Alibakhshi, Alireza, Hussain, Arshian, Dadashi Serej, Nasim, Ufumaka, Isreal, Ghasemi, Hadis, Khalili, Hosseinali and Zolgharni, Massoud ORCID: https://orcid.org/0000-0003-0904-2904 (2026) Orthogonal array-based feature selection for traumatic brain injury prognosis: a metaheuristic benchmarking framework. In: Artificial Intelligence in Healthcare. Springer, pp. 205-216.

Full text not available from this repository.

Abstract

Traumatic brain injury (TBI) outcome prediction remains a challenging problem because recovery is influenced by a wide range of demographic, clinical, laboratory, imaging, and treatment-related factors. This study investigated the use of metaheuristic feature selection methods for predicting 5-class functional outcome after TBI, measured using the Glasgow Outcome Scale Extended (GOSE). A Taguchi-inspired orthogonal array-based feature selection method was proposed and compared with four established metaheuristics: Genetic Algorithm, Grey Wolf Optimizer, Bayesian search, and Whale Optimization. The study used a cohort of 1,645 adult patients admitted to the neuro-intensive care units of Rajaee Hospital between 2016 and 2021, with each patient described by 65 features. Performance was evaluated using repeated stratified cross-validation (30 repetitions x 5 folds) across six classifiers. Multiple evaluation metrics were considered, including accuracy, balanced accuracy, macro F1-score, Matthews correlation coefficient, AUC, ROC analysis, calibration curves, and feature overlap analysis. The proposed Taguchi-based approach showed competitive and consistent performance across several metrics. In particular, the Taguchi-NaïveBayes subset achieved the highest mean accuracy (0.6337), balanced accuracy (0.3413), macro F1-score (0.3363), and MCC (0.3615), while the Bayesian search subset achieved the highest mean AUC (0.7211). These findings suggest that structured orthogonal array-based feature selection is a promising and efficient alternative to conventional stochastic metaheuristics for TBI outcome prediction.

Item Type:	Book Chapter or Section
Identifier:	10.1007/978-3-032-35387-0_15
Additional Information:	Book chapter of a conference paper presented in the International Conference on AI in Healthcare (AIiH), 28-28 August, London, United Kingdom.
Subjects:	Computing > Intelligent systems Medicine and health
Related URLs:	Organisation
Date Deposited:	24 Sep 2026
Dates:	Date Publication status 22 May 2026 Accepted 12 August 2026 Published Online
School, department or research centre:	THRIVE (The Centre for Translational Healthcare Research, Innovation, Vision, and Excellence)
URI:	https://repository.uwl.ac.uk/id/eprint/15319

Actions (admin access)