An approach for Ewing test selection to support the clinical assessment of cardiac autonomic neuropathy

摘要: Objective: This article addresses the problem of determining optimal sequences tests for clinical assessment cardiac autonomic neuropathy (CAN). We investigate accuracy using only one recommended Ewing to classify CAN and additional obtained by adding remaining battery. is important as not all five can always be applied in each situation practice. Methods material: used new unique database diabetes screening research initiative project, which more than ten times larger data set his original investigation CAN. utilized decision trees path finder (ODPF) procedure identifying tests. Results: present experimental results on that achieved found best cost-function equal number The accuracies initial segments 2, 3 4 categories are 80.80, 91.33, 93.97 94.14, respectively, 79.86, 89.29, 91.16 91.76, 78.90, 86.21, 88.15 88.93. They show significant improvement compared sequence considered previously literature mathematical expectations a random complete outcomes subsets features required any with use ODPF procedure. have also two most increase when some attributes cannot obtained. Conclusions: determine individual following single test diagnosing deep breathing heart rate variation test. Optimal guarantee after provide an comparison previous ordering or sequence.