Size-based quality-informed framework for quantitative optimization of pediatric CT.

Area of Science:

• Medical Imaging
• Radiology
• Pediatric Imaging

Background:

• Optimizing computed tomography (CT) protocols for pediatric populations is crucial for balancing diagnostic accuracy and radiation dose.
• Existing CT protocols may not provide consistent performance across diverse pediatric age and size groups.
• Quantitative methods are needed to systematically relate diagnostic performance to radiation dose in pediatric CT.

Purpose of the Study:

• To develop a systematic, evidence-based method for optimizing CT imaging in pediatric populations.
• To establish a multidimensional system that relates quantitative diagnostic performance to radiation dose.
• To enable consistent diagnostic performance across a wide range of pediatric body sizes.

Main Methods:

• Assessed organ doses, effective dose (E), and risk index for 30 pediatric patients across nine age-size groups.
• Utilized simulated lesions and added noise to assess nodule detection accuracy via observer receiving operating characteristic (ROC) studies.
• Developed continuous accuracy-dose relationships to optimize individual scan parameters.

Main Results:

• Before optimization, CT protocols showed similar effective doses but decreasing accuracy with age (0.89 youngest to 0.67 oldest).
• After optimization, a target accuracy of 0.83 was achieved for all pediatric categories with effective doses ranging from 1 to 10 mSv.
• Alternative isogradient operating points demonstrated a consistent ratio of accuracy-per-unit-dose across patient groups.

Conclusions:

• The developed model provides a framework for optimizing individual CT scan parameters in pediatric imaging.
• The method ensures consistent diagnostic performance regardless of patient size or age.
• This approach facilitates safer and more effective CT examinations for children.