Sensitivity and specificity are fundamental characteristics of diagnostic imaging tests.
The two characteristics derive from a 2x2 box of basic, mutually exclusive outcomes from a diagnostic test:
- true positive (TP): an imaging test is positive and the patient has the disease/condition
- false positive (FP): an imaging test is positive and the patient does not have the disease/condition
- true negative (TN): an imaging test is negative and the patient does not have the disease/condition
- false negative (FN): an imaging test is negative and the patient has the disease/condition
On a first pass, we don't assume some relationship between the test and the disease/condition, but we hope there will be some relationship between the test and the disease/condition, because otherwise the test would be worthless.
For a given test and disease/condition, its sensitivity is how well it can be positive among all those with the condition. Therefore:
- sensitivity = TP / (TP + FN)
- true positives / (all those with the disease)
For a given test and disease/condition, its specificity is how well it can distinguish those with disease from those without. The test must not just fail to pick up a segment of the population (that might be poor sensitivity), it must distinguish those without the disease... the true negatives (TNs). Therefore:
- specificity = TN / (TN + FP)
- true negatives / (all those without the disease)
SpPin and SnNout rule
SnNout: if a diagnostic test, characterized by high sensitivity (Sn), returns the negative value (N), then it excludes the diagnosis (out) 2-3.
SpPin: if a diagnostic test, characterized by high specificity (Sp), returns the positive value (P), then it admits the diagnosis (in) 2-3.
- 1. Weinstein S, Obuchowski NA, Lieber ML. Clinical evaluation of diagnostic tests. AJR Am J Roentgenol. 2005;184 (1): 14-9. doi:10.2214/ajr.184.1.01840014 - Pubmed citation
- 2. Carl Heneghan and Douglas Badenoch: Evidence-based Medicine Toolkit, Second Edition. 2006, Blackwell Publishing Ltd.
- 3. Baeyens JP, Serrien B, Goossens M, Clijsen R. Questioning the "SPIN and SNOUT" rule in clinical testing. (2019) Archives of physiotherapy. 9: 4. doi:10.1186/s40945-019-0056-5 - Pubmed
Related Radiopaedia articles
- clinical trials
- descriptive studies
- Bayes' theorem
- sensitivity and specificity
- positive predictive value (PPV)
- negative predictive value (NPV)
- likelihood ratio (LR)
- normal distribution
- type I error
- type II error
- confidence interval
- ROC curve
- retrospective studies
- prospective studies
- analyzes of variance
- non-parametric statistics
- cognitive bias in image perception