How many patients should be studied? This is one of the most commonly asked questions in designing a research study. New imaging modalities are being developed to answer clinical questions with lower radiation doses and decreased scanner time. Although there are many articles and textbooks written about sample size selection and statistical power, few of these articles describe sample size selection to evaluate safer, faster technologies with equivalent accuracy. The purpose of this exhibit is to review the basic principles underlying sample size selection and use those principles to describe the process of sample size selection for common study designs that evaluate new imaging technologies.
1. Basic Principles of Sample Size Calculation
Type I Error
Type II Error
Statistical Power
Types of Variables
2. Case Studies
Non-inferiority of Diagnostic Tests (Is my new imaging test/protocol at least as good as the old one?)
Diagnostic Accuracy Studies (Is my new imaging test better than the old one?)
Imaging Biomarkers as Disease Predictors (Does my imaging test predict disease outcomes?)
3. Resources
Web based
Articles
Narayan, A,
Eng, J,
Sample Size Selection to Evaluate Emerging Imaging Technologies: A Reader's Guide. Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL.
http://archive.rsna.org/2014/14015772.html
