ROC Curve and Diagnostic Performance Software
Establish diagnostic accuracy, compare test performance, and determine optimal decision thresholds — following CLSI EP24-A2 and EP12-A2, all within Excel.
Try free for 15 days — no credit card required
Rated best ROC curve software — Clinical Chemistry, March 2003.
Trusted by 75,000 scientists at most of the top-10 IVD manufacturers, and at thousands of ISO 15189, ISO/IEC 17025, and CLIA-regulated laboratories worldwide for over 30 years.
Complete diagnostic performance evaluation
Sensitivity, specificity, likelihood ratios, predictive values, AUC with confidence intervals, and the full set of EP24-A2 and EP12-A2 statistical tests — for both quantitative and qualitative diagnostic tests, in a single validated package.
Results you can stand behind
Every calculation runs in Analyse-it’s own validated engine — no Excel formulas, no third-party functions. Validated against CLSI EP24-A2 and EP12-A2 reference datasets before every release. Defensible in a 510(k), CE-IVD submission, or peer-reviewed publication.
Your data never leaves your environment
Analyse-it runs entirely on your PC. Nothing is sent to a server, nothing is exposed to a cloud service - important when working with pre-submission or patient-adjacent data under data governance constraints.
Establishing that a diagnostic test works is only the first question. The harder ones come next: how does it compare to the existing test or the gold standard? At what threshold does it perform best, and what does “best” mean in your clinical context — minimising false negatives, controlling false positives, or optimising across both factoring in the costs of misdiagnosis? And how does performance vary across different patient populations or clinical subgroups?
ROC curve analysis is the standard framework for working through all of these questions. CLSI EP24-A2 defines the statistical approach for quantitative tests; EP12-A2 covers qualitative test evaluation. Analyse-it implements both — AUC estimation, confidence intervals, test comparison, threshold optimisation, subgroup analysis, and the full decision plots that show how sensitivity, specificity, likelihood ratios, and predictive values shift across every possible threshold.
Every calculation runs in Analyse-it’s own validated engine. Results are validated against CLSI reference datasets before every release. Whether the analysis is going into a 510(k) submission, a CE-IVD dossier, or a peer-reviewed publication, the numbers are ones you can defend.
Seen in the field
What's included
-
Quantitative diagnostic performance — EP24-A2
ROC curve analysis for up to 10 paired or independent tests. Sensitivity, specificity, likelihood ratios, predictive values, odds ratio, and Youden’s index. Wilcoxon-Mann-Whitney AUC with DeLong-DeLong-Clarke-Pearson confidence intervals. Bi-histogram and dot-plot of positive and negative outcome distributions. Full decision plot across all thresholds showing sensitivity, specificity, likelihood ratios, predictive values, or cost-weighted accuracy.
-
Qualitative test performance — EP12-A2
Sensitivity and specificity with Clopper-Pearson exact or Wilson score confidence intervals. Likelihood ratios, predictive values, and odds ratio. Comparison of two paired or independent qualitative tests with Newcombe or Tango score confidence intervals on differences. McNemar-Mosteller exact, Fisher exact, and Score Z tests for equality of sensitivity and specificity.
-
Test comparison
Overlay and compare ROC curves for multiple tests or subgroups. DeLong-DeLong-Clarke-Pearson difference in AUC with confidence intervals. Test for equality, equivalence, or non-inferiority — so the statistical basis for your conclusions about which test performs better is explicit and defensible.
-
Optimal threshold determination
Find the optimal decision threshold using the Youden index, cost-based optimisation factoring in the relative costs of false positives and false negatives, or fixed constraints on sensitivity or specificity. Predict sensitivity and specificity at any threshold, or find the threshold that meets a specified performance requirement.
-
Subgroup and population analysis
Assess ROC performance across different patient populations, clinical strata, or demographic subgroups. Understand whether test performance is consistent across the groups your clinical decision affects.
-
Publication-ready reports
Generate clean, formatted reports and plots directly from your analysis. No reformatting, no copy-paste errors — whether the output is going into a regulatory dossier, a clinical study report, or a peer-reviewed journal submission.
-
Validated before every release
Every calculation is performed by Analyse-it’s own engine — no Excel formulas or functions. Results are validated against the example datasets in the CLSI EP24-A2 and EP12-A2 guidelines before every release, so you can trust what you’re signing off on. See our development and validation process →
-
Your data never leaves your environment
Analyse-it runs locally on your PC. No cloud processing, no data transmission, no exposure risk — your patient data stays where it belongs.
-
More than just ROC curves
Diagnostic performance is one part of the Method Validation edition, which covers the full CLSI method validation workflow in a single validated package — measurement system analysis (EP05, EP06, EP15, EP17), method comparison (EP09, EP21), reference intervals (EP28), and total analytical error (EP21). One tool for everything your laboratory or IVD submission requires.
Join 75,000 scientists who rely on Analyse-it. No credit card required. Full functionality from day one.
Try free for 15 days — no credit card required
