Automated voice-based detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) could facilitate the screening for COVID19. A dataset of cellular phone recordings from 88 subjects was recently collected. The dataset included vocal utterances, speech and coughs that were self-recorded by the subjects in either hospitals or isolation sites. All subjects underwent nasopharyngeal swabbing at the time of recording and were labelled as SARS-CoV-2 positives or negative controls. The present study harnessed deep machine learning and speech processing to detect the SARS-CoV-2 positives. A three-stage architecture was implemented. A self-supervised attention-based transformer generated embeddings from the audio inputs. Recurrent neural networks were used to produce specialized sub-models for the SARS-CoV-2 classification. An ensemble stacking fused the predictions of the sub-models. Pre-training, bootstrapping and regularization techniques were used to prevent overfitting. A recall of 78% and a probability of false alarm (PFA) of 41% were measured on a test set of 57 recording sessions. A leave-one-speaker-out cross validation on 292 recording sessions yielded a recall of 78% and a PFA of 30%. These preliminary results imply a feasibility for COVID19 screening using voice.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8769003PMC
http://dx.doi.org/10.1109/OJEMB.2020.3026468DOI Listing

Publication Analysis

Top Keywords

sars-cov-2 positives
8
recall 78%
8
recording sessions
8
sars-cov-2
5
sars-cov-2 detection
4
detection voice
4
voice automated
4
automated voice-based
4
voice-based detection
4
detection severe
4

Similar Publications

Background: COVID-19 has caused over 46,000 deaths in New York City, with a disproportional impact on certain communities. As part of the COVID-19 response, the city has directly administered over 6 million COVID-19 tests (in addition to millions of indirectly administered tests not covered in this analysis) at no cost to individuals, resulting in nearly half a million positive results. Given that the prevalence of testing, throughout the pandemic, has tended to be higher in more affluent areas, these tests were targeted to areas with fewer resources.

View Article and Find Full Text PDF

Assessing Relationship Between Goat and Sheep Farmers' Stress and Their Demographics: A Pilot Study.

J Agric Saf Health

August 2024

Department of Agricultural Economics, Sociology, and Education, The Pennsylvania State University, University Park, Pennsylvania, USA.

Highlights: Limited leisure time, insufficient sleep, and family members' health conditions were the top personal stressors. Occupational stressors were too much to do in so little time, worrying about the farm's future and financial issues. Governmental regulation, market prices, and unpredictable weather conditions were off-farm occupational stressors.

View Article and Find Full Text PDF

Background: This study aims to determine the prevalence of long COVID two years after initial infection, identify prognostic factors, and assess its impact on quality of life.

Methodology: An ambispective cohort study was conducted with patients aged = 18 years from two health centers in Málaga, Spain, who tested positive for SARS-CoV-2 between October 2020 and May 2021. Systematic random sampling was performed in October 2022, with 5% precision, 5% alpha error, and 25% expected losses.

View Article and Find Full Text PDF

Femtosecond laser-ablative aqueous synthesis of multi-drug antiviral nanoparticles.

Nanomedicine (Lond)

March 2025

Department of Chemistry and The Institute for Laser, Photonics, and Biophotonics, University at Buffalo, The State University of New York, Buffalo, NY, USA.

Background: Nanomedicine offers a number of innovative strategies to address major public health burdens, including complex respiratory illnesses. In this work, we introduce a multi-drug nanoparticle fabricated using femtosecond laser ablation for the treatment of influenza, SARS-CoV-2, and their co-infections.

Methods: The SARS-CoV-2 antiviral, remdesivir; the influenza antiviral, baloxavir marboxil; and the anti-inflammatory, dexamethasone, were co-crystalized and then ablated in aqueous media using a femtosecond pulsed laser and subsequently surface modified with the cationic polymer, chitosan, or poly-d-lysine.

View Article and Find Full Text PDF

The coronavirus disease 2019 (COVID-19) pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has severely impacted public health. In 2022, the Omicron variant of SARS-CoV-2 rapidly became the dominant circulating variant in the local COVID-19 outbreaks in Tianjin Municipality, China. To gain a deeper understanding of the genetic variations of the Omicron variant in Tianjin, specimens from individuals who tested positive for SARS-CoV-2 between December 2021 and November 2022 were used for virus whole genome sequencing and phylogenetic analysis.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!