Exploring Automatic Diagnosis of COVID-19 from Crowdsourced Respiratory Sound Data
Open accessThe DOI matches its Crossref record; 35 of the 35 references it cites were found in a registry or an index.
Abstract
Audio signals generated by the human body (e.g., sighs, breathing, heart, digestion, vibration sounds) have routinely been used by clinicians as indicators to diagnose disease or assess disease progression. Until recently, such signals were usually collected through manual auscultation at scheduled visits. Research has now started to use digital technology to gather bodily sounds (e.g., from digital stethoscopes) for cardiovascular or respiratory examination, which could then be used for automatic analysis. Some initial work shows promise in detecting diagnostic signals of COVID-19 from voice and coughs. In this paper we describe our data analysis over a large-scale crowdsourced dataset of respiratory sounds collected to aid diagnosis of COVID-19. We use coughs and breathing to understand how discernible COVID-19 sounds are from those in asthma or healthy controls. Our results show that even a simple binary machine learning classifier is able to classify correctly healthy and COVID-19 sounds. We also show how we distinguish a user who tested positive for COVID-19 and has a cough from a healthy user with a cough, and users who tested positive for COVID-19 and have a cough from users with asthma and a cough. Our models achieve an AUC of above 80% across all tasks. These results are preliminary and only scratch the surface of the potential of this type of data and audio-based machine learning. This work opens the door to further investigation of how automatically analysed respiratory patterns could be used as pre-screening signals to aid COVID-19 diagnosis.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures








Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“COVID-19 has been studied using different audio sounds, including voice, breathing and cough sounds [10,29,79,80].”
From Human-centred artificial intelligence for mobile health sensing: challenges and opportunities (2023) · cites this paper as [79]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Davis, S., & Mermelstein, P. (1980) · IEEE Transactions on Acoustics, Speech, and Signal Processing · doi:10.1109/tassp.1980.1163420 · Cited by 5,465 · cited 1×
“ents obtained from the short-term power spectrum, based on a linear cosine transform of the log power spectrum on a nonlinear Mel scale. MFCCs are amongst the most common features in audio processing [9]. We use the first 13 components. • ∆-MFCC: the temporal differential (delta) of the MFCC [1]. ∆2-MFCC: the differen…”
Backgroundlibrosa: Audio and Music Signal Analysis in Python
McFee, B., Raffel, C., Liang, D., Ellis, D., McVicar, M., Battenberg, E., & Nieto, O. (2015) · Proceedings of the Python in Science Conference · doi:10.25080/majora-7b98e3ed-003 · Cited by 3,033 · cited 1×
“ustrates the data processing pipelines. 4.1 Feature extraction HandcraftedFeatures.The raw sound waveform recorded by the apps is resampled to 22kHz, a standard value for audio tasks. We used librosa [20] as our audio processing library. From the resampled audio various handcrafted features are extracted at the frame…”
MethodInfluentialCNN architectures for large-scale audio classification
Hershey, S., Chaudhuri, S., Ellis, D. P. W., Gemmeke, J. F., Jansen, A., Moore, R. C., Plakal, M., Platt, D., Saurous, R. A., Seybold, B., et al. (2017) · 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2017.7952132 · Cited by 2,023
(2010) · doi:10.5555/1756006.1859921 · Cited by 1,924
Real-time tracking of self-reported symptoms to predict potential COVID-19
Menni, C., Valdes, A. M., Freidin, M. B., Sudre, C. H., Nguyen, L. H., Drew, D. A., Ganesh, S., Varsavsky, T., Cardoso, M. J., El-Sayed Moustafa, J. S., et al. (2020) · Nature Medicine · doi:10.1038/s41591-020-0916-2 · Cited by 1,517
KERNEL METHODS IN MACHINE LEARNING
Hofmann, T., Schölkopf, B., & Smola, A. J. (2008) · Cited by 1,408
AI4COVID-19: AI enabled preliminary diagnosis for COVID-19 from cough samples via an app
Imran, A., Posokhova, I., Qureshi, H. N., Masood, U., Riaz, M. S., Ali, K., John, C. N., Hussain, M. I., & Nabeel, M. (2020) · Informatics in Medicine Unlocked · doi:10.1016/j.imu.2020.100378 · Cited by 621 · cited 2×
“In [17] a study of detection of coughs related to COVID-19 is presented using a cohort of 48 COVID-19 tested patients versus other pathology coughs on which an AI model is trained.”
MethodEmotionSense: a mobile phones based adaptive platform for experimental social psychology research
Rachuri, K. K., Musolesi, M., Mascolo, C., Rentfrow, P. J., Longworth, C., & Aucinas, A. (2010) · Proceedings of the 12th ACM international conference on Ubiquitous computing · doi:10.1145/1864349.1864393 · Cited by 508 · cited 1×
“In Emotionsense [26], the phone microphone is used as a sensor for users’ emotion detection in-thewild through Gaussian mixture models.”
MethodProceedings of the 13th International Society for Music Information Retrieval Conference, ISMIR 2012
Gouyon, F., Herrera, P., Martins, L. G., & Müller, M. A. (2012) · Cited by 446
Beat Tracking by Dynamic Programming
Ellis, D. P. W. (2007) · Journal of New Music Research · doi:10.1080/09298210701653344 · Cited by 391 · cited 1×
“• Tempo : the beats of the signals by measuring the onset strength, commonly used in music information retrieval [11].”
Cited by
The COUGHVID crowdsourcing dataset, a corpus for the study of large-scale cough analysis algorithms
Orlandic, L., Teijeiro, T., & Atienza, D. (2021) · Scientific Data · doi:10.1038/s41597-021-00937-4 · Cited by 312
“Several research groups have begun developing algorithms for diagnosing COVID-19 from cough sounds 8,9 .”
BackgroundCOVID-19 cough classification using machine learning and global smartphone recordings
Pahar, M., Klopper, M., Warren, R., & Niesler, T. (2021) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2021.104572 · Cited by 290
COVID-19 open source data sets: a comprehensive survey
Shuja, J., Alanazi, E., Alasmary, W., & Alashaikh, A. (2021) · Applied Intelligence · doi:10.1007/s10489-020-01862-6 · Cited by 274
“Researchers [37] developed a cross-platform application for crowd-sourced collection of voice sounds (cough and breath) to distinguish healthy and unhealthy persons.”
BackgroundApplications of artificial intelligence in battling against covid-19: A literature review
Tayarani N., M.-H. (2021) · Chaos, Solitons & Fractals · doi:10.1016/j.chaos.2020.110338 · Cited by 210
“In order to diagnose the disease via coughs and breathing, a inary machine learning classifier is used in [83] .”
Alafif, T., Tehame, A. M., Bajaba, S., Barnawi, A., & Zia, S. (2021) · International Journal of Environmental Research and Public Health · doi:10.3390/ijerph18031117 · Cited by 162
Lella, K. K., & Alphonse, P. J. A. (2021) · Alexandria Engineering Journal · doi:10.1016/j.aej.2021.06.024 · Cited by 168
COVID-19 Detection System using Recurrent Neural Networks
Hassan, A., Shahin, I., & Alsabek, M. B. (2020) · 2020 International Conference on Communications, Computing, Cybersecurity, and Informatics (CCCI) · doi:10.1109/ccci49893.2020.9256562 · Cited by 155
A Survey on Applications of Artificial Intelligence in Fighting Against COVID-19
Chen, J., Li, K., Zhang, Z., Li, K., & Yu, P. S. (2022) · ACM Computing Surveys · doi:10.1145/3465398 · Cited by 135
Applications of Big Data Analytics to Control COVID-19 Pandemic
Alsunaidi, S. J., Almuhaideb, A. M., Ibrahim, N. M., Shaikh, F. S., Alqudaihi, K. S., Alhaidari, F. A., Khan, I. U., Aslam, N., & Alshahrani, M. S. S. (2021) · Sensors · doi:10.3390/s21072282 · Cited by 161
“However, many people are not willing to share their health information with others, as well as other personal information like gender and location [45].”
BackgroundMethodInfluentialMachine learning for detecting COVID-19 from cough sounds: An ensemble-based MCDM method
Chowdhury, N. K., Kabir, M. A., Rahman, M. M., & Islam, S. M. S. (2022) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2022.105405 · Cited by 138
“[12] proposed a binary predictive model in which they used cough and breath to distinguish the sound of COVID-19 from people with asthma or healthy people.”
InfluentialRasheed, J., Jamil, A., Hameed, A. A., Aftab, U., Aftab, J., Shah, S. A., & Draheim, D. (2020) · Chaos, Solitons & Fractals · doi:10.1016/j.chaos.2020.110337 · Cited by 121
DiCOVA Challenge: Dataset, Task, and Baseline System for COVID-19 Diagnosis Using Acoustics
Muguli, A., Pinto, L., R, N., Sharma, N., Krishnan, P., Ghosh, P. K., Kumar, R., Bhat, S., Chetupalli, S. R., Ganapathy, S., et al. (2021) · Interspeech 2021 · doi:10.21437/interspeech.2021-74 · Cited by 101
Chaudhari, G., Jiang, X., Fakhry, A., Han, A., Xiao, J., Shen, S., & Khanzada, A. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.13320 · Cited by 129
“Various groups across the world have been focused on collecting copious high-quality data to train cough-based machine learning models, including Cambridge University in England, Carnegie Mellon University and the Massachusetts Institute of Technology (MIT) in the US, and Afeka College of Engineering in Israel.”
BackgroundCOVID-19 detection in cough, breath and speech using deep transfer learning and bottleneck features
Pahar, M., Klopper, M., Warren, R., & Niesler, T. (2022) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2021.105153 · Cited by 103
Coppock, H., Gaskell, A., Tzirakis, P., Baird, A., Jones, L., & Schuller, B. (2021) · BMJ Innovations · doi:10.1136/bmjinnov-2021-000668 · Cited by 108
Feature-Based Fusion Using CNN for Lung and Heart Sound Classification
Tariq, Z., Shah, S. K., & Lee, Y. (2022) · Sensors · doi:10.3390/s22041521 · Cited by 116
AI-enabled remote monitoring of vital signs for COVID-19: methods, prospects and challenges
Rohmetra, H., Raghunath, N., Narang, P., Chamola, V., Guizani, M., & Lakkaniga, N. R. (2021) · Computing · doi:10.1007/s00607-021-00937-7 · Cited by 120
“Researchers at the University of Cambridge are aiming to create a crowd-sourced dataset from recordings provided by participants through the COVID-19 Sounds App [15].”
MethodKhan, J. I., Khan, J., Ali, F., Ullah, F., Bacha, J., & Lee, S. (2022) · IEEE Access · doi:10.1109/access.2022.3181605 · Cited by 116
“Methods in feature extraction from audio signals include segmentation and normalization [571], harmonic to noise ratio [571], Variable Markov Oracle Method [572], empirical mode decomposition [573], Mel Spectrograms [574].”
MethodRespiratory Diseases Diagnosis Using Audio Analysis and Artificial Intelligence: A Systematic Review
Kapetanidis, P., Kalioras, F., Tsakonas, C. G., Tzamalis, P., Kontogiannis, G., Karamanidou, T., Stavropoulos, T. G., & Nikoletseas, S. (2024) · Sensors · doi:10.3390/s24041173 · Cited by 113
“The studies [53,54] use the ResNet model architecture, whereas [49,58] use the VGG-16, VGGish, and VGGish-BiGRU topologies, respectively.”
MethodInfluentialAutomated detection of COVID-19 cough
Tena, A., Claria, F., & Solsona, F. (2021) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2021.103175 · Cited by 107
“[13] used coughs and breathing to understand how discernible COVID-19 sounds are from those in asthma or healthy controls.”
Pal, A., & Sankarasubbu, M. (2021) · Proceedings of the 36th Annual ACM Symposium on Applied Computing · doi:10.1145/3412841.3441943 · Cited by 62
“This leads to irritation of the lung tissue that manifests into one of the earliest symptoms of COVID-19, dry cough, which is quite persistent.”
BackgroundCough Against COVID: Evidence of COVID-19 Signature in Cough Sounds
Bagad, P., Dalmia, A., Doshi, J., Nagrani, A., Bhamare, P., Mahale, A., Rane, S., Agarwal, N., & Panicker, R. (2020) · arXiv · doi:10.48550/arxiv.2009.08790 · Cited by 91
Sounds of COVID-19: exploring realistic performance of audio-based digital testing
Han, J., Xia, T., Spathis, D., Bondareva, E., Brown, C., Chauhan, J., Dang, T., Grammenos, A., Hasthanasombat, A., Floto, A., et al. (2022) · npj Digital Medicine · doi:10.1038/s41746-021-00553-x · Cited by 90
“In particular, advances in machine learning have demonstrated the potential of automated auscultation of respiratory sounds and brought about new possibilities for fully automated COVID-19 screening [12, 13, 14, 15, 16, 17, 18, 19].”
BackgroundSelf-citationSchuller, B. W., Batliner, A., Bergler, C., Mascolo, C., Han, J., Lefter, I., Kaya, H., Amiriparian, S., Baird, A., Stappen, L., et al. (2021) · Interspeech 2021 · doi:10.21437/interspeech.2021-19 · Cited by 89
Self-citationDetection of COVID-19 from voice, cough and breathing patterns: Dataset and preliminary results
Despotović, V., Ismael, M., Cornil, M., Call, R. M., & Fagherazzi, G. (2021) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2021.104944 · Cited by 106
A systematic review on AI/ML approaches against COVID-19 outbreak
Dogan, O., Tiwari, S., Jabbar, M. A., & Guggari, S. (2021) · Complex & Intelligent Systems · doi:10.1007/s40747-021-00424-8 · Cited by 87
“[304] COVID-19 diagnosis Boosting Trees, SVM Speech”
Lella, K. K., & Alphonse, P. J. A. (2021) · AIMS Public Health · doi:10.3934/publichealth.2021019 · Cited by 91
“Authors have illustrated the significant impact of RNN (Recurrent Neural Network) with the use of SSP (Speech Signal Processing) to detect the disease and specifically, this LSTM (Long Short-Term Memory) is used to evaluate the acoustic feature maps of patients’ cough, breathing, and voice, in the process of early scr…”
BackgroundMethodInfluentialReview on the Advancements of Stethoscope Types in Chest Auscultation
Seah, J. J., Zhao, J., Wang, D. Y., & Lee, H. P. (2023) · Diagnostics · doi:10.3390/diagnostics13091545 · Cited by 86
“described a simple binary machine learning classifier that can utilize cough and breathing sounds to understand and discern those from COVID-19 against sounds attributable to asthma or healthy controls [52].”
BackgroundExploring Automatic COVID-19 Diagnosis via Voice and Symptoms from Crowdsourced Data
Han, J., Brown, C., Chauhan, J., Grammenos, A., Hasthanasombat, A., Spathis, D., Xia, T., Cicuta, P., & Mascolo, C. (2021) · ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp39728.2021.9414576 · Cited by 84
“Besides, in [4], the analysis based on cough and breathing sounds, achieved the sensitivity of 69% and ROC-AUC of 80% , on a different sub-set of users though.”
BackgroundMethodInfluentialSelf-citationAutomatic cough classification for tuberculosis screening in a real-world environment
Pahar, M., Klopper, M., Reeve, B. W. P., Warren, R. M., Theron, G., & Niesler, T. (2021) · Physiological Measurement · doi:10.1088/1361-6579/ac2fb8 · Cited by 98
Automatic Lung Health Screening Using Respiratory Sounds
Mukherjee, H., Sreerama, P., Dhar, A., Obaidullah, S. M., Roy, K., Mahmud, M., & Santosh, K. (2021) · Journal of Medical Systems · doi:10.1007/s10916-020-01681-9 · Cited by 83
“As we observed that COVID-19 could possibly screened by analyzing respiratory sound [5], we are now extending our experiments on COVID-19 [21, 22].”
MethodThe Promise of AI in Detection, Diagnosis, and Epidemiology for Combating COVID-19: Beyond the Hype
Abdulkareem, M., & Petersen, S. E. (2021) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2021.652669 · Cited by 96
“Other models that use respiratory or coughing data for COVID-19 diagnosis can be found in Brown et al. (2020), Imran et al. (2020), and Jiang et al. (2020b).”
MethodBiometrics in the Era of COVID-19: Challenges and Opportunities
Gomez‐Barrero, M., Drozdowski, P., Rathgeb, C., Patino, J., Todisco, M., Nautsch, A., Damer, N., Priesnitz, J., Evans, N., & Busch, C. (2022) · IEEE Transactions on Technology and Society · doi:10.1109/tts.2022.3203571 · Cited by 92
“The data typically includes recordings of coughs [124], [125], [126], breathing sounds [127], [128] as well as speech excerpts [129].”
BackgroundA study of using cough sounds and deep neural networks for the early detection of Covid-19
Islam, R., Abdel‐Raheem, E., & Tarique, M. (2022) · Biomedical Engineering Advances · doi:10.1016/j.bea.2022.100025 · Cited by 93
“The system achieves even higher accuracy with the time-domain and mixed-feature vector than the works published in [ 31 , 34-35 , 64 ].”
MethodResultManshouri, N. (2021) · Cognitive Neurodynamics · doi:10.1007/s11571-021-09695-w · Cited by 85
“As a result of the lack of a wide range of COVID-19 studies based on cough sound classification, we do not find very accurate the detailed comparison of our proposed classification study with a few classification studies [9], [94], [45], [98] because each study focuses on different datasets.”
BackgroundResultInfluentialOESense: employing occlusion effect for in-ear human sensing
Ma, D., Ferlini, A., & Mascolo, C. (2021) · International Conference on Mobile Systems, Applications, and Services · doi:10.1145/3458864.3467680 · Cited by 82
“The detailed description of these features can be found in [23].”
BackgroundMethodSelf-citationDetection of COVID-19 from speech signal using bio-inspired based cepstral features
Dash, T. K., Mishra, S., Panda, G., & Satapathy, S. C. (2021) · Pattern Recognition · doi:10.1016/j.patcog.2021.107999 · Cited by 76
“It is re-ported that a maximum accuracy of 80% in Receiver Operator Characteristics Area Under Curve (AUC) [8] .”
BackgroundMethodInfluentialSait, U., Gokul Lal, K., V, Shivakumar, S., Kumar, T., Bhaumik, R., Prajapati, S. P., Bhalla, K., & Chakrapani, A. (2021) · Applied Soft Computing · doi:10.1016/j.asoc.2021.107522 · Cited by 72
“It was revealed that signal analysis proved the reliability of common abnormalities in breathing sounds such as crackles, vocal resonance and murmurs for Covid-19 detection [38,39].”
Hamdi, S., Oussalah, M., Moussaouı, A., & Saidi, M. (2022) · Journal of Intelligent Information Systems · doi:10.1007/s10844-022-00707-7 · Cited by 77
“Imran et al. (Imran et al., 2020) (Brown et al., 2020) from University of Cambridge proposed a large-scale crowdsourced dataset of respiratory sounds collected using either web or mobile apps.”
BackgroundResultInfluentialBelkacem, A. N., Ouhbi, S., Lakas, A., Benkhelifa, E., & Chen, C. (2021) · Frontiers in Medicine · doi:10.3389/fmed.2021.585578 · Cited by 62
“Brown et al. (34) have analyzed a large-scale crowdsourced dataset of respiratory sounds that were collected to assist in diagnosis of COVID-19.”
BackgroundDetecting COVID-19 from Breathing and Coughing Sounds using Deep Neural Networks
Nessiem, M. A., Mohamed, M. M., Coppock, H., Gaskell, A., & Schuller, B. W. (2021) · Proceedings - IEEE Symposium on Computer-Based Medical Systems · doi:10.1109/cbms52027.2021.00069 · Cited by 72
From Personalized Medicine to Population Health: A Survey of mHealth Sensing Techniques
Wang, Z., Xiong, H., Zhang, J., Yang, S., Boukhechba, M., Zhang, D., Barnes, L. E., & Dou, D. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3161046 · Cited by 76
“[119] proposed methodologies to detect diagnostic signs of COVID-19 from voice and coughs, which well differentiated COVID-19 from other viral infections.”
MethodTransformer-Based Multilingual Speech Emotion Recognition Using Data Augmentation and Feature Fusion
Al-onazi, B. B., Nauman, M. A., Jahangir, R., Malik, M. M., Alkhammash, E. H., & Elshewey, A. M. (2022) · Applied Sciences · doi:10.3390/app12189188 · Cited by 73
The Dawn of the Human-Machine Era: A forecast of new and emerging language technologies
Sayers, D., Sousa‐Silva, R., Höhn, S., Ahmedi, L., Allkivi-Metsoja, K., Anastasiou, D., Beňuš, Š., Bowker, L., Bytyçi, E., Catalá, A., et al. (2021) · doi:10.17011/jyx/reports/20210518/1 · Cited by 84
Dash, T. K., Chakraborty, C., Mahapatra, S., & Panda, G. (2022) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2022.3197910 · Cited by 68
“These features are combined with an LGM classifier and three baseline classifiers such as Random Forest (RF) [9], SVM [10], [11], and K-Nearest Neighbor (KNN) [44] used for the speech classification task.”
BackgroundMethodInfluentialCOVID-19 in the Age of Artificial Intelligence: A Comprehensive Review
Rasheed, J., Jamıl, A., Hameed, A. A., Al-Turjman, F. M., & Rasheed, A. (2021) · Interdisciplinary Sciences Computational Life Sciences · doi:10.1007/s12539-021-00431-w · Cited by 68
Comprehensive Survey of Using Machine Learning in the COVID-19 Pandemic
El-Rashidy, N. M., Abdelrazek, S. M., Abuhmed, T., Amer, E., Ali, F., Hu, J. W., & El–Sappagh, S. (2021) · Diagnostics · doi:10.3390/diagnostics11071155 · Cited by 70
“In another dataset [76], th authors collected 7000 sound samples hat included 200 confirmed COVID-19 subj s.”
BackgroundMethodInfluentialAI-Based human audio processing for COVID-19: A comprehensive overview
Deshpande, G., Batliner, A. M., & Schuller, B. W. (2021) · Pattern Recognition · doi:10.1016/j.patcog.2021.108289 · Cited by 69
“Especially Brown et al. [18] reported that breathing signals are better suited for distinguishing C19 positive users from C19 negative users having asthma and cough.”
InfluentialDetection of COVID-19 in smartphone-based breathing recordings: A pre-screening deep learning tool
Alkhodari, M., & Khandoker, A. H. (2022) · PLoS ONE · doi:10.1371/journal.pone.0262448 · Cited by 63
“Biological respiratory signals, such as coughing and breathing sounds, could be another promising tool to indicate the existence of the viral infection [29], as these signals have a direct connection with lungs.”
BackgroundSerrurier, A., Neuschaefer‐Rube, C., & Röhrig, R. (2022) · Sensors · doi:10.3390/s22082896 · Cited by 65
“[70,79,92–97] X X [98] X X [87] X X [99] X X [88] X X [33,100] X X [100] X X [29] X X X [30,101–106] X X [59,85,107] X [59,84,85,89] X [85,90,91,108] X [105,109] X X [110] X X [83,111] X [112] X X X [81] X X [113–115] X X [80] X [58] X X [116] X X X [100] X X [86] X X X [105] X X [59] X [85] X [105,109] X X [85] X [10…”
Studying the Similarity of COVID-19 Sounds based on Correlation Analysis of MFCC
Alsabek, M. B., Shahin, I. M. A., & Hassan, A. (2020) · International Conference on Communications, Computing, Cybersecurity, and Informatics (CCCI) · doi:10.1109/ccci49893.2020.9256700 · Cited by 57
“However, we only focused on extracting MFCCs since they are considered the most important features in distinguishing COVID from non-COVID sounds [5].”
MethodInfluentialCao, L., & Liu, Q. (2024) · ACM Computing Surveys · doi:10.1145/3686150 · Cited by 61
“Infection diagnosis Shallow learners [25, 26, 43, 253], CNN and RNN variants like LSTM and GRU, and pretrained CNNbased image nets like ResNet, MobileNetV2 and SqueezeNet, etc.”
COVID-19 detection from audio: seven grains of salt
Coppock, H., Jones, L., Kiskin, I., & Schuller, B. (2021) · The Lancet Digital Health · doi:10.1016/s2589-7500(21)00141-2 · Cited by 55
“…of the labels used in the COVID-19 audio datasets collected so far is questionable because most datasets allow participants to self-report their COVID-19 status and fail to record the type of test participants had; 2,8,9 polymerase chain reaction (PCR) or, the less accurate, lateral flow test.”
Hemdan, E. E., El‐Shafai, W., & Sayed, A. (2022) · Journal of Ambient Intelligence and Humanized Computing · doi:10.1007/s12652-022-03732-0 · Cited by 58
Sunitha, G., Arunachalam, R., Abd‐Elnaby, M., Eid, M. M. A., & Rashed, A. N. Z. (2022) · International Journal of Imaging Systems and Technology · doi:10.1002/ima.22749 · Cited by 55
“In Reference [23], this study used a significant dataset for COVID-19 based on infected people's breath and cough sounds.”
MethodLonini, L., Shawen, N., Botonis, O. K., Fanton, M., Jayaraman, C., Mummidisetty, C. K., Shin, S. Y., Rushin, C. C., Jenz, S. T., Xu, S., et al. (2021) · IEEE Journal of Translational Engineering in Health and Medicine · doi:10.1109/jtehm.2021.3058841 · Cited by 54
“Recent studies showed that crowdsourced data from smart-phones or consumer-grade wearables on either respiratory, cardiac or cough sounds [17], [32]–[35] could potentially be used to develop biomarkers to predict the onset or detect the presence of COVID-19.”
Ponomarchuk, A., Burenko, I., Malkin, E., Nazarov, I., Kokh, V., Avetisian, M., & Евгеньевич, Ж. Л. (2022) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2022.3142514 · Cited by 54
“The ablation study with stratified grouped cross-validation in [30] demonstrates that coughnon-cough pretraining and ensembling contribute positively to the performance of their final stacked composite COVIDclassifier, consisting of the deep ResNet-18 classifier and shallow models on acoustic features from [33].”
BackgroundMethodInfluentialParticle Swarm Optimization-Based Extreme Learning Machine for COVID-19 Detection
Albadr, M. A. A., Tiun, S. B., Ayob, M. B., & AL‐Dhief, F. T. (2022) · Cognitive Computation · doi:10.1007/s12559-022-10063-x · Cited by 55
“The authors in [20] proposed a COVID-19 detection sys-tem by implementing several handcrafted features and logistic regression (LR) classifiers.”
MethodAly, M., Rahouma, K. H., & Ramzy, S. M. (2021) · Alexandria Engineering Journal · doi:10.1016/j.aej.2021.08.070 · Cited by 52
“It has not been a long time since researchers from Cambridge University [12] established a COVID-19 sound collection framework [13] to collect coughs and breathing sounds.”
InfluentialThe safety of dorsal penile nerve block for neonatal circumcision.
Fontaine, P., Dittberner, D., & Scheltema, K. E. (1994) · PubMed · Cited by 50
A literature review on COVID-19 disease diagnosis from respiratory sound data
Lella, K. K., & Alphonse, P. J. A. (2021) · AIMS bioengineering · doi:10.3934/bioeng.2021013 · Cited by 50
“Data review of a large crowdsourced respiratory speech/sounds dataset that has been obtained to accurately diagnose COVID positive cases was mentioned in [9].”
BackgroundMethodInfluentialCao, L., & Liu, Q. (2022) · doi:10.1101/2022.08.22.22279022 · Cited by 49
AI diagnosis of heart sounds differentiated with super StethoScope
Ogawa, S., Namino, F., Mori, T., Sato, G., Yamakawa, T., & Saito, S. (2023) · Journal of Cardiology · doi:10.1016/j.jjcc.2023.09.007 · Cited by 48
Manzella, F., Pagliarini, G., Sciavicco, G., & Stan, I. E. (2023) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2022.102486 · Cited by 57
“For the datasets, CAM stands for Cambridge [8], V for Virufy [9], CO for COUGHVID [10], CW for Coswara [11], NO for NoCoCoDa [12], C19S for COVID-19 Sounds [13], DI for DiCOVA [14], CCS and CSS for ComParE2021 CCS and CSS [15].”
InfluentialXia, T., Han, J., & Mascolo, C. (2022) · Experimental Biology and Medicine · doi:10.1177/15353702221115428 · Cited by 71
“44,45 There are many existing libraries that can be leveraged to automatically extract those features from raw signals, among which Librosa is a well-known Python-based programming tool.”
BackgroundSelf-citationLella, K. K., & Alphonse, P. J. A. (2022) · The European Physical Journal Special Topics · doi:10.1140/epjs/s11734-022-00432-w · Cited by 52
“SVM with PCA [24] Crowdsourced COVID-19”
BackgroundMethodInfluentialSantosh, K., Rasmussen, N. R., Mamun, M., & Aryal, S. (2022) · PeerJ Computer Science · doi:10.7717/peerj-cs.958 · Cited by 51
“Performance Authors (year) Data type (sample size) ACC AUC SEN SPEC Han et al. (2021) Multiple data types* (828) – 0.79 0.62 0.74 Brown et al. (2020) Multiple data types* (430) – 0.80 0.69 – Vrindavanam et al. (2021) Cough sounds (150) 0.84 0.88 0.81 – Anupam et al. (2021) Cough sounds (640) 0.97…”
MethodInfluentialDetection of Covid-19 Through the Analysis of Vocal Fold Oscillations
Ismail, M. A., Deshmukh, S., & Singh, R. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9414201 · Cited by 46
“Researchers have also used crowd-sourced data [5, 6] with data-driven end-to-end deep learning methods for this purpose.”
BackgroundDiagnosis of COVID-19 and non-COVID-19 patients by classifying only a single cough sound
Melek, M. (2021) · Neural Computing and Applications · doi:10.1007/s00521-021-06346-3 · Cited by 45
Automatic Detection of COVID-19 Based on Short-Duration Acoustic Smartphone Speech Analysis
Stasak, B., Huang, Z., Razavi, S., Joachim, D., & Epps, J. (2021) · Journal of Healthcare Informatics Research · doi:10.1007/s41666-020-00090-4 · Cited by 45
“However, unlike [4, 20], SHC audio recordings were collected less invasively (i.e., collected over short chosen period, chosen location/time, under naturalistic conditions) and speech data was automatically analyzed using only 6 s per recording (i.e., did not require searching hours of speech data…”
BackgroundMethodResultInfluentialDeep Neural Network-Based Respiratory Pathology Classification Using Cough Sounds
Balamurali, B. T., Hee, H. I., Kapoor, S., Teoh, O. H., Teng, S. S., Lee, K. P., Herremans, D., & Chen, J. (2021) · Sensors · doi:10.3390/s21165555 · Cited by 45
“Features were extracted (both handcrafted and transfer learned) from a crowd-sourced database containing breathing and cough sounds [18] and were used to train a support vector machine and ensemble classifiers to screen COVID-19 individuals from healthy controls.”
MethodSanabria, A. R., Zambonelli, F., Dobson, S., & Ye, J. (2021) · Pervasive and Mobile Computing · doi:10.1016/j.pmcj.2021.101477 · Cited by 43
“With pervasive availability of smart sensing technologies such as wearables and environmental sensors [30], we have witnessed increasing deployment of HAR systems in real-world environments [8].”
BackgroundVocalsound: A Dataset for Improving Human Vocal Sounds Recognition
Gong, Y., Yu, J., & Glass, J. (2022) · ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp43922.2022.9746828 · Cited by 43
“Recently, there are a few efforts to collect cough samples for building COVID-19 classification models [12, 13, 14, 15, 16, 17, 18].”
Deep-learning Based Approach to Identify Covid-19
Feng, K., He, F., Steinmann, J., & Demirkiran, I. (2021) · IEEE SoutheastCon · doi:10.1109/southeastcon45413.2021.9401826 · Cited by 50
“Research from the Massachusetts Institute of Technology [15] also shows that machine learning has the potential to detect COVID-19.”
BackgroundUpdate in COVID-19 in the intensive care unit from the 2020 HELLENIC Athens International symposium
Rello, J., Belliato, M., Dimopoulos, M. A., Giamarellos‐Bourboulis, E. J., Jaksic, V., Martín‐Loeches, I., Mporas, I., Pelosi, P. P., Poulakou, G., Pournaras, S., et al. (2020) · Anaesthesia Critical Care & Pain Medicine · doi:10.1016/j.accpm.2020.10.008 · Cited by 51
COVID-19 Detection From Respiratory Sounds With Hierarchical Spectrogram Transformers
Aytekin, I., Dalmaz, O., Gönç, K., Ankışhan, H., Sarıtaş, E. Ü., Bağcı, U., Celik, H., & Çukur, T. (2023) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2023.3339700 · Cited by 47
“SVM: A total of 733 hand-crafted and data-driven features were taken as inputs [38].”
BackgroundMethodInfluentialExploring Self-Supervised Representation Ensembles for COVID-19 Cough Classification
Xue, H., & Salim, F. D. (2021) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3447548.3467263 · Cited by 40
“[5] combines hand-crafted features and VGGish [15] (pre-trained on Audioset [10]) extracted deep learning features.”
BackgroundMethodInfluentialÇelik, G. (2023) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2023.107153 · Cited by 41
“[30], Covid-19 detection was carried out through ML techniques using sample data obtained from cough and breath.”
Face mask recognition from audio: The MASC database and an overview on the mask challenge
Mohamed, M. M., Nessiem, M. A., Batliner, A. M., Bergler, C., Hantke, S., Schmitt, M., Baird, A., Mallol-Ragolta, A., Karas, V., Amiriparian, S., et al. (2021) · Pattern Recognition · doi:10.1016/j.patcog.2021.108361 · Cited by 41
“Examples of speech-based applications can be found in [30,31] .”
Liu, S., Han, J., Puyal, E. L., Kontaxis, S., Sun, S., Locatelli, P., Dineley, J., Pokorny, F. B., Costa, G. D., Leocani, L., et al. (2021) · Pattern Recognition · doi:10.1016/j.patcog.2021.108403 · Cited by 43
“Recent work has investigated data streams that could potentially be used to detect COVID-19 and can be easily captured using smart devices and wearable equipment [17,18] , including recordings of coughing and breathing [19] and speech signals [20,21] .”
Self-citationSharma, N. K., Chetupalli, S. R., Bhattacharya, D., Dutta, D., Mote, P., & Ganapathy, S. (2022) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp43922.2022.9747188 · Cited by 44
“This has garnered interest in the speech and audio research community, and there have been several studies [7, 8, 9, 10] gathering insights on the possibility of acoustics based COVID-19 diagnosis.”
BackgroundCoppock, H., Nicholson, G., Kiskin, I., Koutra, V., Baker, K., Budd, J., Payne, R., Karoune, E., Hurley, D., Titcomb, A., et al. (2024) · Nature Machine Intelligence · doi:10.1038/s42256-023-00773-8 · Cited by 50
Multi-Modal Point-of-Care Diagnostics for COVID-19 Based on Acoustics and Symptoms
Chetupalli, S. R., Krishnan, P., Sharma, N. K., Muguli, A., Kumar, R., Nanda, V., Pinto, L., Ghosh, P., & Ganapathy, S. (2023) · IEEE Journal of Translational Engineering in Health and Medicine · doi:10.1109/jtehm.2023.3250700 · Cited by 40
“We compared the performance of the proposed multi-modal approach with (i, ii) the approaches in [29] and [30] which use the breathing and cough modalities and (iii) the approach in [39] using speech and symptoms.”
InfluentialDang, T., Han, J., Xia, T., Spathis, D., Bondareva, E., Siegele-Brown, C., Chauhan, J., Grammenos, A., Hasthanasombat, A., Floto, R., et al. (2022) · Journal of Medical Internet Research · doi:10.2196/37004 · Cited by 35
“The existing literature has mainly investigated the information content of different audio modalities (eg, cough, breathing, and voice) [8-12] and the power of various machine learning techniques, especially deep learning for COVID-19 detection [10,13-19].”
BackgroundSelf-citationDar, J. A., Srivastava, K. K., & Lone, S. A. (2022) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2022.106123 · Cited by 37
“The utilization of human engendered audio as biomarker for different illnesses affords massive possible for premature analysis [17].”
Exploring racial and gender disparities in voice biometrics
Chen, X., Li, Z., Setlur, S., & Xu, W. (2022) · Scientific Reports · doi:10.1038/s41598-022-06673-y · Cited by 36
QUCoughScope: An Intelligent Application to Detect COVID-19 Patients Using Cough and Breath Sounds
Rahman, T., Ibtehaz, N., Khandakar, A. A., Hossain, M. S. A., Mekki, Y. M. S., Ezeddin, M., Bhuiyan, E., Ayari, M. A., Tahir, A., Qiblawey, Y., et al. (2022) · Diagnostics · doi:10.3390/diagnostics12040920 · Cited by 39
“Cambridge dataset: The Cambridge dataset was designed for developing a diagnostic tool for COVID-19 based on cough and breath sounds [55].”
BackgroundMethodInfluentialVoice analytics in the wild: Validity and predictive accuracy of common audio-recording devices
Busquet, F., Efthymiou, F., & Hildebrand, C. (2023) · Behavior Research Methods · doi:10.3758/s13428-023-02139-9 · Cited by 41
“…et al., 2016a, b), assess early onset of physical health issues (Kelchner et al., 2010), mental disorders (Faurholt-Jepsen et al., 2016), early diagnosis of the Parkinson’s disease (Brabenec et al., 2017; Erdogdu Sakar et al., 2017), or a COVID-19 infection (Brown et al., 2020; Han et al., 2021).”
BackgroundEnsemble multimodal deep learning for early diagnosis and accurate classification of COVID-19
Kumar, S., Gupta, S. K., Kumar, V., Kumar, M., Chaube, M. K., & Naik, N. S. (2022) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2022.108396 · Cited by 36
(Un)fair Devices: Moving Beyond AI Accuracy in Personal Sensing
Yfantidou, S., Constantinides, M., Spathis, D., Vakali, A. I., Quercia, D., & Kawsar, F. (2026) · ACM Journal on Responsible Computing · doi:10.1145/3796223 · Cited by 33
“Additionally, the ML tasks addressed by the personal devices community diverge from those typically explored in fairness research. characteristic and refrained from providing performance breakdowns conditioned on demographics, adhering solely to aggregate evaluations [4].”
Self-citationTBscreen: A passive cough classifier for tuberculosis screening with a controlled dataset
Sharma, M., Nduba, V., Njagi, L. N., Murithi, W. B., Mwongera, Z., Hawn, T. R., Patel, S., & Horne, D. J. D. L. (2024) · Science Advances · doi:10.1126/sciadv.adi0282 · Cited by 31
Deep Learning Audio Spectrograms Processing to the Early COVID-19 Detection
Rodriguez, C. R., Angeles, D., Chafloque, R., Kaseng, F., & Pandey, B. (2020) · International Conference on Computational Intelligence and Communication Networks (CICN) · doi:10.1109/cicn49253.2020.9242583 · Cited by 31
Chen, Z., Li, M., Wang, R., Sun, W., Liu, J., Li, H., Wang, T., Lian, Y., Zhang, J., & Wang, X. (2022) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2022.104078 · Cited by 32
Audio, Speech, Language, & Signal Processing for COVID-19: A Comprehensive OverviewPreprint
Deshpande, G., & Schuller, B. W. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.14445 · Cited by 33
“However, in the work presented in [10], only cough and breathing sounds are considered.”
BackgroundMethodInfluentialHuman-centred artificial intelligence for mobile health sensing: challenges and opportunities
Dang, T., Spathis, D., Ghosh, A., & Mascolo, C. (2023) · Royal Society Open Science · doi:10.1098/rsos.230806 · Cited by 30
“COVID-19 has been studied using different audio sounds, including voice, breathing and cough sounds [10,29,79,80].”
BackgroundSelf-citationOptimizing MFCC parameters for the automatic detection of respiratory diseases
Yan, Y., Simons, S. O., van Bemmel, L., Reinders, L. G., Franssen, F. M. E., & Urovi, V. (2024) · Applied Acoustics · doi:10.1016/j.apacoust.2024.110299 · Cited by 33
“Initially proposed by Davis and Mermelstein in the 1980s [16], it has been utilized in various audio recognition tasks, including speech emotion recognition [17], speaker recognition [18] and pathology voice identification [19] in traditional pipeline systems.”
Battling COVID-19 using machine learning: A review
Chadaga, K., Prabhu, S., Vivekananda Bhat, K., Niranjana, S., & Umakanth, S. (2021) · Cogent Engineering · doi:10.1080/23311916.2021.1958666 · Cited by 35
“A system which used both coughing and breathing was created by a research group at the University of Cambridge (Brown et al., 2020).”
MethodRobotti, C. A., Costantini, G., Saggio, G., Cesarini, V., Calastri, A., Maiorano, E., Piloni, D., Perrone, T., Sabatini, U., Ferretti, V. V., et al. (2021) · Journal of Voice · doi:10.1016/j.jvoice.2021.11.004 · Cited by 35
“48,49 However, comparisons with literature appear scarcely feasible, primarily for methodological reasons.”
BackgroundPrandi, C., Nisi, V., Ribeiro, M., & Nunes, N. (2021) · Journal Of Big Data · doi:10.1186/s40537-021-00442-w · Cited by 29
“During the COVID-19 pandemic, smartphone sensors and crowdsourcing have been employed to collect respiratory sound data [79], and contact tracing data [80].”
MethodDeep Learning with hyper-parameter tuning for COVID-19 Cough Detection
Rao, S., Narayanaswamy, V., Esposito, M., Thiagarajan, J. J., & Spanias, A. (2021) · International Conference on Information, Intelligence, Systems & Applications (IISA) · doi:10.1109/iisa52424.2021.9555564 · Cited by 34
Metadata
- Authors
- Brown, C., Chauhan, J., Grammenos, A., Han, J., Hasthanasombat, A., Spathis, D., Xia, T., Cicuta, P., & Mascolo, C.
- Affiliations
- University of Cambridge (GB)
- University of Cambridge, Cambridge, United Kingdom
- ORCID
- Venue
- Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining
- Publisher
- ACM
- Published
- 2020-08-20
- Pages
- 3474-3484
- OpenAlex
- W3035378948
- Cited by
- 447
- References
- 34
- Citations by year
- 2026: 17 · 2025: 26 · 2024: 49 · 2023: 78 · 2022: 124 · 2021: 136 · 2020: 15
- FWCI
- 253.82
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- Cambridge University Engineering Department Publications Database (submittedVersion)
- Apollo (University of Cambridge) (submittedVersion, cc-by)
- arXiv (Cornell University)
- Apollo (cc-by)
- Funders
- Cystic Fibrosis Trust
- European Research Council (833296)
- Topics
- Health Sciences › Medicine › Pulmonary and Respiratory Medicine › Phonocardiography and Auscultation Techniques
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › COVID-19 diagnosis using AI
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Keywords
- COVID-19 diagnosis, lung sounds, cough, digital stethoscope, asthma, binary classification, prescreening
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.