JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

Identifying patients with neurofibromatosis type 1 related optic pathway glioma using the OMOP CDM

  • 0Department of General Paediatrics, Erasmus MC-Sophia Children's Hospital, Rotterdam, the Netherlands; The ENCORE Expertise Centre for Neurodevelopmental Disorders, Erasmus MC, Rotterdam, the Netherlands.
European Journal of Medical Genetics +

|

March 19, 2025

|

March 19, 2025

View abstract on PubMed

Summary

This summary is machine-generated.

Computable phenotype algorithms using OMOP successfully identified patients with Neurofibromatosis type 1-related optic pathway gliomas (NF1-OPG) in electronic health records. These algorithms can efficiently screen for rare NF1-OPG cases and discover new ones.

Area Of Science

  • Medical Informatics
  • Oncology
  • Genetics

Background

  • Neurofibromatosis type 1 (NF1) is a rare genetic disorder predisposing individuals to tumors, including optic pathway gliomas (OPG).
  • Efficiently identifying patients with NF1-related OPG in clinical settings is crucial for timely diagnosis and management.
  • Existing methods for patient identification may not be optimal for rare conditions like NF1-OPG.

Purpose Of The Study

  • To develop and evaluate computable phenotype algorithms for identifying patients with NF1-related OPG in electronic health records (EHR).
  • To assess the performance of these algorithms using diagnosis codes, clinical visits, and radiologic procedures.
  • To determine the utility of these algorithms in discovering previously unidentified NF1-OPG cases.

Main Methods

  • Developed computable phenotype algorithms based on the Observational Medical Outcome Partnership (OMOP) Common Data Model.
  • Applied algorithms to an EHR-derived database from an academic hospital.
  • Validated algorithm performance using precision, recall, and F2 scores against a clinician-provided list of known cases (n=61).
  • Manually reviewed algorithm-identified potential cases to find additional NF1-OPG patients.

Main Results

  • The algorithm using diagnosis codes 'Neurofibromatosis syndrome' and 'Neoplasm of optic nerve' demonstrated the highest performance (precision=1.000, recall=0.614, F2-score=0.665).
  • An algorithm combining 'Neurofibromatosis syndrome' with specific visit and imaging criteria achieved a precision of 0.489 and recall of 0.511.
  • Manual review of algorithm outputs identified 27 additional NF1-OPG cases not in the initial known case list.
  • A trade-off was observed between precision and recall, with higher precision generally leading to lower recall.

Conclusions

  • OMOP computable phenotype algorithms are effective tools for identifying NF1-related OPG patients within EHR databases.
  • These algorithms provide rapid insights into case counts and can uncover previously unrecognized cases.
  • The developed phenotype algorithms hold significant potential for facilitating patient screening in rare disease research, particularly for multi-centric trials.

Keywords: