A Novel Swarm Intelligence-Driven Feature Selection for Interpretable Machine Learning in Multiparametric MRI-Based GBM Overall Survival Analysis.
Authors
Affiliations (3)
Affiliations (3)
- School of Engineering, Cardiff University, Cardiff CF24 3AA, UK.
- School of Computer Science and Informatics, Cardiff University, Cardiff CF24 4AG, UK.
- Department of Oncology, Velindre University NHS Trust, Cardiff CF14 2TL, UK.
Abstract
In this study, we develop and validate an interpretable machine learning (ML) model that integrates a hybrid swarm intelligence (SI)-based feature selection method with multiparametric magnetic resonance imaging (MRI)-derived RFs to estimate overall survival (OS) in glioblastoma multiforme (GBM) patients. A cohort of 276 GBM patients with open-access pre-treatment MRI data was used to perform comprehensive radiomic analysis. In the training (discovery) dataset, we employed five-fold cross-validation combined with bootstrapping to ensure robust methodological validation. Model evaluation covered the concordance index (C-index) with 95% confidence intervals (CIs). Additionally, survival stratification was performed using Kaplan-Meier curves and the log-rank test to separate patients into low- and high-risk groups for OS. The final survival model integrates patient age and ten independent RFs. The model's performance in the holdout test dataset was evaluated by a C-index of 0.71 (95% CI: 0.63-0.80), exhibiting statistically significant risk stratification (<i>p</i> = 3 × 10<sup>-4</sup>). Upon external validation, the model achieved a C-index of 0.67, maintaining statistical significance (<i>p</i> = 1 × 10<sup>-2</sup>). The research combined a traditional regularized Cox regression (Cox-LASSO) model with a new SI-based LASSO-PSO method, yielding significant stratification. To our knowledge, the present study offers one of the first studies to document the use of an interpretable ML model with an SI-based approach for successful risk stratification based on OS.