JMIR Publications

ABSTRACT

Background:

Although electronic health records (EHRs) have been widely used for secondary use of EHRs, clinical documents are relatively less utilized due to the lack of standardized clinical text frameworks across different institutions.

Objective:

To develop a framework for processing unstructured clinical documents of EHRs and integration with standardized structured data.

Methods:

We developed a framework known as Staged Optimization of Curation, Regularization, and Annotation of clinical text (SOCRATex). SOCRATex has four components: (1) clinical document extraction and preprocessing using text mining methods, (2) defining annotation schema with hierarchical structure (3) document-level annotation using the annotation schema, and (4) indexing the annotations for search engine system. To test the usability of the proposed framework, proof-of-concept studies were performed on EHRs. We defined three distinctive patient groups and extracted their clinical documents (i.e., pathology reports, radiology reports, and admission notes). The documents were annotated and integrated into the Observational Medical Outcomes Partnership (OMOP)-common data model (CDM) database. The annotations were used for performing Cox proportional hazard models with different settings of clinical analyses, measuring 1) all-cause mortality, 2) thyroid cancer recurrence, and 3) 30-day hospital readmission.

Results:

Overall, 1,055 clinical documents of 923 patients were extracted and annotated using the defined annotation schemas. The generated annotations were indexed into an unstructured textual data repository. Using the annotations of pathology reports, we identified that node metastasis and lymphovascular invasion of tumor were associated with all-cause mortality of colon and rectum cancer patients (All P <0.05). The other analyses of measuring thyroid cancer recurrence using radiology reports and 30-day hospital readmission using admission notes of depressive disorder patients were also showing consistent results with previously known knowledge.

Conclusions:

We propose a framework for hierarchical annotation of textual data and integration into the standardized OMOP-CDM medical database. The proof-of-concept studies demonstrated that our framework can effectively process and integrate diverse clinical documents with standardized structured data for clinical research.