How to use
- Enter TP, FP, TN, and FN from one binary decision table.
- Optionally rename the positive and negative labels.
- Read DOR together with LR+ and LR- so the large ratio stays interpretable.
Wave 8 diagnostic metrics
One summary ratio for threshold separation
Diagnostic odds ratio is convenient when you want a single threshold number, but it should still be read beside LR+, LR-, and the raw counts.
Inputs
Run a calculation to review DOR together with LR+, LR-, sensitivity, and specificity.
Why DOR can be useful
Diagnostic odds ratio folds both positive and negative evidence into one threshold summary. Larger values mean the current operating point separates positive and negative cases more strongly.
Because DOR can hide whether the strength comes from positive calls, negative calls, or both, keep LR+ and LR- visible as companion metrics.
Frequently asked questions
What is diagnostic odds ratio?
Diagnostic odds ratio combines the evidence from positive and negative results into one ratio. Larger values mean the test or model separates positive and negative cases more strongly.
How is DOR related to LR+ and LR-?
DOR equals LR+ divided by LR-. That means it summarizes both how convincing positive results are and how reassuring negative results are.
When can DOR become infinite?
DOR can become very large or infinite when false positives or false negatives are zero. That is possible with small samples, so keep the raw counts visible.
Does the share URL include my counts or labels?
No. The share URL stores only lightweight settings such as decimal places. Counts and custom labels stay in your browser.
Related
Comments (optional)
To reduce load, comments are fetched only when needed.