Clinical Research Informatics | Data Architecture | Integration Engineering | OMOP / Cohort Discovery

I am a Registered Nurse and health data scientist with a long background in software engineering, clinical data systems, quantitative analysis, and integration architecture.

My work sits at the intersection of clinical practice, research informatics, and software engineering. I have decades of scripting and programming experience across shell, Perl, Tcl/Tk, Python, SQL, and related technologies, along with experience in statistical modeling, machine learning, healthcare analytics, and teaching quantitative methods in higher education.

I am the original architect of a metadata-driven, provenance-preserving, TQIP-compliant trauma registry platform designed around configurable validation, governed data definitions, and hierarchical dictionary inheritance. The architecture was intentionally extensible beyond trauma, allowing the underlying platform to support additional clinical registry domains.

I have also designed and implemented integration pipelines across dozens of technical contexts. My approach emphasizes explicit data contracts, programmatic transformation, validation, automation, and reproducibility rather than dependence on any particular vendor or integration tool.

Clinical data across boundaries

I am particularly interested in the problems that arise where clinical knowledge, research requirements, terminology, and technical systems meet.

Healthcare data rarely exists in one clean form. It moves through EHRs, registries, research systems, databases, HL7 interfaces, XML documents, FHIR resources, APIs, and analytical platforms. I build systems that make those transitions explicit, testable, and reproducible.

That work requires more than technical access to clinical data. It requires respect for institutional governance, HIPAA requirements, IRB protocols, clinical definitions, security controls, and the expertise of the people who own each domain.

My role is not to replace that expertise. It is to collaborate across domains to translate clinical questions into computable definitions, terminology into data models, research protocols into cohort logic, and heterogeneous source data into structures suitable for reproducible analysis.

Current focus: oncology cohort discovery

I am currently developing a portfolio project exploring how oncology clinical-trial eligibility criteria can be translated into reproducible cohort definitions against the OMOP Common Data Model.

Automated Oncology Cohort Discovery in OMOP

The project is being built in Python to retrieve trial criteria from ClinicalTrials.gov, represent selected inclusion and exclusion criteria as structured rules, resolve oncology concepts using terminologies such as ICD-O-3, HemOnc, LOINC, and RxNorm, and execute temporal cohort logic against a synthetic OMOP database.

The goal is not to build a clinical decision-support system. It is a research-informatics reference implementation designed to demonstrate:

• terminology-aware clinical data modeling

• OMOP vocabulary mapping and cohort definition

• explicit and reproducible eligibility logic

• temporal reasoning across diagnoses, treatments, and laboratory results

• provenance and criterion-level explainability

• Python/SQL integration against relational clinical data

Project in development — coming soon.