Book contents
- Frontmatter
- Contents
- Preface
- Acknowledgements
- Glossary
- Abbreviations
- 1 Basics of breast MRI
- 2 Imaging-related anatomy and pathology
- 3 Interpreting breast MRI studies
- 4 MRI-guided biopsy techniques
- 5 High-risk screening using breast MRI
- 6 Preoperative staging with breast MRI
- 7 Problem-solving applications of breast MRI
- 8 MRI after breast augmentation
- Answers to multiple choice questions
- Appendices
- 1 Nephrogenic systemic fibrosis
- 2 Sensitivity and specificity
- 3 TNM classification
- 4 Overview of surgical procedures
- 5 Overview of radiation therapy
- 6 Overview of systemic therapy
- 7 Primary prevention strategies for high-risk women
- 8 Breast cancer genes and genetic testing
- Index
- Plate section
2 - Sensitivity and specificity
Published online by Cambridge University Press: 05 March 2012
- Frontmatter
- Contents
- Preface
- Acknowledgements
- Glossary
- Abbreviations
- 1 Basics of breast MRI
- 2 Imaging-related anatomy and pathology
- 3 Interpreting breast MRI studies
- 4 MRI-guided biopsy techniques
- 5 High-risk screening using breast MRI
- 6 Preoperative staging with breast MRI
- 7 Problem-solving applications of breast MRI
- 8 MRI after breast augmentation
- Answers to multiple choice questions
- Appendices
- 1 Nephrogenic systemic fibrosis
- 2 Sensitivity and specificity
- 3 TNM classification
- 4 Overview of surgical procedures
- 5 Overview of radiation therapy
- 6 Overview of systemic therapy
- 7 Primary prevention strategies for high-risk women
- 8 Breast cancer genes and genetic testing
- Index
- Plate section
Summary
A number of useful statistics can be derived from a matrix table which shows the performance of a diagnostic test compared to a defined reference or “gold” standard.
Total subjects tested = TP + FN + FP + TN
Number of subjects with disease = TP + FN
Number of subjects without disease = TN + FP
Sensitivity = percentage correctly identified by the test = TP / (TP + FN)
Specificity = TN / (TN + FP)
PPV = TP / (TP + FP)
NPV = TN / (TN + FN)
Accuracy = percentage of correct results in all tests = (TP + TN) / TOTAL
Note that the prevalence of the target disease in the study population (or pre-test probability) has a significant effect on both PPV and NPV. For example, the PPV of an abnormal MRI finding is greater in a high-risk population than it is in the general population at average risk [5].
- Type
- Chapter
- Information
- Handbook of Breast MRI , pp. 191Publisher: Cambridge University PressPrint publication year: 2011