Mining association rules in incomplete information systems

Based on the rough set theory which is a powerful tool in dealing with vagueness and uncertainty, an algorithm to mine association rules in incomplete information systems was presented and the support and confidence were redefined. The algorithm can mine the association rules with decision attributes directly without processing missing values. Using the incomplete dataset Mushroom from UCI machine learning repository, the new algorithm was compared with the classical association rules mining algorithm based on Apriori from the number of rules extracted, testing accuracy and execution time. The experiment results show that the new algorithm has advantages of short execution time and high accuracy.

QTc Interval Predicts Outcome of Catheter Ablation in Paroxysmal Atrial Fibrillation Patients with Type 2 Diabetes Mellitus

Catheter ablation has been recommended as a treatment option for paroxysmal atrial fibrillation（PAF） patients complicated with type 2 diabetes mellitus（T2DM）. PAF patients with T2 DM have a higher recurrence rate after catheter ablation. Prolongation of corrected QT（QTc） interval has been linked to poor outcomes in T2 DM patients. Whether the abnormal QTc interval is associated with the ablation outcome in the PAF patients with T2 DM remains unknown. In this study, 134 PAF patients with T2 DM undergoing primary catheter ablation were retrospectively enrolled. Pre-procedural QTc interval was corrected by using the Bazett＇s formula. Cox proportional hazards models were constructed to assess the relationship between QTc interval and the recurrence of AF. After a 29.1-month follow-up period, 61 patients experienced atrial tachyarrhythmia recurrence. Recurrent patients had a longer QTc interval than non-recurrent patients（425.2±21.5 ms vs. 414.1±13.4 ms, P=0.002）. Multivariate Cox regression analysis revealed that QTc interval [hazard ratio（HR）=1.026, 95% confidence interval（CI） 1.012–1.040, P=0.005] and left atrial diameter（LAD）（HR=1.125, 95% CI 1.062–1.192, P=0.003） were independent predictors of recurrent atrial tachyarrhythmia. Receiver operating characteristic analysis demonstrated that the cut-off value of QTc（418 ms） predicted arrhythmia recurrence with a sensitivity of 55.7% and a specificity of 69.9%. A combination of LAD and QTc was more effective than LAD alone（P〈0.001） in predicting arrhythmia recurrence after the procedure. QTc interval could be used as an independent predictor of arrhythmia recurrence in T2 DM patients undergoing AF ablation, thus providing a simple method to identify those patients who likely have a better outcome following the procedure.

Static Frame Model Validation with Small Samples Solution Using Improved Kernel Density Estimation and Confidence Level Method

An improved method using kernel density estimation （KDE） and confidence level is presented for model validation with small samples. Decision making is a challenging problem because of input uncertainty and only small samples can be used due to the high costs of experimental measurements. However, model validation provides more confidence for decision makers when improving prediction accuracy at the same time. The confidence level method is introduced and the optimum sample variance is determined using a new method in kernel density estimation to increase the credibility of model validation. As a numerical example, the static frame model validation challenge problem presented by Sandia National Laboratories has been chosen. The optimum bandwidth is selected in kernel density estimation in order to build the probability model based on the calibration data. The model assessment is achieved using validation and accreditation experimental data respectively based on the probability model. Finally, the target structure prediction is performed using validated model, which are consistent with the results obtained by other researchers. The results demonstrate that the method using the improved confidence level and kernel density estimation is an effective approach to solve the model validation problem with small samples.