Evaluating statistics – explanations of terminology in general use
1. Reliability – reproducibility of results. (These may be from the same observer or from different observers.) Assessment of this can be built into a study of diagnostic accuracy of a technique, or evaluated beforehand.
2. Accuracy: ‘proportion of results (positive and negative) which agree with the final diagnosis’, i.e.
3. Sensitivity – ‘proportion of diseased patients who are reported as positive’, i.e.
4. Specificity – ‘proportion of disease-free patients who are reported as negative’, i.e.
Buy Membership for Radiology Category to continue reading. Learn more here
