PhEMA: Phenotype Execution and Modeling Architecture

PhEMA: Phenotype Execution and Modeling
Architecture
NIGMS: R01 GM105688
projectphema.org
Jyotishman Pathak – P.I. (Mayo Clinic)
Joshua Denny – P.I. (Vanderbilt University)
William Thompson – P.I. (NorthShore University HealthSystem)
Enid Montague – Human Factors (Site P.I. at Northwestern)
Luke Rasmussen – Architect
Huan Mo – Execution Engine Development
Guoqian Jiang – Standards Interoperability
April 23, 2015
EHR-Driven Phenotpying
I
Phenotyping is the process of selecting cohorts based on
clinically relevant observable features
I
The identification of patient cohorts for clinical and genomic
research is a costly and time-consuming process
I
Leveraging EHRs for identifying patient cohorts has become an
increasingly attractive option
From a research perspective, integrated data constitute a
computable collection of fine-grained longitudinal phenotypic
profiles, facilitating cohort-wide investigations and
knowledge discovery on an unprecedented scale.
P. B. Jensen, L. J. Jensen, and Brunak (2012) “Mining electronic health records”
Current Practice of EHR-Driven Phenotpying
I
A key component for identifying patient cohorts in the EHR is to
define inclusion and exclusion criteria that select sets of patients
based on stored clinical data
I
Phenotyping logic can be quite complex, and typically includes
both Boolean and temporal operators applied to multiple clinical
events.
I
The phenotyping algorithm development process is a
multi-disciplinary team effort, including clinicians, domain experts,
and informaticians
I
The typical way to share phenotyping algorithms across
institutions is through the use of informal free text descriptions of
algorithm logic, possibly augmented with graphical flowcharts and
simple lists of structured codes.
I
Algorithms are operationalized as database queries and
software, customized to the local EHR environment.
Electronic Medical Records and Genomics Network
Gottesman, Kuivaniemi, et al. (2013) “The Electronic Medical Records and Genomics
(eMERGE) Network”
I NHGRI funded project in 8th
year (2nd funding cycle; 3rd
starting soon)
I Development of methods and
best practices for using the EHR
as a tool for genomic research
I Network members link EHR data
to DNA biobanks, pool results
across sites
I Several dozen electronic
phenotypes completed and in
progress (www.PheKB.org)
Example Phenotypes: Type 2 Diabetes,
Peripheral Arterial Disease, QRS Duration,
Colon Polyps, Hypothyroidism
Example Algorithm: Hypothyroidism
Malinowski, Denny, et al. (2014) “Genetic variants associated with serum thyroid
stimulating hormone (TSH) levels in European Americans and African Americans
from the eMERGE Network”
Phenotyping Workflow
Phenotyping Workflow
Thompson, Rasmussen, et al. (2012) “An evaluation of the NQF Quality Data Model
for representing Electronic Health Record driven phenotyping algorithms”
Key Lessons Learned from eMERGE
I
Algorithm design and transportability
I
I
I
I
I
Non-trivial; requires significant expert involvement
Highly iterative process
Time-consuming manual chart reviews
Representation of “phenotype logic” is critical
Standardized data access and representation
I
I
I
Importance of unified vocabularies, data elements, and value sets
Questionable reliability of ICD & CPT codes (e.g., billing the wrong
code since it is easier to find)
Natural Language Processing (NLP) is critical
Kho, J. A. Pacheco, et al. (2011) “Electronic Medical Records for Genetic Research”
Target Phenotyping Workflow
Project Goal and Aims
Goal:
The proposed project will design, build and promote an open-access
community infrastructure for standards-based development and
sharing of phenotyping algorithms, as well as provide tools and
resources for investigators, researchers and their informatics support
staff to implement and execute the algorithms on native EHR data.
Aims:
I
To create a standards-based information model for representing
phenotyping algorithms.
I
To create an open-access repository and infrastructure for
authoring, sharing and accessing computable, standardized
phenotyping algorithms.
I
To develop informatics methods and tools for translating
phenotyping algorithmic criteria into EHR-based executable
queries.
Phenotype KnowledgeBase (PheKB)
Phenotype Authoring Tool
Phenotype Execution Environment – JBoss Drools
Li, Endle, et al. (2012) “Modeling and executing electronic health records driven
phenotyping algorithms using the NQF Quality Data Model and JBoss Drools Engine”
Phenotype Execution Environment – KNIME
Mo, J. Pacheco, et al. (2015) “A prototype for executable and portable electronic
clinical quality measures using the KNIME analytics platform”
References I
Omri Gottesman et al. “The Electronic Medical Records and
Genomics (eMERGE) Network: past, present, and future”. In:
Genetics in medicine: official journal of the American College of
Medical Genetics (June 6, 2013). ISSN: 1530-0366. DOI:
10.1038/gim.2013.72.
Peter B. Jensen, Lars J. Jensen, and Søren Brunak. “Mining
electronic health records: towards better research applications
and clinical care”. In: Nature Reviews. Genetics 13.6 (June
2012), pp. 395–405. ISSN: 1471-0064. DOI:
10.1038/nrg3208.
References II
A. N. Kho et al. “Electronic Medical Records for Genetic
Research: Results of the eMERGE Consortium”. In: Science
Translational Medicine 3.79 (Apr. 20, 2011), 79re1–79re1. ISSN:
1946-6234, 1946-6242. DOI:
10.1126/scitranslmed.3001807. URL:
http://stm.sciencemag.org/cgi/doi/10.1126/
scitranslmed.3001807 (visited on 07/18/2014).
Dingcheng Li et al. “Modeling and executing electronic health
records driven phenotyping algorithms using the NQF Quality
Data Model and JBoss Drools Engine”. In: AMIA ... Annual
Symposium proceedings / AMIA Symposium. AMIA Symposium
2012 (2012), pp. 532–541. ISSN: 1942-597X.
References III
Jennifer R. Malinowski et al. “Genetic variants associated with
serum thyroid stimulating hormone (TSH) levels in European
Americans and African Americans from the eMERGE Network”.
In: PloS One 9.12 (2014), e111301. ISSN: 1932-6203. DOI:
10.1371/journal.pone.0111301.
Huan Mo et al. “A prototype for executable and portable
electronic clinical quality measures using the KNIME analytics
platform”. In: AMIA Summits Transl Sci Proc. 2015.
William K. Thompson et al. “An evaluation of the NQF Quality
Data Model for representing Electronic Health Record driven
phenotyping algorithms”. In: AMIA Annual Symposium
Proceedings. Vol. 2012. 2012, p. 911. URL: http://www.
ncbi.nlm.nih.gov/pmc/articles/PMC3540514/
(visited on 08/29/2013).