Toward Data-Driven Radiology Education-Early Experience Building Multi-Institutional Academic Trainee Interpretation...

Area of Science:

• Medical Informatics
• Radiology Education
• Big Data Analytics

Background:

• The Accreditation Council for Graduate Medical Education (ACGME) collects resident exam volume data to set minimum requirements.
• ACGME data and methodologies are not publicly available, hindering assessment of data integrity and relevance to resident experience.

Purpose of the Study:

• To describe the experience of creating a multi-institutional case log system for diagnostic radiology residency programs.
• To aggregate resident case log data from three institutions into a centralized database.

Main Methods:

• Established automated query pipelines from radiology information systems at three institutions to create resident-specific databases.
• Aggregated three institutional resident case log databases into a single, centralized database schema.
• Catalogued 330 residents and 2,905,923 radiologic examinations over a 4-year period using 11 ACGME categories.

Main Results:

• Successfully created a centralized database aggregating data from three diagnostic radiology residency programs.
• Catalogued a large volume of radiologic examinations (2,905,923) across 330 residents over four years.
• Identified significant big data challenges, including internal data heterogeneity and external data discrepancies.

Conclusions:

• Multi-institutional case log systems can be developed to enhance data transparency in graduate medical education.
• Automated data pipelines and centralized aggregation are feasible for large-scale resident data collection.
• Addressing data heterogeneity and discrepancies is crucial for informatics researchers in medical education.