Analyse-it® Method Evaluation Edition
Evaluate diagnostic method performance to determine decision levels or choose the
best among many tests with ROC (receiver operating characteristics) curve plots
& analysis.
Determine the diagnostic accuracy of qualitative and quantitative diagnostic methods.
Methods can be compared to choose the best performing test, decision levels can
be determined, or diagnostic ability of new methods determined.
Rated #1 in Journal of Clinical Chemistry review of ROC curve software
The reviewers rated Analyse-it #1, better than both AccuRoc(#2) and MedCalc(#3).
And that was before we improved it! Analyse-it is now unsurpassed for ROC curve
analysis, supporting all the latest recommendations.
- Describe performance over all decision levels
For each decision level the true/false-positive & true/false-negative rates, predictive
values, likelihood ratios and cost can be shown.
- Visualise test performance with ROC curve plot and decision plots
Upto 6 ROC curves can be plotted, allowing the performance of upto 6 methods to
be assessed visually. Decision plots visualise performance across decision levels
in terms of sensitivity & specificity, positive & negative likelihood ratios,
positive & negative predictive values or cost.
- Compare upto 6 correlated diagnostic tests
Uses the DeLong, Delong, Clarke-Pearson curve comparision method -- the best method
for comparing correlated ROC curves, a less parametric approach than the early 1980's
Hanley & McNeil approach (in fact, now recommend by Hanley & McNeil themselves).
- Incorporate prevalence and costs
Optionally include prevalance of the condition, or costs of misdiagnosis for more
accurate evaluation of the performance of decision levels.
- Describe the performance of a qualitative test
Satisfies CLSI EP12-A protocol. Calculates sensitivity/specificity, likelihood ratios,
and predictive values.
What next?
Download a free 30-day evaluation