Optimization of CT protocols using cause-and-effect analysis of outliers

Area of Science:

• Radiology
• Medical Physics
• Health Informatics

Background:

• Computed Tomography (CT) examinations generate significant radiation dose data.
• Optimizing CT protocols is crucial for patient safety and effective diagnostics.
• Existing methods like Dose Reference Levels assess general protocol performance.

Purpose of the Study:

• To implement and evaluate an outlier marking and analysis methodology for CT examination protocols.
• To optimize CT Head examination protocols through detailed outlier analysis.
• To identify root causes of dose metric outliers in CT examinations.

Main Methods:

• Collected CT Head examination data, including dose metrics and technical parameters, in an automatic dose registry.
• Calculated outlier thresholds using the medcouple parameter to account for distribution skewness.
• Performed Cause-and-Effect analysis on marked outliers from a 4-month period with a multidisciplinary committee.

Main Results:

• Analyzed dose distributions from 3690 CT Head examinations, noting skewness in CTDIvol and DLP.
• Identified inadequate protocol selection, foreign objects in the field-of-view, and abnormal scan region diameter as primary causes of upper outliers.
• Determined that 90% of lower outliers were due to the inclusion of extra series in the head protocol.

Conclusions:

• The proposed outlier analysis method offers a significant advancement over traditional Dose Reference Level methods for CT protocol optimization.
• This methodology effectively detects incorrect CT utilization stemming from inadequate technical knowledge.
• The findings provide a basis for refining CT protocols and improving the quality of examinations.