Oncology Cohort

Release Version: May 2026 (IRB76049)

Population Summary

Patients are included in the Oncology-OMOP dataset if they have a case in the Neural Frame Stanford Cancer Registry (regardless of case status) or if they have a tumor board encounter in Epic Clarity.

As of the May 2026 OMOP population, 228,102 patients met the criteria for cancer cohort.

  • Neural Frame: Defined as research-eligible patients with case records in Neural Frame (aka Stanford Cancer Registry).
  • Tumor Board encounter: These encounters are defined using the EPIC patient encounter data tables.
  • Philips ISPM: Includes genomic testing information from the Philips IntelliSpace Precision Medicine (ISPM) genomics database at Stanford.
  • OMOP Image Occurrence: The OMOP image occurrence table captures records of medical imaging events at the series level providing a path to deidentified DICOM images.
  • OMOP Variant Occurrence: The OMOP variant occurrence table contains records of genomic variants identified via the STANFORD ACTIONABLE MUTATION PANEL FOR SOLID TUMORS (STAMP) test, sourced from Epic Genomics Suite (August 2025 onwards).
  • OMOP Whole Slide Imaging: The OMOP whole slide imaging table contains records of digitized pathology slides and currently, the table contains manually scanned slides that have been deidentified and converted into TIFF and JSON file formats.
  • OMOP Episode Table: The OMOP episode table captures records of clinical episodes, which are defined as clusters of healthcare events related to a specific condition or treatment course. Episode table is currently populated only with episodes representing treatment plans for patients in the oncology cohort.
  • Data Coverage Trends: Longitudinal view of data evolution across all platforms from February 2025 to May 2026.
Data Source Number of Patients Percent (%)
Oncology OMOP (Cohort) 228,102 100
• With Tumor Board Encounter 27,442 12
• With Image Occurrence 202,098 89
Neural Frame 208,991 92
Philips ISPM 14,783 6

Overlap between Data Sources

The following plot represents patients overlap between different data sources:

  • The connected dots in the middle show which combinations of data sources exist

  • The bars show how many patients are in each combination

For example, if dots are connected for “Oncology OMOP” and “Neural Frame”, the bar above shows how many patients are in both sources.

Patient Distribution Across Data Sources

Source Codes:

The source codes for this page can be found here, and the sql queries that support the metrics can be found here.