17 January 2025 in Revue Africaine de La Recherche En Informatique Et Mathématiques Appliquées
Soumission à EpisciencesCough is the most common symptom of lung disease. COVID-19, a respiratory illness, has caused over 700 million positive cases and 7 million deaths worldwide. An effective, affordable, and widely available diagnostic tool is crucial in combating lung disease and the COVID-19 pandemic. Deep learning and machine learning algorithms could be used to analyze the cough sounds of infected patients and make predictions. Our research lab and the COUGHVID research lab provide the cough data. This diagnostic approach can distinguish cough sounds from COVID-19 patients and people suffering from other ailments as well as healthy people using deep learning and feature extraction from Mel spectrograms. The model used is a variant of ConvNet. This ConvNet model can easily capture features in MFCC Vectors and enable convolution parallelism, which increases processing speed. ConvNet attains translational invariance in features through the sharing of weights between layers. During data acquisition for model training, it is important to consider quiet environments to reduce errors in audio quality. The architecture of the convolutional neural networks gives an F1-score of 89%, an accuracy of 90.33% and sensitivity of 87.3%. This system has the potential to significantly impact society by reducing virus transmission, expediting patient treatment, and freeing up hospital resources. Early detection of COVID-19 can prevent disease progression and enhance screening effectiveness.La toux est le symptôme le plus courant des maladies pulmonaires. Le COVID-19, une maladie respiratoire, a provoqué plus de 700 millions de cas positifs et 7 millions de décès dans le monde. Un outil de diagnostic efficace, abordable et largement disponible est essentiel pour lutter contre les maladies pulmonaires et la pandémie de COVID-19. Des algorithmes d'apprentissage profond et d'apprentissage automatique pourraient être utilisés pour analyser les bruits de toux des patients infectés et faire des prédictions. Notre laboratoire de recherche et le laboratoire de recherche COUGHVID fournissent les données sur la toux. Cette approche diagnostique peut distinguer les bruits de toux des patients atteints de COVID-19 et des personnes souffrant d'autres maladies ainsi que des personnes en bonne santé en utilisant l'apprentissage profond et l'extraction de caractéristiques à partir des spectrogrammes de Mel. Le modèle utilisé est une variante de ConvNet. Ce modèle ConvNet peut facilement capturer les caractéristiques des vecteurs MFCC et permettre le parallélisme de convolution, ce qui augmente la vitesse de traitement. ConvNet atteint l'invariance translationnelle des caractéristiques grâce au partage des poids entre les couches. Lors de l'acquisition des données pour l'entraînement du modèle, il est important de prendre en compte les environnements calmes pour réduire les erreurs de qualité audio. L'architecture des réseaux neuronaux convolutionnels donne un score F1 de 89 %, une précision de 90,33 % et une sensibilité de 87,3 %. Ce système peut avoir un impact significatif sur la société en réduisant la transmission des virus, en accélérant le traitement des patients et en libérant des ressources hospitalières. La détection précoce du COVID-19 peut prévenir la progression de la maladie et améliorer l'efficacité du dépistage.
28 September 2026 in Canadian Journal of Artificial Intelligence for Learning and Engineering Innovation
This study introduces a hybrid architecture designed to address the challenges of automated cough sound analysis by combining unsupervised detection with supervised classification. The system integrates a One-Class Support Vector Machine (SVM) trained exclusively on positive cough recordings to establish an acoustic boundary that isolates true cough events (inliers) while rejecting out-of-distribution background noise (outliers) from raw audio streams, and a convolutional neural network (CNN) to classify the extracted cough sounds into three clinically relevant categories. The preprocessing pipeline is integral to the system's effectiveness. The process involves extracting static Mel-Frequency Cepstral Coefficients (MFCCs), followed by standardization and dimensionality alignment to ensure compatibility with the CNN. These steps ensure that the spectral and temporal characteristics of cough sounds are accurately represented while reducing redundancy and enhancing separability in the feature space. Experimental evaluation demonstrates the reliability of this hybrid framework. As a standalone classifier on clean, isolated cough sounds, the DenseNet achieves an accuracy of 96.0% and an F1-score of 0.96. When evaluated on complex, unsegmented audio streams, the end-to-end hybrid system achieves an overall accuracy of 93.0% and a corresponding F1-score of 0.91. These results underscore the system's resilience and capacity for generalization, even when confronted with noisy or heterogeneous data. The incorporation of One-Class SVM for unsupervised detection ensures that cough events are effectively isolated, while the CNN provides precise classification, making the system suitable for real-world applications. Generalizability requires further validation on prospective external cohorts. Beyond its clinical relevance, this work also holds pedagogical value for engineering education. It illustrates the integration of signal processing, machine learning, and system design in a coherent pipeline, serving as a practical case study for students and researchers. By integrating theoretical concepts with applied methodologies, the system illustrates how hybrid architectures can be utilized to address complex challenges in health monitoring. Overall, the proposed framework offers a reliable foundation for preliminary screening and educational purposes.