Large-scale optimization is a challenging problem because it involves a large number of decision variables. In this paper, a simple but effective method, called hierarchical sorting swarm optimizer (HSSO), is proposed for large-scale optimization. As a variant of representative particle swarm optimizer (PSO), HSSO first sorts the initial particles according to their fitness values, and then partitions the sorted particles into two groups, namely, the good group corresponding to better fitness values, and the bad group with worse fitness values. The bad group is then updated by learning from the good one. After that, we take the good group as a new swarm and conduct the sorting and learning procedures. The aforementioned operations are repeated several times until only one particle left to form a hierarchical structure. In the experiments, HSSO is applied to optimize 39 benchmark test functions. The comparative results with several existing algorithms demonstrate that, despite its simplicity, HSSO shows improved performance in terms of both exploration and exploitation. INDEX TERMS Large-scale optimization, hierarchical sorting swarm optimizer, hierarchical learning, particle swarm optimization.
BackgroundAcute kidney injury (AKI) is the most common major complication of cardiac surgery field. The purpose of this study is to investigate the association between acute kidney injury and the prognoses of cardiac surgery patients in the Medical Information Mart for Intensive Care III (MIMIC-III) database.MethodsClinical data were extracted from the MIMIC-III database. Adult (≥18 years) cardiac surgery patients in the database were enrolled. Multivariable logistic regression analyses were employed to assess the associations between acute kidney injury (AKI) comorbidity and 30-day mortality, 90-day mortality and hospital mortality. Different adjusting models were used to adjust for potential confounders.ResultsA total of 6,002 patients were involved, among which 485 patients (8.08%) had comorbid AKI. Patients with AKI were at higher risks of prolonged ICU stay, hospital mortality, 90-day mortality (all P < 0.001), and 30-day mortality (P = 0.008). AKI was a risk factor for hospital mortality [Model 1, OR (95% CI) = 2.50 (1.45–4.33); Model 2, OR (95% CI) = 2.44 (1.48–4.02)], 30-day mortality [Model 1, OR (95% CI) = 1.84 (1.05–3.24); Model 2, OR (95% CI) = 1.96 (1.13–3.22)] and 90-day mortality [Model 1, OR (95% CI) = 2.05 (1.37–3.01); Model 2, OR (95% CI) = 2.76 (1.93–3.94)]. Higher hospital mortality, 30-day mortality and 90-day mortality was observed in higher KDIGO grade for cardiac surgery patients with AKI (all P < 0.05).ConclusionComorbid AKI increased the risk of hospital mortality, 30-day mortality, and 90-day mortality of cardiac surgery patients in the MIMIC-III database.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.