Analytical Method Performance Evaluation (AMPE) – A software tool for analytical method validation.


2007 - Journal of AOAC INTERNATIONAL, 90, 1432-1438
Acutis, M., Trevisiol, P., Confalonieri, R., Bellocchi, G., Grazioli, E., van den Eede, G., Paoletti, C.

Abstract:

A Windows-based software tool [Analytical Method Performance Evaluation (AMPE)] was developed to support the validation of analytical methods. The software implements standard statistical approaches commonly adopted in validation studies to estimate analytical method performance (limits of detection and quantitation, accuracy specificity, working range, and linearity of responses) according to ISO 5725. In addition, AMPE proposes the application of innovative and unique approaches for the assessment of analytical method performance. Specifically, AMPE proposes the use of difference-based indexes to quantify the agreement between measurements and reference values, the use of pattern indexes to quantify methods bias with respect to specific external variables, and the application of fuzzy logic to aggregate into synthetic indicators the information collected independently via the different performance statistics traditionally estimated in validation studies. Aggregated
measures are particularly useful for methods comparison, when more than one method is available for a specific analysis and itmay be of
interest to identify the best performing one taking into account, simultaneously, the information available from different performance statistics.
Illustrative examples of the type of outputs expected from AMPE-based validation sessions are given. The extensive data handling capabilities
and the wide range of statistics supplied in the software package makes AMPE suitable for specific needs that may arise in different validation
studies. The installation package, complete with a fully documented help file, is distributed free of charge to interested users along with input files
exemplary of the type of entry data required to run validation data analyses.


Keywords: no keywords
DOI: no doi