Transferability of neural network-based decision support algorithms for early assessment of chest-pain patients

The present investigation concerns methodological and epidemiological aspec ts of the transferability of artificial neural network-based algorithms, as key-components for classification in decision support systems (DSS). The p revalence of pathological conditions to be detected must be known in order to tune an artificial neural networks (ANN)-decision algorithm so that the predictive values of the outcome fulfil medical requirements. Another aspec t of transferability, when clinical laboratory results are used, concerns d ifferences in analytical performance of measuring instruments. The relative bias between two instruments is not known exactly, but must be estimated a nd corrected for. A general method, based on original measured data sets an d statistical modeling, was developed for simulating the impact of various correction procedures when using different analytical instruments. The simu lation methodology was applied to a real clinical problem of ruling-in/ruli ng-out of patients with suspected acute myocardial infarction (AMI) by bioc hemical monitoring. The recommended correction procedure was based on metho d comparison with use of five duplicate measurements on a common set of pat ient samples covering the relevant measuring interval. Transferability of l aboratory data over time was also studied. The design of quality assurance procedures should be based on analytical quality requirement specifications related to medical needs. Limits of critically sized systematic errors wer e assessed by calculating the decrease in diagnostic performance of the ANN -algorithm as a result of temporary analytical disturbances. The consequenc es for the design of QA procedures was illustrated. It is concluded that th e actual ANN-decision algorithm for early assessment of chest-pain patients should be possible to transfer to new sites under realistic conditions. (C ) 2000 Elsevier Science Ireland Ltd. All rights reserved.