Artificial Intelligence-based Speech Signal for COVID-19 Diagnostics

The speech signal has numerous features that represent the characteristics of a specific language and recognize emotions. It also contains information that can be used to identify the mental, psychological, and physical states of the speaker. Recently, the acoustic analysis of speech signals offers a practical, automated, and scalable method for medical diagnosis and monitoring symptoms of many diseases. In this paper, we explore the deep acoustic features from confirmed positive and negative cases of COVID-19 and compare the performance of the acoustic features and COVID-19 symptoms in terms of their ability to diagnose COVID-19. The proposed methodology consists of the pre-trained Visual Geometry Group (VGG-16) model based on Mel spectrogram images to extract deep audio features. In addition to the K-means algorithm that determines effective features, followed by a Genetic Algorithm-Support Vector Machine (GA-SVM) classifier to classify cases. The experimental findings indicate the proposed methodology's capability to classify COVID-19 and NOT COVID-19 from acoustic features compared to COVID-19 symptoms, achieving an accuracy of 97%. The experimental results show that the proposed method remarkably improves the accuracy of COVID-19 detection over the handcrafted features used in previous studies. © 2022 ACM.

Acoustic-Based Deep Learning Architectures for Lung Disease Diagnosis: A Comprehensive Overview.

Lung auscultation has long been used as a valuable medical tool to assess respiratory health and has gotten a lot of attention in recent years, notably following the coronavirus epidemic. Lung auscultation is used to assess a patient's respiratory role. Modern technological progress has guided the growth of computer-based respiratory speech investigation, a valuable tool for detecting lung abnormalities and diseases. Several recent studies have reviewed this important area, but none are specific to lung sound-based analysis with deep-learning architectures from one side and the provided information was not sufficient for a good understanding of these techniques. This paper gives a complete review of prior deep-learning-based architecture lung sound analysis. Deep-learning-based respiratory sound analysis articles are found in different databases including the Plos, ACM Digital Libraries, Elsevier, PubMed, MDPI, Springer, and IEEE. More than 160 publications were extracted and submitted for assessment. This paper discusses different trends in pathology/lung sound, the common features for classifying lung sounds, several considered datasets, classification methods, signal processing techniques, and some statistical information based on previous study findings. Finally, the assessment concludes with a discussion of potential future improvements and recommendations.

Some considerations on the use of space sound absorbers with next-generation materials reflecting COVID situations in Japan: additional sound absorption for post-pandemic challenges in indoor acoustic environments.

In this study, we first point out the possible acoustic problems associated with the post-pandemic operation of built environments. In particular, we focus on the problem of acoustic deficiency due to the lack of absorption. This deficiency, which is likely to be encountered in most enclosed spaces in a range of establishments, is due to the reduced number of audience members or users of the space as a result of social distancing. As one of the promising solutions to this problem, we introduce a sound absorption technique using three-dimensional (3D) space sound absorbers developed through our recent research projects. Significantly, the type of sound absorber proposed herein is made of materials that are especially suited to hygiene considerations. The materials are microperforated panels (MPPs) and permeable membranes (PMs), both of which are easily washable and sanitised. Furthermore, we point out that 3D-MPP or PM space absorbers possess the additional value of aesthetic designability.

How did the 'state of emergency' declaration in Japan due to the COVID-19 pandemic affect the acoustic environment in a rather quiet residential area?

The COVID-19 pandemic caused lockdowns in many countries worldwide. Acousticians have made surveys to monitor how cities became quieter under the lockdown, mainly in central areas in cities. However, there have been few studies on the changes in the acoustic environment due to the pandemic in the usually quieter residential areas. It may be expected to be different from the effect in 'originally noisy' areas. Also, the effect could be different in Japan, because the 'state of emergency' declaration there was different to lockdowns elsewhere. Considering these circumstances, this article reports the results of noise monitoring and makes some observations on the acoustic environment in residential areas far from city centres, to provide an example of how the acoustic environment was affected by the state of emergency declaration due to the COVID-19 pandemic in Japan. The results showed that the reduction of noise levels was somewhat less than that reported in large cities. Also, comparing the results after the cancellation of the state of emergency, the noise level increased again. However, observations of noise sources imply that a possible change in human behaviour may have also affected the acoustic environment.

Brazilian Portuguese-Russian (BraPoRus) corpus: automatic transcription and acoustic quality of elderly speech during the COVID-19 pandemic

This article presents the Brazilian Portuguese-Russian (BraPoRus) corpus, whose goal is to collect, analyze, and preserve for posterity the spoken heritage Russian still used today in Brazil by approximately 1,500 elderly bilingual heritage Russian-Brazilian Portuguese speakers. Their unique 100-year-old variety of moribund Russian is disappearing because it has not been passed to their descendants born in Brazil. During the COVID-19 pandemic, we remotely collected 170 h of speech samples in heritage Russian from 26 participants (M (age) = 75.7 years) in naturalistic settings using Zoom or a phone call. To estimate the quality of collected data, we focus on two methodological challenges, automatic transcription and acoustic quality of remote recordings. First, we find that among commercially available transcription programs, Sonix far outperforms Google Transcribe and Vocalmatic on the measure of word error rate (WER). Second, we also establish that the acoustic quality of the remote recordings was adequate for intonational and speech rate analysis. Moreover, this remote method of collecting and analyzing speech samples works successfully with elderly bilingual participants who speak a heritage language different from their dominant societal language, and it can become a new norm when face-to-face communication with elderly participants is not possible.

Effects of various respiratory activities on human aerosol emissions: experimental design and preliminary results

This study combines particle measurements and acoustic measurements to study aerosols generated in breathing, speaking, singing and coughing. Particle measurements are carried out using a portable measurement chamber designed specially for the study. Acoustic measurements of voice production are conduced to standardize measurements in human aerosol emission and to reveal possible reasons for the individual differences in particle generation. Understanding mechanisms of human aerosol generation is important in trying to understand how the airborne transmission of pathogens takes place and furthermore in assessing how to minimize the risk of transmission. The results can be used in the context of all airborne diseases. © 2022 17th International Conference on Indoor Air Quality and Climate, INDOOR AIR 2022. All rights reserved.

Emergency Auscultation Sound Transmission System Using Web Conferencing Services in Japan

The impact of COVID-19 has led to a rapid increase in the demand for home medicine and telemedicine at a global level. This is also true in Japan. However, telemedicine in Japan is mainly carried out by telephone or video calls, and does not cover the provision of information by percussion, palpation, or auscultation. As a result, remote auscultation services are not widely available, and we were unable to use them for "COVID-19."In this paper, we focus on auscultation sounds and construct a mechanism to remotely transmit auscultation sounds, with the aim of increasing information that physicians can use to make judgments and thus contribute to more accurate diagnoses. Based on this, we shall consider a system that can use existing web conferencing services to transmit auscultation and enable remote auscultation in countries where specialist services cannot be used (i.e., in the case of disasters) or where such specialist services are not widely available. © 2022 IEEE.

Pareto-Optimized AVQI Assessment of Dysphonia: A Clinical Trial Using Various Smartphones

Multiparametric indices offer a more comprehensive approach to voice quality assessment by taking into account multiple acoustic parameters. Artificial intelligence technology can be utilized in healthcare to evaluate data and optimize decision-making processes. Mobile devices provide new opportunities for remote speech monitoring, allowing the use of basic mobile devices as screening tools for the early identification and treatment of voice disorders. However, it is necessary to demonstrate equivalence between mobile device signals and gold standard microphone preamplifiers. Despite the increased use and availability of technology, there is still a lack of understanding of the impact of physiological, speech/language, and cultural factors on voice assessment. Challenges to research include accounting for organic speech-related covariables, such as differences in conversing voice sound pressure level (SPL) and fundamental frequency (f0), recognizing the link between sensory and experimental acoustic outcomes, and obtaining a large dataset to understand regular variation between and within voice-disordered individuals. Our study investigated the use of cellphones to estimate the Acoustic Voice Quality Index (AVQI) in a typical clinical setting using a Pareto-optimized approach in the signal processing path. We found that there was a strong correlation between AVQI results obtained from different smartphones and a studio microphone, with no significant differences in mean AVQI scores between different smartphones. The diagnostic accuracy of different smartphones was comparable to that of a professional microphone, with optimal AVQI cut-off values that can effectively distinguish between normal and pathological voice for each smartphone used in the study. All devices met the proposed 0.8 AUC threshold and demonstrated an acceptable Youden index value.

Evaluation of the changing urban soundscape in Turkey during Covid-19 pandemic through online survey

The disease caused by the virus named Covid-19 and declared as a pandemic has shortly spread worldwide. Measures taken during the pandemic has exceedingly affected the acoustic environment of the cities. Sounds are a part of the human activities in the cities;therefore, they contain information regarding city life. It is possible to understand the positive or negative impacts of the pandemic on social life by analyzing the acoustic life throughout the process. Within the scope of the study, the impacts of the changing social life in Turkey on the city acoustic were studied physically, socially, and psychologically with the soundscape approach. The study conducted accordingly is designed to understand how the Covid-19 process affected the urban soundscape. For this reason, the focus was on the answers given to the participants on their level of pleasantness with the acoustical environment before and during the pandemic, the change in the sound sources they heard, and the sounds they were pleased to hear during the pandemic. Due to Covid-19 restrictions, the study was conducted by using an online Internet survey with 690 participants across Turkey. In addition to the cities with high participation in the study and a general evaluation was made. At the end of the study, it was seen that the change in sound environment pleasantness was more pronounced, especially in the cities of high population density. In general, the audibility of nature-based sounds increased and nature-based sounds were found to be pleasing during the pandemic. © 2023, Istanbul Teknik Universitesi, Faculty of Architecture. All rights reserved.

Relationship Between Vocal Fatigue Index and Acoustic Voice Scales in Patients With Coronavirus Infection.

OBJECTIVES: The voice quality of patients with Coronavirus Disease 2019 (COVID-19) seems to be affected due to lower and upper respiratory involvement. Patient-based voice assessment scales are important clinical measures to diagnose voice disorders and monitor treatment outcomes in COVID-19 patients. This study compared vocal fatigue between COVID-19 patients and those with normal voices. Furthermore, the relationship between vocal fatigue and acoustic voice parameters of COVID-19 patients was evaluated. METHODS: This cross-sectional study enrolled 30 laboratory-confirmed patients with COVID-19 (18 males and 12 females) and 30 healthy individuals with normal voices (14 males and 16 females) to compare their respiratory or phonatory parameters. The Persian versions of the Consensus Auditory Perceptual Evaluation of Voice (CAPE-V) and the vocal fatigue index (VFI) were conducted before and after reading the text. The Jitter, shimmer, maximum phonation time, and harmonic-to-noise ratio (HNR) were analyzed by Praat software based on the recorded voices of CAPE-V tasks. The acoustic assessment and VFI questionnaire results were compared between COVID-19 patients and the control group. RESULTS: There were significant differences between COVID-19 patients and their healthy counterparts in all VFI subscales (P < 0.001). Moreover, after reading the text, we found significant differences between the two groups regarding Jitter, shimmer, and HNR of /a/ and /i/ vowels (P < 0.05). Our findings also indicated a significant correlation between symptom improvement with rest and acoustic parameters in all tasks, except the Jitter of /a/ before reading the text. CONCLUSION: Patients with COVID-19 showed significantly more vocal fatigue than people with normal voices after reading the text. Moreover, there was a significant relationship between Jitter, shimmer, and HNR and the tiredness of voice and physical discomfort subscales of VFI.

Assessment of humoral and cellular immunity after bivalent BNT162b2 vaccination and potential association with reactogenicity

Nowadays, new challenges around increasing production quality and productivity, and decreasing energy consumption, are growing in the manufacturing industry. In order to tackle these challenges, it is of vital importance to monitor the health of critical components. In the machine tool sector, one of the main aspects is to monitor the wear of the cutting tools, as it affects directly to the fulfillment of tolerances, production of scrap, energy consumption, etc. Besides, the prediction of the remaining useful life (RUL) of the cutting tools, which is related to their wear level, is gaining more importance in the field of predictive maintenance, being that prediction is a crucial point for an improvement of the quality of the cutting process. Unlike monitoring the current health of the cutting tools in real time, as tool wear diagnosis does, RUL prediction allows to know when the tool will end its useful life. This is a key factor since it allows optimizing the planning of maintenance strategies. Moreover, a substantial number of signals can be captured from machine tools, but not all of them perform as optimum predictors for tool RUL. Thus, this paper focuses on RUL and has two main objectives. First, to evaluate the optimum signals for RUL prediction, a substantial number of them were captured in a turning process and investigated by using recursive feature elimination (RFE). Second, the use of bidirectional recurrent neural networks (BRNN) as regressive models to predict the RUL of cutting tools in machining operations using the investigated optimum signals is investigated. The results are compared to traditional machine learning (ML) models and convolutional neural networks (CNN). The results show that among all the signals captured, the root mean squared (RMS) parameter of the forward force ( F-y ) is the optimum for RUL prediction. As well, the bidirectional long-short term memory (BiLSTM) and bidirectional gated recurrent units (BiGRU), which are two types of BRNN, along with the RMS of F-y signal, achieved the lowest root mean squared error (RMSE) for tool RUL, being also computationally the most demanding ones.

Vessel traffic influences distribution of Aotearoa New Zealand's endemic dolphin (Cephalorhynchus hectori)

Cetacean tourism and vessel traffic have grown considerably around the world in recent decades. At Akaroa Harbor, Aotearoa New Zealand, recreational vessel traffic, dolphin tourism, and cruise ship presence increased substantially between 2008 and 2020. We examined the relationship between vessel traffic parameters and the presence of Hector's dolphins (Cephalorhynchus hectori) during the austral summer 2019–2020, using automated vessel tracking and autonomous passive acoustic monitoring. Data were collected between December 2019 and May 2020, including the entirety of the first COVID‐19 nationwide lockdown. Generalized additive models revealed that increasing levels of motor vessel traffic, the presence of cruise ships, and high levels of dolphin tour vessel traffic resulted in decreases in acoustic detections of dolphins. Our findings suggest that Hector's dolphins at Akaroa Harbor were displaced from core habitat in response to each of these vessel traffic parameters. We recommend that managers use immediately actionable tools to reduce the impacts of vessels on these dolphins.

Influence of interventions against spread of COVID-19 on noise levels in student restaurant

One of the main interventions against the spread of COVID-19 is so called social distancing, which has rather large consequence on activities held in the building interior. The direct impact can be seen in the limitations of the number of people present in an enclosed space at the same time in order to fulfil the necessary distance between persons. However, these rather unpleasant restrictions with dramatical impact on functioning of the so-called HORECA (hotel-restaurant-café) services, turned out to have also a positive impact on the indoor acoustic comfort. The main difference can be seen in the decreased noise levels in restaurants. This article compares acoustic performance of one space in two situations (before and during pandemic) and shows to what extent a simulation can be used for prediction of noise from quasi dynamic sound sources (i.e. talking people). © International Building Performance Simulation Association, 2022

Adaptive Noise Cancellation Using a Fully Connected Network: A Lesson Learned

Ambient noise causes annoying difficulty for listeners, especially in online learning and work-from-home environments such as during the COVID-19 pandemic. The aim of this work was to employ the neural network to mitigate such ambient noise in the online environment. The software was designed, implemented, and tested on 4 types of noise. The algorithm used was a fully connected network. The results indicated that the standard fully connected network might not be an effective solution for a specific situation. Nonetheless, the processing time was very low, making it possible for real-time application on standalone devices. The implementation using leaky ReLu, creating leaky networks, offered slightly better results in English speeches, i.e. an average of 1.382 and 0.4389 in the PESQ and STOI, respectively. The Thai leaky networks, on another hand, exhibited an average of 3.111 and 0.7096 in PESQ and STOI, respectively. © 2022 IEEE.

A Survey on Perceived Indoor Acoustic Quality by Workers from Home during COVID-19 Lockdown in Italy

From March 2020, Italians experienced lockdown due to COVID-19 pandemic. People had to share common living spaces with family members for an extended period converting their home into workplaces. This resulted in changes to everyday life noises with implications in terms of perception of indoor acoustic quality. An online survey was designed and distributed to Italian residents to assess how they perceived the indoor quality of domestic spaces when working from home. A total of 330 questionnaires were collected and analyzed. The paper reports the results of the analyses carried out, focusing on the acoustic quality in home spaces and the satisfaction of the respondents, including an analysis of the housing context. Most respondents attach great importance to the acoustic aspects in judging the quality of the living environment and believe that the acoustic quality can improve the performance of their work. The comparison between pre-lockdown and lockdown periods shows that noises inside the building prevail over those coming from the outside and annoyance is mainly due to noise from shared spaces. The results of this study highlighted how the COVID-19 lockdown was a unique opportunity to draw attention to the importance of the indoor acoustic quality. © 2023 by the authors.

Chronic sleep disturbance among adult Canadians: Associations and implications to the evaluation of noise impacts under Canada's Impact Assessment Act

The Canadian Perspectives on Environmental Noise Survey was completed online by 6647 randomly selected Canadians 18 years of age and older between April 12 and May 25, 2021. The survey objective was to explore attitudes, perceptions, and expectations toward environmental noise in rural and non-rural Canada. The questionnaire assessed self-reported high sleep disturbance (HSD) in the previous year, at home. The prevalence of HSD was 7.8% overall. A list of potential sources of sleep disturbance was provided to the full sample, where 6.1%, 5.2%, and 3.0% reported HSD by noisy neighbors, road traffic noise and indoor noise, respectively. Stress/anxiety or worrying about something was selected most frequently at 12.9%. Finally, 7.6% and 5.5% reported pain/illness and partner's sleep disturbance, respectively, as sources of HSD. Reported HSD was significantly higher among respondents below 55 years of age, females, lower income groups, unemployed respondents, those on paid leave (sick, maternity, disability), and living in an urban area. Expectations of quiet, perceiving nighttime noise to have increased over time, high noise sensitivity, hearing and being highly annoyed by road traffic noise was also associated with an increased prevalence of reporting HSD. In contrast to hearing impairment and heart disease (including high blood pressure);rated physical health, mental health, anxiety/depression, and reporting a sleep disorder, were associated with increased HSD. The perceived affects of the COVID-19 pandemic on health and annoyance toward environmental and indoor noise also influenced HSD. In the fully adjusted multivariate logistic regression model, the effect of age, gender, changes in nighttime noise, road traffic noise annoyance, noise sensitivity and sleep disorder remained statistically significant. The univariate and multivariate models showed a similar prevalence of HSD between Indigenous Peoples and non-Indigenous Canadians. Results are discussed in relation to the provision of advice on sleep and health under Canada's Impact Assessment Act. © 2023

Volatile Organic Compound Exhaled Breath Sensing

This book chapter presents a broad overview of the application of nanotechnology in the biomedical area, exemplified by the application of several gas sensors (electrochemical sensors, piezoelectric sensors, optical, chemoresistive, metal oxide sensors, surface acoustic wave sensors) and focusing on the study of volatile organic compounds (VOCs) in exhaled breath for the screening of diseases of worldwide interest such as breast cancer, lung cancer, COVID-19, post COVID-19 syndrome, colorectal cancer, prostate cancer, diabetes, chronic obstructive disease, among others. This document aims to provide the state of the art in disruptive technologies based on nanosensors, especially electronic noses and the advances and perspectives in this field. The present work represents an important tool for researchers who are in the field of the development of sensing disruptive technologies for the study of VOCs in biological matrices (i.e., exhaled breath). Thus, the application of gas sensors has proven to be feasible in the biomedical area and a promising area within the diagnosis of communicable and non-communicable diseases, to be applied in POC settings, clinics, hospitals, doctors' offices, and especially in-field applications for less-favored populations where they lack the minimum resources to achieve universal health coverage. © 2023 Elsevier Ltd. All rights reserved

Design of ear-contactless stethoscope and improvement in the performance of deep learning based on CNN to classify the heart sound.

Cardiac-related disorders are rapidly growing throughout the world. Accurate classification of cardiovascular diseases is an important research topic in healthcare. During COVID-19, auscultating heart sounds was challenging as health workers and doctors wear protective clothing, and direct contact with patients can spread the outbreak. Thus, contactless auscultation of heart sound is necessary. In this paper, a low-cost ear contactless stethoscope is designed where auscultation is done with the help of a bluetooth-enabled micro speaker instead of an earpiece. The PCG recordings are further compared with other standard electronic stethoscopes like Littman 3 M. This work is made to improve the performance of deep learning-based classifiers like recurrent neural networks (RNN) and convolutional neural networks (CNN) for different valvular heart problems using tuning of hyperparameters like learning rate of optimizers, dropout rate, and hidden layer. Hyper-parameter tuning is used to optimize the performances of various deep learning models and their learning curves for real-time analysis. The acoustic, time, and frequency domain features are used in this research. The investigation is made on the heart sounds of normal and diseased patients available from the standard data repository to train the software models. The proposed CNN-based inception network model achieved an accuracy of 99.65 ± 0.06% on the test dataset with a sensitivity of 98.8 ± 0.05% and specificity of 98.2 ± 0.19%. The proposed hybrid CNN-RNN architecture attained 91.17 ± 0.03% accuracy on test data after hyperparameter optimization, whereas the LSTM-based RNN model achieved 82.32 ± 0.11% accuracy. Finally, the evaluated results were compared with machine learning algorithms, and the improved CNN-based Inception Net model is the most effective among others.

The effects of audio-visual perceptual characteristics on environmental health of pedestrian streets with traffic noise: A case study in Dalian, China.

The COVID-19 pandemic has affected city dwellers' physical and mental health and has raised concerns about the health of urban public spaces. This field investigation research in Dalian, China, examined the perceived audio-visual environment characteristics of urban pedestrian streets with traffic noise and their influences on the environmental health of the pedestrian streets. Five indicators reflecting psychological responses to environmental characteristics (willingness to walk, relaxation, safety, beauty, and comprehensive comfort) were used to measure environmental health of pedestrian streets with traffic noise. The results showed that safety was rated the highest, and willingness to walk was evaluated as the lowest among health evaluation indicators. The imageability and openness of the streetscape were associated with each health evaluation indicator. In contrast, the rhythm and continuity of the street buildings had a greater effect on willingness to walk than the other health indicators. There were negative correlations between L Aeq for traffic noise and health evaluations. Positive health evaluations were observed when L Aeq was less than 55 dBA. In contrast, soundscape indicators showed positive correlations with health evaluations, and acoustic comfort and noise annoyance, rather than sound preference and subjective loudness were associated with each health evaluation indicator. In terms of the combined audio-visual factors, acoustic comfort, the quantity of greening, annoyance, sky visibility, spatial scale, and building distance were examined as the determining factors affecting health evaluations, and 55.40% of the variance in health evaluations was explained by the soundscape and streetscape indicators. The findings provide references for better understanding the relationships between healthy experience and audio-visual perceptions. Moreover, they enable environmental health quality optimisation of pedestrian spaces considering audio-visual indicators and approaches in the post-epidemic era.