Qualifying the use of RIS data for patient dose by comparison with DICOM header data

Area of Science:

• Medical Physics
• Radiology Informatics
• Health Informatics

Background:

• A 2008 system for patient dose audits utilized Radiology Information System (RIS) data, but suffered from user-entered data quality issues.
• Digital Imaging and Communication in Medicine (DICOM) header data has demonstrated high accuracy for dose audits.
• Existing methods highlight the need for improved data accuracy and efficiency in patient dose auditing.

Purpose of the Study:

• To evaluate the viability and advantages of using RIS data for patient dose audits compared to DICOM header data.
• To develop and implement a new system for easily extracting DICOM header data.
• To compare the consistency and utility of data fields from both RIS and DICOM sources.

Main Methods:

• Developed a new system to extract header data from DICOM images.
• Collected data for common examinations using both RIS and DICOM sources.
• Performed a result-by-result comparison of RIS and DICOM data for consistency and unique field value.

Main Results:

• RIS data, while less accurate than DICOM, provides sufficient accurate data for dose audits.
• RIS data offers distinct advantages over relying solely on DICOM data.
• Identified inconsistencies in common fields between RIS and DICOM, alongside valuable unique data fields.

Conclusions:

• RIS data presents a practical and advantageous alternative for patient dose audits.
• A hybrid approach, potentially utilizing Modality Performed Procedure Step (MPPS), may offer the optimal solution ('best of both worlds').
• Further investigation into MPPS integration is recommended for comprehensive dose auditing.