Clinical Scorecard: Optimizing Abdominal CT Protocols Through Crowd-Sourced Analysis of 908,000 Imaging Exams in a Comprehensive Radiation Dose Database
At a Glance
Category
Detail
Condition
Routine abdomen computed tomography (CT) imaging
Key Mechanisms
Variation in radiation dose driven by local protocol decisions rather than patient or machine factors; use of crowd-sourced registry data and cluster analysis to identify low-dose optimized protocols
Hospitals and outpatient imaging facilities across multiple countries
Key Highlights
A six-fold variation in average radiation dose exists across 1033 unique abdomen CT protocols after adjusting for patient size.
Local clinical decisions on protocol parameters, including phase selection and acquisition settings, primarily drive dose variation.
Crowd-sourced analysis of a large international dose registry enables identification of optimized low-dose protocols.
Guideline-Based Recommendations
Diagnosis
Exclude abdomen CT exams for angiography, trauma, low-dose indications (e.g., suspected renal stones), and high-dose indications (e.g., liver cancer detection) when analyzing routine abdomen CT protocols.
Management
Implement standardized protocols to minimize radiation exposure while maintaining diagnostic image quality.
Utilize dose reduction technologies such as automated kV modulation and limit unindicated scan phases.
Monitoring & Follow-up
Use size-adjusted radiation dose metrics (CTDIvol and DLP) to compare and monitor radiation doses across facilities, accounting for patient size.
Regularly update and review protocols using crowd-sourced registry data to identify and adopt optimized low-dose protocols.
Risks
Excess radiation dose from non-standardized protocols increases patient cancer risk.
Inadequate protocol maintenance may lead to inconsistent image quality or unnecessary radiation exposure.
Patient & Prescribing Data
Adults aged 18–99 undergoing routine abdomen CT scans across 132 hospitals or outpatient facilities in 7 countries.
Protocols with at least 20 examinations were considered routine and sufficient for diagnosis; dose variation is mainly influenced by protocol choices rather than patient or machine characteristics.
Clinical Best Practices
Normalize radiation dose metrics for patient size to enable fair comparison across facilities.
Exclude outlier scans and rarely used machine models to ensure data quality in protocol optimization.
Leverage cluster analysis to group protocols by technical parameters and identify low-dose clusters.
Engage multidisciplinary teams including radiologists, medical physicists, and technologists in protocol development and maintenance.
Adopt a consensus-based approach to protocol standardization to reduce resource burden and dose variation.
A VHA study across 11 vendors finds AI-generated primary care notes score lower than clinician-written notes, with the largest deficits in thoroughness, organization, and usefulness
