This chapter provides a framework for establishing the validity of the mathematical models that are used in radiological assessment. Although model validation can be a contentious topic, largely because different groups of scientists have attached different meanings to the term, a useful working definition of model validation for radiological assessment is provided. The importance of specifying the purpose of the model before its validity can be assessed is stressed. The process of model validation, which is divided into two parts (an analysis of the model's composition and an analysis of the model's performance), is described in detail. Quantitative measures of model performance based on bias, scatter, correlation, and regression are presented. The difference between model calibration and model validation is described and examples given. The final part of the chapter discusses the different reasons for poor model performance.
Oxford Scholarship Online requires a subscription or purchase to access the full text of books within the service. Public users can however freely search the site and view the abstracts and keywords for each book and chapter.
If you think you should have access to this title, please contact your librarian.