Evaluating data capture methods for the establishment of diagnostic reference levels in CT scanning

Area of Science:

• Medical Imaging
• Radiological Physics
• Health Informatics

Background:

• Diagnostic reference levels for CT scanning are crucial for radiation dose management.
• Self-complete surveys are commonly used but may lack accuracy.
• Radiological Information System/Picture Archive Communication Systems (RIS/PACS) offer an alternative data source.

Purpose of the Study:

• To compare radiation dosimetry derived from self-complete surveys with data from RIS/PACS.
• To evaluate the accuracy and reliability of survey data for CT radiation dose assessment.

Main Methods:

• Data collected from adult CT examinations at a tertiary hospital.
• Radiation dose calculated from both survey and RIS/PACS datasets.
• Inter-rater agreement evaluated for both data collection methods.

Main Results:

• Significant systematic and proportional differences found between survey and RIS/PACS dosimetry.
• Discrepancies varied across CT examination types, not allowing simple correction.
• Organ dose showed greater variability than effective dose due to reliance on generic limits in surveys.

Conclusions:

• Caution is advised when interpreting radiation dose studies using self-complete surveys due to observed bias.
• Electronic databases like RIS/PACS provide higher quality, cost-effective data for CT radiation dose monitoring.
• Establishment of national and local databases is recommended for routine data capture and dose management.