What does this research mean for the field?

An interpretable EEG-based depression recognition method utilizing a hybrid feature selection approach of RankSearch and Genetic Algorithm achieves 95.08% accuracy and an AUC of 0.9514 on the HUSM dataset while offering lower computational complexity than deep learning models. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

The aim is to develop an efficient and interpretable method for recognizing depression using EEG signals.

April 3, 2026Bioengineering0 citationsOpen Access

An Effective and Interpretable EEG-Based Depression Recognition Method Using Hybrid Feature Selection

XXXun XuShanghai University QFQiuyun FanTianjin University SJShanjing JuNanjing University of Posts and Telecommunications

Key Result

An interpretable EEG-based depression recognition method using hybrid feature selection achieved 95.08% accuracy, 95.99% sensitivity, and an AUC of 0.9514 on the public HUSM dataset.

Key Points

The aim is to develop an efficient and interpretable method for recognizing depression using EEG signals.
Extract multi-domain features from EEG signals
Utilize RankSearch for preliminary feature screening
Apply Genetic Algorithm to explore inter-feature relationships
Evaluate performance using multiple machine learning classifiers
Conduct rigorous cross-validation on the HUSM dataset
Achieved 95.08% accuracy in depression classification
Sensitivity of 95.99% and specificity of 94.30%
F1-score of 95% and AUC of 0.9514
Confirmed interpretability through feature importance analysis
Lower computational complexity compared to existing models

Structured PICO

Does a hybrid feature selection method using RankSearch and GA improve EEG-based depression recognition compared to existing models?

Population

EEG signals from the public HUSM dataset

Intervention

Hybrid feature selection method integrating RankSearch and Genetic Algorithm (GA) with machine learning classifiers

Comparator

Existing complex deep learning models

Outcome

Classification performance (accuracy, sensitivity, specificity, F1-score, AUC)surrogate

A hybrid feature selection method combining RankSearch and Genetic Algorithm provides highly accurate and interpretable EEG-based depression recognition with lower computational complexity than deep learning models.

Main Result

Absolute Event Rate: 0% vs 0%

Abstract

Recent studies on EEG-based automated depression detection have primarily depended on complex deep learning models. While these methods improve classification performance, their practical application is limited by high computational complexity, challenging training processes, and poor interpretability. This paper proposes an efficient method for depression recognition, which extracts multi-domain features from preprocessed EEG signals and selects the most discriminative feature subset by integrating the rapid preliminary screening capability of RankSearch with the interactive optimization ability of the Genetic Algorithm (GA). Our approach first eliminates redundant features efficiently through RankSearch, then deeply explores inter-feature relationships via GA, significantly enhancing classification performance while maintaining feature-level interpretability. Using the optimized feature subset, we evaluate performance with multiple machine learning classifiers (Decision Tree, KNN, Random Forest, SVM, XGBoost). Experiments on the public HUSM dataset demonstrate superior performance under rigorous cross-validation (accuracy = 95.08%, sensitivity = 95.99%, specificity = 94.30%, F1-score = 95%, AUC = 0.9514), with feature importance analysis further confirming interpretability. Compared to existing models, our method achieves lower computational complexity and higher clinical practicality, offering a more efficient technical solution for objective depression diagnosis.

AI से पूछें

Bookmark

View Full Paper