Multi-Modal Point-of-Care Diagnostics for COVID-19 Based on Acoustics and Symptoms

The research direction of identifying acoustic bio-markers of respiratory diseases has received renewed interest following the onset of COVID-19 pandemic. In this paper, we design an approach to COVID-19 diagnostic using crowd-sourced multi-modal data. The data resource, consisting of acoustic signals like cough, breathing, and speech signals, along with the data of symptoms, are recorded using a web-application over a period of ten months. We investigate the use of statistical descriptors of simple time-frequency features for acoustic signals and binary features for the presence of symptoms. Unlike previous works, we primarily focus on the application of simple linear classifiers like logistic regression and support vector machines for acoustic data while decision tree models are employed on the symptoms data. We show that a multimodal integration of acoustics and symptoms classifiers achieves an area-under-curve (AUC) of 92.40, a significant improvement over any individual modality. Several ablation experiments are also provided which highlight the acoustic and symptom dimensions that are important for the task of COVID-19 diagnostics.

[1]  Sriram Chellappan,et al.  TussisWatch: A Smart-Phone System to Identify Cough Episodes as Early Symptoms of Chronic Obstructive Pulmonary Disease and Congestive Heart Failure , 2019, IEEE Journal of Biomedical and Health Informatics.

[2]  Victor M Corman,et al.  Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR , 2020, Euro surveillance : bulletin Europeen sur les maladies transmissibles = European communicable disease bulletin.

[3]  H. B. Curry.  The method of steepest descent for non-linear minimization problems , 1944 .

[4]  Muhammad Nabeel,et al.  AI4COVID-19: AI enabled preliminary diagnosis for COVID-19 from cough samples via an app , 2020, Informatics in Medicine Unlocked.

[5]  Vinayak Swarnkar,et al.  Automatic Identification of Wet and Dry Cough in Pediatric Patients with Respiratory Diseases , 2013, Annals of Biomedical Engineering.

[6]  Characteristics of SARS-CoV-2 and COVID-19 , 2020, Nature reviews. Microbiology.

[7]  Brian Subirana,et al.  COVID-19 Artificial Intelligence Diagnosis Using Only Cough Recordings , 2020, IEEE Open Journal of Engineering in Medicine and Biology.

[8]  Vinayak Swarnkar,et al.  Cough Sound Analysis Can Rapidly Diagnose Childhood Pneumonia , 2013, Annals of Biomedical Engineering.

[9]  Ahmed Ghoneim,et al.  Exploring the Use of Artificial Intelligence Techniques to Detect the Presence of Coronavirus Covid-19 Through Speech and Voice Analysis , 2021, IEEE Access.

[10]  Ke Feng,et al.  Deep-learning Based Approach to Identify Covid-19 , 2021, SoutheastCon 2021.

[11]  Kun Qian,et al.  Recent Advances in Computer Audition for Diagnosing COVID-19: An Overview , 2020, ArXiv.

[12]  Yang Liu,et al.  An introduction to decision tree modeling , 2004 .

[13]  Björn Schuller,et al.  Opensmile: the munich versatile and fast open-source audio feature extractor , 2010, ACM Multimedia.

[14]  David Atienza,et al.  The COUGHVID crowdsourcing dataset, a corpus for the study of large-scale cough analysis algorithms , 2021, Scientific data.

[15]  R. Peeling,et al.  Scaling up COVID-19 rapid antigen tests: promises and challenges , 2021, The Lancet Infectious Diseases.

[16]  Eduardo Coutinho,et al.  The INTERSPEECH 2016 Computational Paralinguistics Challenge: Deception, Sincerity & Native Language , 2016, INTERSPEECH.

[17]  Hector Rodriguez-Villalobos,et al.  Low performance of rapid antigen detection test as frontline testing for COVID-19 diagnosis , 2020, Journal of Clinical Virology.

[18]  Srikanth Raj Chetupalli,et al.  Coswara - A Database of Breathing, Cough, and Voice Sounds for COVID-19 Diagnosis , 2020, INTERSPEECH.

[19]  J. Borghi,et al.  Epidemiology of COVID‐19: A systematic review and meta‐analysis of clinical characteristics, risk factors, and outcomes , 2020, Journal of medical virology.

[20]  Mehrin Kiani,et al.  A Generic Deep Learning Based Cough Analysis System From Clinically Validated Samples for Point-of-Need Covid-19 Test and Severity Levels , 2021, IEEE Transactions on Services Computing.

[21]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[22]  Renard Xaviero Adhi Pramono,et al.  A Cough-Based Algorithm for Automatic Diagnosis of Pertussis , 2016, PloS one.

[23]  N. Shomron,et al.  Machine learning-based prediction of COVID-19 diagnosis based on symptoms , 2021, npj Digital Medicine.

[24]  P D van Helden,et al.  Detection of tuberculosis by automatic cough sound analysis , 2018, Physiological measurement.

[25]  K. Scherer,et al.  On the Acoustics of Emotion in Audio: What Speech, Music, and Sound have in Common , 2013, Front. Psychol..

[26]  M. Jorge Cardoso,et al.  Real-time tracking of self-reported symptoms to predict potential COVID-19 , 2020, Nature Medicine.

[27]  U. Abeyratne,et al.  Diagnosing Chronic Obstructive Airway Disease on a Smartphone Using Patient-Reported Symptoms and Cough Analysis: Diagnostic Accuracy Study , 2020, JMIR formative research.

[28]  Dorien Herremans,et al.  Development of Machine Learning for Asthmatic and Healthy Voluntary Cough Sounds: A Proof of Concept Study , 2019, Applied Sciences.

[29]  Y. Hu,et al.  Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China , 2020, The Lancet.

[30]  A. Haq,et al.  Wavelet-Based Cough Signal Decomposition for Multimodal Classification , 2020, 2020 17th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP).

[31]  Cecilia Mascolo,et al.  Exploring Automatic Diagnosis of COVID-19 from Crowdsourced Respiratory Sound Data , 2020, KDD.

[32]  Alice Baird,et al.  End-to-end convolutional neural network enables COVID-19 detection from breath and cough audio: a pilot study , 2021, BMJ Innovations.