BACKGROUND
Time-averaged serum albumin (TSA) is commonly associated with clinical outcomes in hemodialysis (HD) patients and considered a surrogate indicator of nutritional status. Whale optimization (WO)-based feature selection algorithm could address the challenges associated with the complex characteristics of multifactor interactions and could be combined with regression models.
OBJECTIVE
The present study aimed to demonstrate an optimal multifactor TSA-associated model, which could be applied in the interpretation of the association between TSA and clinical factors in HD patients.
METHODS
A total of 829 HD patients who met the inclusion criteria were analyzed. Monthly serum albumin data tracked from January 2009 to December 2013 were converted into TSA categories based on a critical value of 3.5 g/dL. Multivariate logistic regression was used to analyze the association between TSA categories and multiple clinical factors using three types of feature selection models, namely the fully adjusted model, stepwise model, and whale optimization algorithm (WOA) model.
RESULTS
The WOA yielded the lowest Akaike Information Criterion (AIC) value, which indicated that the WOA could achieve superior performance in multifactor analysis when compared to the fully adjusted and stepwise models. The significant features in the optimal multifactor TSA-associated model included age, creatinine, potassium, and HD adequacy index (Kt/V level).
CONCLUSIONS
The WOA algorithm could select five features from 15 clinical factors, which is the minimum number of selected features required in multivariate regression models for optimal multifactor model construction to achieve high model performance. Therefore, the application of the optimal multifactor TSA-associated model could facilitate nutritional status monitoring in HD patients.
CLINICALTRIAL
All data were retrospectively collected using an approved data protocol (201800595B0) with a waiver of informed consent from patients.