demography upon cardiovascular health, focusing on the key areas of heart transplant, sinoatrial node dysfunction and various forms of congenital heart disease. The resulting multimorbidity networks make possible wide-ranging explorations of the comorbid and demographic landscapes surrounding these cardiovascular outcomes, and can be distributed as web-based tools for further community-based outcomes research. The ability to transform enormous collections of EHRs into compact, portable tools devoid of Protected Health Information solves many of the legal, technological, and data-scientific challenges associated with large-scale EHR analyses.

A Poisson binomial-based statistical testing framework for comorbidity discovery across electronic health record datasets.

Discovering the concomitant occurrence of distinct medical conditions in a patient, also known as comorbidities, is a prerequisite for creating patient outcome prediction tools. Current comorbidity discovery applications are designed for small datasets and use stratification to control for confounding variables such as age, sex or ancestry. Stratification lowers false positive rates, but reduces power, as the size of the study cohort is decreased. Here we describe a Poisson binomial-based approach to comorbidity discovery (PBC) designed for big-data applications that circumvents the need for stratification. PBC adjusts for confounding demographic variables on a per-patient basis and models temporal relationships. We benchmark PBC using two datasets to compute comorbidity statistics on 4,623,841 pairs of potentially comorbid medical terms. The results of this computation are provided as a searchable web resource. Compared with current methods, the PBC approach reduces false positive associations while retaining statistical power to discover true comorbidities.

Implementing the VMC Specification to Reduce Ambiguity in Genomic Variant Representation.

Current methods used for representing biological sequence variants allow flexibility, which has created redundancy within variant archives and discordance among variant representation tools. While research methodologies have been able to adapt to this ambiguity, strict clinical standards make it difficult to use this data in what would otherwise be useful clinical interventions. We implemented a specification developed by the GA4GH Variant Modeling Collaboration (VMC), which details a new approach to unambiguous representation of variants at the allelic level, as a haplotype, or as a genotype. Our implementation, called the VMC Test Suite (http://vcfclin.org), offers web tools to generate and insert VMC identifiers into a VCF file and to generate a VMC bundle JSON representation of a VCF file or HGVS expression. A command line tool with similar functionality is also introduced. These tools facilitate use of this standard-an important step toward reliable querying of variants and their associated annotations.

ClinVar Miner: Demonstrating utility of a Web-based tool for viewing and filtering ClinVar data.

ClinVar Miner is a Web-based suite that utilizes the data held in the National Center for Biotechnology Information's ClinVar archive. The goal is to render the data more accessible to processes pertaining to conflict resolution of variant interpretation as well as tracking details of data submission and data management for detailed variant curation. Here, we establish the use of these tools to address three separate use cases and to perform analyses across submissions. We demonstrate that the ClinVar Miner tools are an effective means to browse and consolidate data for variant submitters, curation groups, and general oversight. These tools are also relevant to the variant interpretation community in general.