A particle swarm optimization levy flight algorithm for imputation of missing creatinine dataset

Abstract

Clinicians could intervene during what may be a crucial stage for preventing permanent kidney injury if patients with incipient Acute Kidney Injury (AKI) and those at high risk of developing AKI could be identified. This paper proposes an improved mechanism to machine learning imputation algorithms by introducing the Particle Swarm Levy Flight algorithm. We improve the algorithms by modifying the Particle Swarm Optimization Algorithm (PSO), by enhancing the algorithm with levy flight (PSOLF). The creatinine dataset that we collected, including AKI diagnosis and staging, mortality at hospital discharge, and renal recovery, are tested and compared with other machine learning algorithms such as Genetic Algorithm and traditional PSO. The proposed algorithms' performances are validated with a statistical significance test. The results show that SVMPSOLF has better performance than the other method. This research could be useful as an important tool of prognostic capabilities for determining which patients are likely to suffer from AKI, potentially allowing clinicians to intervene before kidney damage manifests.

Missing Value Imputation Designs and Methods of Nature-Inspired Metaheuristic Techniques: A Systematic Review

“…Multi objective approaches Hybrid approaches GA [16]- [18] GP [19] PSO [20] MOGA-II [21]- [22] MOPSO [23] Bayesian ACO+ Bayesian [24] ABC+ Bayesian [25] Max-min ACO +bayesian [26]- [27] Bayesian+ tensor+chaotic PSO [28] Probabilistic GA+KNN [29] GMSA+MPSO+ WKNN [30] PSO+ covariance matrix [32] IDW+TR+ PSO [33] Clustering ACO+ clustering [34] FCM+GA [35][36] FCM+ SVR+GA [37] GA+SOM [38] FCM+PSO [39]- [42] GFM+PSO [43] PSO-ECM+ AAELM [44] ELM+PSO+ FCM [45] PSO+K-means+ ontology [46] SOM+FOA +LSSVM [47] DE+ clustering [48] GA+RF [49] GP+wrapper [55] [56] Neural network GSO+MLP [57] GA+MLP, SA+MLP, PSO+MLP, RF+MLP [58] SC-FITNET [59] SC-FDO+ MLP [60] DL-CS [61] DL-BAT [62] DL-GSA [63] PSO+LSVM [54] PSO+levy flight+SVM [53] MAIS+GA [50] GA+ARO [51] GP+tree vector [52] KNN+LAHC AWOA [31] The proposed approach enhanced imputation for missing multivariate data…”

“…In [53], Ismail et al incorporated levy flight into PSO to improve global exploration of PSO and helped PSO to escape from local optimum. The results indicated that support vector machine (SVM) imputation, optimized by levy flight PSO achieved the lowest error for filling the incomplete creatinine data than KNN, naïve Bayes, and decision tree imputation.…”

“…The results revealed that the DL-GSA imputation method produced more substantial correlation results and lower MSE than the works in [61] and [62]. Some researchers also adopted MAE to measure the proposed imputation methods in terms of the average magnitude of the errors for continuous variables [53], [58]- [60].…”

“…It is not envisaged that some features are missing for all patients in medical datasets. In real-life cases, some features may be missing by certain patients [50], [53], [72] in medical datasets. The occurrence of the missingness pattern depends on the observed values of other features in the dataset.…”

An Improved particle swarm optimization based on lévy flight and simulated annealing for high dimensional optimization problem

“…Lévy flight is a type of random walk introduced by Paul Lévy in 1937 and has the characteristic of intensive probability in its movement [30]- [33]. Yang, Ting, and Karamanoglu [31] implied that Lévy flight has an excellent capability to explore the search space.…”