Modified binary PSO for feature selection using SVM applied to mortality prediction of septic patients

• Septic shock patients’ outcome prediction.
• Modified binary particle swarm optimization method.
• Feature selection with the simultaneous optimization of SVM kernel parameters.
• Improved results for sepsis outcome prediction.

This paper proposes a modified binary particle swarm optimization (MBPSO) method for feature selection with the simultaneous optimization of SVM kernel parameter setting, applied to mortality prediction in septic patients. An enhanced version of binary particle swarm optimization, designed to cope with premature convergence of the BPSO algorithm is proposed. MBPSO control the swarm variability using the velocity and the similarity between best swarm solutions. This paper uses support vector machines in a wrapper approach, where the kernel parameters are optimized at the same time. The approach is applied to predict the outcome (survived or deceased) of patients with septic shock. Further, MBPSO is tested in several benchmark datasets and is compared with other PSO based algorithms and genetic algorithms (GA). The experimental results showed that the proposed approach can correctly select the discriminating input features and also achieve high classification accuracy, specially when compared to other PSO based algorithms. When compared to GA, MBPSO is similar in terms of accuracy, but the subset solutions have less selected features.

As can be seen in this figure, even if most of the particles converge to a suboptimal xsb (a), when the mechanism of local search is applied the particles are displaced (b). This mechanism can occur simultaneously with the reset of the swarm best xsb (b). The algorithm will be able to converge towards a better solution (c). Therefore, we will be able to refine the results around the best position reducing the risk of causing premature convergence of the algorithm.Figure optionsDownload as PowerPoint slide