Alban Glangetas1, Mary-Anne Hartley2, Aymeric Cantais1, Delphine S Courvoisier3, David Rivollet4, Deeksha M Shama2, Alexandre Perez5, Hervé Spechbach6, Véronique Trombert7, Stéphane Bourquin8, Martin Jaggi2, Constance Barazzone-Argiroffo9, Alain Gervaix1, Johan N Siebert10. 1. Division of Paediatric Emergency Medicine, Department of Women, Child and Adolescent, Geneva University Hospitals, 47 Avenue de la Roseraie, 1205, Geneva, Switzerland. 2. Intelligent Global Health, Machine Learning and Optimization (MLO) Laboratory, Swiss Federal Institute of Technology (EPFL), Lausanne, Switzerland. 3. Quality of Care Unit, Geneva University Hospitals, Geneva, Switzerland. 4. Essential Tech Centre, Swiss Federal Institute of Technology (EPFL), Lausanne, Switzerland. 5. Geneva University Hospitals, Geneva, Switzerland. 6. Division of Primary Care Medicine, Department of Community Medicine, Geneva University Hospitals, Geneva, Switzerland. 7. Department of Internal Medicine and Rehabilitation, Geneva University Hospitals, Geneva, Switzerland. 8. Department of Micro-Engineering, Geneva School of Engineering, Architecture and Landscape (HEPIA), Geneva, Switzerland. 9. Paediatric Pulmonology Unit, Department of Women, Child and Adolescent, University Hospitals of Geneva, Geneva, Switzerland. 10. Division of Paediatric Emergency Medicine, Department of Women, Child and Adolescent, Geneva University Hospitals, 47 Avenue de la Roseraie, 1205, Geneva, Switzerland. Johan.Siebert@hcuge.ch.
Abstract
BACKGROUND: Lung auscultation is fundamental to the clinical diagnosis of respiratory disease. However, auscultation is a subjective practice and interpretations vary widely between users. The digitization of auscultation acquisition and interpretation is a particularly promising strategy for diagnosing and monitoring infectious diseases such as Coronavirus-19 disease (COVID-19) where automated analyses could help decentralise care and better inform decision-making in telemedicine. This protocol describes the standardised collection of lung auscultations in COVID-19 triage sites and a deep learning approach to diagnostic and prognostic modelling for future incorporation into an intelligent autonomous stethoscope benchmarked against human expert interpretation. METHODS: A total of 1000 consecutive, patients aged ≥ 16 years and meeting COVID-19 testing criteria will be recruited at screening sites and amongst inpatients of the internal medicine department at the Geneva University Hospitals, starting from October 2020. COVID-19 is diagnosed by RT-PCR on a nasopharyngeal swab and COVID-positive patients are followed up until outcome (i.e., discharge, hospitalisation, intubation and/or death). At inclusion, demographic and clinical data are collected, such as age, sex, medical history, and signs and symptoms of the current episode. Additionally, lung auscultation will be recorded with a digital stethoscope at 6 thoracic sites in each patient. A deep learning algorithm (DeepBreath) using a Convolutional Neural Network (CNN) and Support Vector Machine classifier will be trained on these audio recordings to derive an automated prediction of diagnostic (COVID positive vs negative) and risk stratification categories (mild to severe). The performance of this model will be compared to a human prediction baseline on a random subset of lung sounds, where blinded physicians are asked to classify the audios into the same categories. DISCUSSION: This approach has broad potential to standardise the evaluation of lung auscultation in COVID-19 at various levels of healthcare, especially in the context of decentralised triage and monitoring. TRIAL REGISTRATION: PB_2016-00500, SwissEthics. Registered on 6 April 2020.
BACKGROUND: Lung auscultation is fundamental to the clinical diagnosis of respiratory disease. However, auscultation is a subjective practice and interpretations vary widely between users. The digitization of auscultation acquisition and interpretation is a particularly promising strategy for diagnosing and monitoring infectious diseases such as Coronavirus-19 disease (COVID-19) where automated analyses could help decentralise care and better inform decision-making in telemedicine. This protocol describes the standardised collection of lung auscultations in COVID-19 triage sites and a deep learning approach to diagnostic and prognostic modelling for future incorporation into an intelligent autonomous stethoscope benchmarked against human expert interpretation. METHODS: A total of 1000 consecutive, patients aged ≥ 16 years and meeting COVID-19 testing criteria will be recruited at screening sites and amongst inpatients of the internal medicine department at the Geneva University Hospitals, starting from October 2020. COVID-19 is diagnosed by RT-PCR on a nasopharyngeal swab and COVID-positivepatients are followed up until outcome (i.e., discharge, hospitalisation, intubation and/or death). At inclusion, demographic and clinical data are collected, such as age, sex, medical history, and signs and symptoms of the current episode. Additionally, lung auscultation will be recorded with a digital stethoscope at 6 thoracic sites in each patient. A deep learning algorithm (DeepBreath) using a Convolutional Neural Network (CNN) and Support Vector Machine classifier will be trained on these audio recordings to derive an automated prediction of diagnostic (COVID positive vs negative) and risk stratification categories (mild to severe). The performance of this model will be compared to a human prediction baseline on a random subset of lung sounds, where blinded physicians are asked to classify the audios into the same categories. DISCUSSION: This approach has broad potential to standardise the evaluation of lung auscultation in COVID-19 at various levels of healthcare, especially in the context of decentralised triage and monitoring. TRIAL REGISTRATION: PB_2016-00500, SwissEthics. Registered on 6 April 2020.
Entities:
Keywords:
Artificial intelligence; Auscultation; COVID-19; Deep learning; Pneumonia; Respiratory sounds; Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)
Authors: Marko Topalovic; Nilakash Das; Pierre-Régis Burgel; Marc Daenen; Eric Derom; Christel Haenebalcke; Rob Janssen; Huib A M Kerstjens; Giuseppe Liistro; Renaud Louis; Vincent Ninane; Christophe Pison; Marc Schlesser; Piet Vercauter; Claus F Vogelmeier; Emiel Wouters; Jokke Wynants; Wim Janssens Journal: Eur Respir J Date: 2019-04-11 Impact factor: 16.671
Authors: Tulin Ozturk; Muhammed Talo; Eylul Azra Yildirim; Ulas Baran Baloglu; Ozal Yildirim; U Rajendra Acharya Journal: Comput Biol Med Date: 2020-04-28 Impact factor: 4.589
Authors: Marco Francone; Franco Iafrate; Giorgio Maria Masci; Simona Coco; Francesco Cilia; Lucia Manganaro; Valeria Panebianco; Chiara Andreoli; Maria Chiara Colaiacomo; Maria Antonella Zingaropoli; Maria Rosa Ciardi; Claudio Maria Mastroianni; Francesco Pugliese; Francesco Alessandri; Ombretta Turriziani; Paolo Ricci; Carlo Catalano Journal: Eur Radiol Date: 2020-07-04 Impact factor: 5.315
Authors: Nour Kasim; Noa Bachner-Hinenzon; Shay Brikman; Ori Cheshin; Doron Adler; Guy Dori Journal: Biomed Signal Process Control Date: 2022-06-27 Impact factor: 5.076
Authors: Elena A Lapteva; Olga N Kharevich; Victoria V Khatsko; Natalia A Voronova; Maksim V Chamko; Irina V Bezruchko; Elena I Katibnikova; Elena I Loban; Mostafa M Mouawie; Helena Binetskaya; Sergey Aleshkevich; Aleksey Karankevich; Vitaly Dubinetski; Jørgen Vestbo; Alexander G Mathioudakis Journal: Eur Respir J Date: 2021-12-02 Impact factor: 16.671