Detection of tuberculosis using cough audio analysis: a deep learning approach with capsule networks
Abstract Purpose Tuberculosis (TB) is a widespread infectious disease that requires early detection for effective treatment and control. This study aims to improve TB detection using cough audio analysis, comparing the performance of capsule networks to other deep learning models. Methods We used co...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2024-11-01
|
| Series: | Discover Artificial Intelligence |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s44163-024-00179-4 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850196013194674176 |
|---|---|
| author | Sakthi Jaya Sundar Rajasekar Anu Rithiga Balaraman Deepa Varnika Balaraman Saleem Mohamed Ali Kannan Narasimhan Narayanasamy Krishnasamy Varalakshmi Perumal |
| author_facet | Sakthi Jaya Sundar Rajasekar Anu Rithiga Balaraman Deepa Varnika Balaraman Saleem Mohamed Ali Kannan Narasimhan Narayanasamy Krishnasamy Varalakshmi Perumal |
| author_sort | Sakthi Jaya Sundar Rajasekar |
| collection | DOAJ |
| description | Abstract Purpose Tuberculosis (TB) is a widespread infectious disease that requires early detection for effective treatment and control. This study aims to improve TB detection using cough audio analysis, comparing the performance of capsule networks to other deep learning models. Methods We used cough audio recordings from 1105 individuals with a new or worsening cough for at least two weeks, totaling 9772 recordings. These recordings were processed into spectral images, and HOG features were extracted. Various models, including Capsule Networks + FCNN, CNN, VGG16, and ResNet50 were trained and evaluated. Results Capsule Networks + FCNN achieved the best performance with an accuracy of 0.97, sensitivity of 0.98, specificity of 0.96, F1 score of 0.97, and precision of 0.97, outperforming other models. This attribute is due to the model’s ability to learn complex features from spectral images. Conclusions This study concludes that Capsule Networks are more effective than typical CNN-based models in diagnosing TB from cough audio. This suggests that advanced deep learning frameworks could significantly enhance TB screening accuracy, especially in resource-limited areas. |
| format | Article |
| id | doaj-art-075491dcfd5f48429f988f4c47082bd4 |
| institution | OA Journals |
| issn | 2731-0809 |
| language | English |
| publishDate | 2024-11-01 |
| publisher | Springer |
| record_format | Article |
| series | Discover Artificial Intelligence |
| spelling | doaj-art-075491dcfd5f48429f988f4c47082bd42025-08-20T02:13:35ZengSpringerDiscover Artificial Intelligence2731-08092024-11-014111110.1007/s44163-024-00179-4Detection of tuberculosis using cough audio analysis: a deep learning approach with capsule networksSakthi Jaya Sundar Rajasekar0Anu Rithiga Balaraman1Deepa Varnika Balaraman2Saleem Mohamed Ali3Kannan Narasimhan4Narayanasamy Krishnasamy5Varalakshmi Perumal6Melmaruvathur Adhiparasakthi Institute of Medical Sciences and ResearchAnna UniversityAnna UniversityMelmaruvathur Adhiparasakthi Institute of Medical Sciences and ResearchMelmaruvathur Adhiparasakthi Institute of Medical Sciences and ResearchThe Tamil Nadu Dr. M.G.R. Medical UniversityAnna UniversityAbstract Purpose Tuberculosis (TB) is a widespread infectious disease that requires early detection for effective treatment and control. This study aims to improve TB detection using cough audio analysis, comparing the performance of capsule networks to other deep learning models. Methods We used cough audio recordings from 1105 individuals with a new or worsening cough for at least two weeks, totaling 9772 recordings. These recordings were processed into spectral images, and HOG features were extracted. Various models, including Capsule Networks + FCNN, CNN, VGG16, and ResNet50 were trained and evaluated. Results Capsule Networks + FCNN achieved the best performance with an accuracy of 0.97, sensitivity of 0.98, specificity of 0.96, F1 score of 0.97, and precision of 0.97, outperforming other models. This attribute is due to the model’s ability to learn complex features from spectral images. Conclusions This study concludes that Capsule Networks are more effective than typical CNN-based models in diagnosing TB from cough audio. This suggests that advanced deep learning frameworks could significantly enhance TB screening accuracy, especially in resource-limited areas.https://doi.org/10.1007/s44163-024-00179-4TuberculosisDeep learningCapsule networksCough audio analysis |
| spellingShingle | Sakthi Jaya Sundar Rajasekar Anu Rithiga Balaraman Deepa Varnika Balaraman Saleem Mohamed Ali Kannan Narasimhan Narayanasamy Krishnasamy Varalakshmi Perumal Detection of tuberculosis using cough audio analysis: a deep learning approach with capsule networks Discover Artificial Intelligence Tuberculosis Deep learning Capsule networks Cough audio analysis |
| title | Detection of tuberculosis using cough audio analysis: a deep learning approach with capsule networks |
| title_full | Detection of tuberculosis using cough audio analysis: a deep learning approach with capsule networks |
| title_fullStr | Detection of tuberculosis using cough audio analysis: a deep learning approach with capsule networks |
| title_full_unstemmed | Detection of tuberculosis using cough audio analysis: a deep learning approach with capsule networks |
| title_short | Detection of tuberculosis using cough audio analysis: a deep learning approach with capsule networks |
| title_sort | detection of tuberculosis using cough audio analysis a deep learning approach with capsule networks |
| topic | Tuberculosis Deep learning Capsule networks Cough audio analysis |
| url | https://doi.org/10.1007/s44163-024-00179-4 |
| work_keys_str_mv | AT sakthijayasundarrajasekar detectionoftuberculosisusingcoughaudioanalysisadeeplearningapproachwithcapsulenetworks AT anurithigabalaraman detectionoftuberculosisusingcoughaudioanalysisadeeplearningapproachwithcapsulenetworks AT deepavarnikabalaraman detectionoftuberculosisusingcoughaudioanalysisadeeplearningapproachwithcapsulenetworks AT saleemmohamedali detectionoftuberculosisusingcoughaudioanalysisadeeplearningapproachwithcapsulenetworks AT kannannarasimhan detectionoftuberculosisusingcoughaudioanalysisadeeplearningapproachwithcapsulenetworks AT narayanasamykrishnasamy detectionoftuberculosisusingcoughaudioanalysisadeeplearningapproachwithcapsulenetworks AT varalakshmiperumal detectionoftuberculosisusingcoughaudioanalysisadeeplearningapproachwithcapsulenetworks |