Complexity-driven feature selection for enhancing tuberculosis detection [r-libre/4226]

Ben Mahjouba, Sana, Ayena, Johannes C, Ouakrim, Youssef, Loulou, Karim, Grandjean Lapierre, Simon, Raberahona, Simon et Mezghani, Neila (2026). Complexity-driven feature selection for enhancing tuberculosis detection. Franklin Open, 17, 100759. 10.1016/j.fraope.2026.100759

Fichier(s) associé(s) à ce document :

PDF - BenMahjouba2026.pdf
Contenu du fichier : Version de l'éditeur
Licence : Creative Commons CC BY.

Télécharger

Catégorie de document :	Articles de revues
Évaluation par un comité de lecture :	Oui
Étape de publication :	Publié
Résumé :	Most existing machine learning approaches for tuberculosis (TB) screening typically utilize large, high-dimensional acoustic feature sets without examining their intrinsic discriminative power. To address this gap, we introduce a complexity-based feature selection approach that evaluates temporal and spectral descriptors using Fisher score (F1), class-distribution overlap (F2), and Shannon entropy (F4). Applied to the CODA-TB dataset (9772 audio recordings from 1105 participants), the proposed method identified 7 highly informative features from the original 26 features, primarily consisting of mel-frequency cepstral coefficients (MFCC) derivatives and spectral-shape measures. The resulting model achieved performance comparable to full-feature baselines while reducing feature dimensionality by 73% and computational cost by up to 14×. Comparative evaluation against four established feature selection techniques, supported by ablation and statistical analyses, confirmed the efficiency and robustness of the complexity-driven strategy, with no statistically significant loss in performance. These findings highlight the potential of lightweight, interpretable, and computationally efficient models for TB cough-based screening in resource-constrained environments
Adresse de la version officielle :	https://www.sciencedirect.com/science/article/pii/...
Déposant:	Ayena, Johannes
Responsable :	Johannes Ayena
Dépôt :	16 sept. 2026 14:11
Dernière modification :	16 sept. 2026 14:11

Actions (connexion requise)

RÉVISER