A systematic review on cough sound analysis for Covid-19 diagnosis and screening: is my cough sound COVID-19?

Longitudinal voice monitoring in a decentralized Bring Your Own Device trial for respiratory illness detection

The Acute Respiratory Illness Surveillance (AcRIS) Study was a low-interventional trial that examined voice changes with respiratory illnesses. This longitudinal trial was the first of its kind, conducted in a fully decentralized manner via a Bring Your Own Device mobile application. The app enabled social-media-based recruitment, remote consent, at-home sample collection, and daily remote voice and symptom capture in real-world settings. From April 2021 to April 2022, the trial enrolled 9151 participants, followed for up to eight weeks. Despite mild symptoms experienced by reverse transcription polymerase chain reaction (RT-PCR) positive participants, two machine learning algorithms developed to screen respiratory illnesses reached the pre-specified success criteria. Algorithm testing on independent cohorts demonstrated that the algorithm's sensitivity increased as symptoms increased, while specificity remained consistent. Study findings suggest voice features can identify individuals with viral respiratory illnesses and provide valuable insights into fully decentralized clinical trials design, operation, and adoption (study registered at ClinicalTrials.gov (NCT04748445) on 5 February 2021).

Acoustic biomarkers in asthma: a systematic review

OBJECTIVE

Current monitoring methods of asthma, such as peak expiratory flow testing, have important limitations. The emergence of automated acoustic sound analysis, capturing cough, wheeze, and inhaler use, offers a promising avenue for improving asthma diagnosis and monitoring. This systematic review evaluated the validity of acoustic biomarkers in supporting the diagnosis of asthma and its monitoring.

DATA SOURCES

A search was performed using two databases (PubMed and Embase) for all relevant studies published before November 2023.

STUDY SELECTION

27 studies were included for analysis. Eligible studies focused on acoustic signals as digital biomarkers in asthma, utilizing recording devices to register or analyze sound.

RESULTS

Various respiratory acoustic signal types were analyzed, with cough and wheeze being predominant. Data collection methods included smartphones, custom sensors and digital stethoscopes. Across all studies, automated acoustic algorithms achieved average accuracy of cough and wheeze detection of 88.7% (range: 61.0 - 100.0%) with a median of 92.0%. The sensitivity of sound detection ranged from 54.0 to 100.0%, with a median of 90.3%; specificity ranged from 67.0 to 99.7%, with a median of 95.0%. Moreover, 70.4% (19/27) studies had a risk of bias identified.

CONCLUSIONS

This systematic review establishes the promising role of acoustic biomarkers, particularly cough and wheeze, in supporting the diagnosis of asthma and monitoring. The evidence suggests the potential for clinical integration of acoustic biomarkers, emphasizing the need for further validation in larger, clinically-diverse populations.

Update on Patient Self-Testing with Portable and Wearable Devices: Advantages and Limitations

Laboratory medicine has undergone a deep and multifaceted revolution in the course of human history, in both organizational and technical terms. Over the past century, there has been a growing recognition of the need to centralize numerous diagnostic activities, often similar or identical but located in different clinical departments, into a common environment (i.e., the medical laboratory service), followed by a progressive centralization of tests from smaller laboratories into larger diagnostic facilities. Nevertheless, the numerous technological advances that emerged at the beginning of the new millennium have helped to create a new testing culture characterized by a countervailing trend of decentralization of some tests closer to patients and caregivers. The forces that have driven this (centripetal) counter-revolution essentially include a few key concepts, namely "home testing", "portable or even wearable devices" and "remote patient monitoring". By their very nature, laboratory medical services and remote patient testing/monitoring are not contradictory, but may well coexist, with the choice of one or the other depending on the demographic and clinical characteristics of the patient, the type of analytical procedure and the logistics and local organization of the care system. Therefore, this article aims to provide a general overview of patient self-testing, with a particular focus on portable and wearable (including implantable) devices.

Recognition of Patient Gender: A Machine Learning Preliminary Analysis Using Heart Sounds from Children and Adolescents

Research has shown that X-rays and fundus images can classify gender, age group, and race, raising concerns about bias and fairness in medical AI applications. However, the potential for physiological sounds to classify sociodemographic traits has not been investigated. Exploring this gap is crucial for understanding the implications and ensuring fairness in the field of medical sound analysis. We aimed to develop classifiers to determine gender (men/women) based on heart sound recordings and using machine learning (ML). Data-driven ML analysis. We utilized the open-access CirCor DigiScope Phonocardiogram Dataset obtained from cardiac screening programs in Brazil. Volunteers < 21 years of age. Each participant completed a questionnaire and underwent a clinical examination, including electronic auscultation at four cardiac points: aortic (AV), mitral (MV), pulmonary (PV), and tricuspid (TV). We used Mel-frequency cepstral coefficients (MFCCs) to develop the ML classifiers. From each patient and from each auscultation sound recording, we extracted 10 MFCCs. In sensitivity analysis, we additionally extracted 20, 30, 40, and 50 MFCCs. The most effective gender classifier was developed using PV recordings (AUC ROC = 70.3%). The second best came from MV recordings (AUC ROC = 58.8%). AV and TV recordings produced classifiers with an AUC ROC of 56.4% and 56.1%, respectively. Using more MFCCs did not substantially improve the classifiers. It is possible to classify between males and females using phonocardiogram data. As health-related audio recordings become more prominent in ML applications, research is required to explore if these recordings contain signals that could distinguish sociodemographic features.

A smartphone-based application for cough counting in patients with acute asthma exacerbation

BACKGROUND

While tools exist for objective cough counting in clinical studies, there is no available tool for objective cough measurement in clinical practice. An artificial intelligence (AI)-based cough count system was recently developed that quantifies cough sounds collected through a smartphone application. In this prospective study, this AI-based cough algorithm was applied among real-world patients with an acute exacerbation of asthma.

METHODS

Patients with an acute asthma exacerbation recorded their cough sounds for 7 days (2 consecutive hours during awake time and 5 consecutive hours during sleep) using CoughyTM smartphone application. During the study period, subjects received systemic corticosteroids and bronchodilator to control asthma. Coughs collected by application were counted by both the AI algorithm and two human experts. Subjects also provided self-measured peak expiratory flow rate (PEFR) and completed other outcome assessments [e.g., cough symptom visual analogue scale (CS-VAS), awake frequency, salbutamol use] to investigate the correlation between cough and other parameters.

RESULTS

A total of 1,417.6 h of cough recordings were obtained from 24 asthmatics (median age =39 years). Cough counts by AI were strongly correlated with manual cough counts during sleep time (rho =0.908, P<0.001) and awake time (rho =0.847, P<0.001). Sleep time cough counts were moderately to strongly correlated with CS-VAS (rho =0.339, P<0.001), the frequency of waking up (rho =0.462, P<0.001), and salbutamol use at night (rho =0.243, P<0.001). Weak-to-moderate correlations were found between awake time cough counts and CS-VAS (rho =0.313, P<0.001), the degree of activity limitation (rho =0.169, P=0.005), and salbutamol use at awake time (rho =0.276, P<0.001). Neither awake time nor sleep time cough counts were significantly correlated with PEFR.

CONCLUSIONS

The strong correlation between cough counts using the AI-based algorithm and human experts, and other indicators of patient health status provides evidence of the validity of this AI algorithm for use in asthma patients experiencing an acute exacerbation. Study findings suggest that CoughyTM could be a novel solution for objectively monitoring cough in a clinical setting.

The value of chest X-ray and CT severity scoring systems in the diagnosis of COVID-19: A review

Coronavirus disease 2019 (COVID-19) is caused by a coronavirus family member known as severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2). The main laboratory test to confirm the quick diagnosis of COVID-19 infection is reverse transcription-polymerase chain reaction (RT-PCR) based on nasal or throat swab sampling. A small percentage of false-negative RT-PCR results have been reported. The RT-PCR test has a sensitivity of 50-72%, which could be attributed to a low viral load in test specimens or laboratory errors. In contrast, chest CT has shown 56-98% of sensitivity in diagnosing COVID-19 at initial presentation and has been suggested to be useful in correcting false negatives from RT-PCR. Chest X-rays and CT scans have been proposed to predict COVID-19 disease severity by displaying the score of lung involvement and thus providing information about the diagnosis and prognosis of COVID-19 infection. As a result, the current study provides a comprehensive overview of the utility of the severity score index using X-rays and CT scans in diagnosing patients with COVID-19 when compared to RT-PCR.

Cough Audio Analysis for COVID-19 Diagnosis

Humanity has suffered catastrophically due to the COVID-19 pandemic. One of the most reliable diagnoses of COVID-19 is RT-PCR (reverse-transcription polymer chain reaction) testing. This method, however, has its limitations. It is time consuming and requires scalability. This research work carries out a preliminary prognosis of COVID-19, which is scalable and less time consuming. The research carried out a competitive analysis of four machine-learning models namely, Multilayer Perceptron, Convolutional Neural Networks, Recurrent Neural Networks with Long Short-Term Memory, and VGG-19 with Support Vector Machines. Out of these models, Multilayer Perceptron outperformed with higher specificity of 94.5% and accuracy of 96.8%. The results show that Multilayer Perceptron was able to distinguish between positive and negative COVID-19 coughs by a robust feature embedding technique.

Coronavirus diagnosis using cough sounds: Artificial intelligence approaches

Introduction

The Coronavirus disease 2019 (COVID-19) pandemic has caused irreparable damage to the world. In order to prevent the spread of pathogenicity, it is necessary to identify infected people for quarantine and treatment. The use of artificial intelligence and data mining approaches can lead to prevention and reduction of treatment costs. The purpose of this study is to create data mining models in order to diagnose people with the disease of COVID-19 through the sound of coughing.

Method

In this research, Supervised Learning classification algorithms have been used, which include Support Vector Machine (SVM), random forest, and Artificial Neural Networks, that based on the standard "Fully Connected" neural network, Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) recurrent neural networks have been established. The data used in this research was from the online site sorfeh.com/sendcough/en, which has data collected during the spread of COVID-19.

Result

With the data we have collected (about 40,000 people) in different networks, we have reached acceptable accuracies.

Conclusion

These findings show the reliability of this method for using and developing a tool as a screening and early diagnosis of people with COVID-19. This method can also be used with simple artificial intelligence networks so that acceptable results can be expected. Based on the findings, the average accuracy was 83% and the best model was 95%.