Accumetric Test Performance Analysis software, receiver operating characteristic analysis, nonparametric, sensitivity, specificity, likelihood ratio, predictive value, kappa, weighted kappa, intraclass correlation

Published Reviews

The predecessor of Accumetric Test Performance Analysis was AccuROC Nonparametric Receiver Operating Characteristic Analysis.

In a review of AccuROC in 2003, Stephan et al had this to say:

"...to our knowledge, at this stage it is the only program that uses" the totally nonparametric method of DeLong et al (DeLong et al, 1988).

"The layout of the program is very well structured, and because of the comprehensive manual and the up-to-date homepage, the program is easy to learn."

"Furthermore, AccuROC can calculate CIs and SD with a bootstrap method."

At this time, AccuROC was ranked second among eight software packages, and was recommended "...if one attaches great importance to highly accurate results and accepts the mentioned drawbacks..."

The mentioned drawbacks included complex data input, ability to print but not to save output, and inablility to easily analyze tests in which test scores are inversely related to the condition being detected.

We are very pleased to report that these drawbacks have been addressed in Accumetric Test Performance Analysis Version 1.1.

First, data input can now be done easily within Accumetric's own data input module or with spreadsheets or text editors. Accumetric's data file format is now "csv" (comma-separated values), a universally accepted format which can easily be exported from (or imported to) any spreadsheet.

Second, all output can now be printed and saved.

Third, Accumetric adds a new variable which specifies the direction of association between test scores and condition being detected (i.e. direct or inverse) and which therefore allows the analysis of tests in which greater or lesser scores are associated with the condition being detected, with no modification of test scores needed.

References

DeLong, ER, DeLong, DM and Clarke-Pearson, DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 1988; 44: 837-45.

Stephan, C, Wesseling, S, Schink, T and Jung, K. Comparison of eight computer programs for receiver-operating characteristic analysis. Clinical Chemistry 2003; 49(3): 433-439.