Deep learning diagnostic and risk-stratification pattern detection for COVID-19 in digital lung auscultations: clinical protocol for a case-control and prospective cohort study

Article Article scientifique Article Deep learning diagnostic and risk-stratification pattern detection for COVID-19 in digital lung auscultations: clinical protocol for a case-control and prospective cohort study eng 2021-03-24 2021-03-24 author Glangetas Alban Division of Paediatric Emergency Medicine, Department of Women, Child and Adolescent, Geneva University Hospitals, 47 Avenue de la Roseraie, 1205, Geneva, Switzerland. 779424 author Hartley Mary-Anne Intelligent Global Health, Machine Learning and Optimization (MLO) Laboratory, Swiss Federal Institute of Technology (EPFL), Lausanne, Switzerland. author Cantais Aymeric 0000-0002-8650-2129 Division of Paediatric Emergency Medicine, Department of Women, Child and Adolescent, Geneva University Hospitals, 47 Avenue de la Roseraie, 1205, Geneva, Switzerland. 930730 author Courvoisier Delphine 0000-0002-1956-2607 Quality of Care Unit, Geneva University Hospitals, Geneva, Switzerland. 94475 Département de médecine (DEMED) author Rivollet David Essential Tech Centre, Swiss Federal Institute of Technology (EPFL), Lausanne, Switzerland. author Shama Deeksha M Intelligent Global Health, Machine Learning and Optimization (MLO) Laboratory, Swiss Federal Institute of Technology (EPFL), Lausanne, Switzerland. author Perez Alexandre Geneva University Hospitals, Geneva, Switzerland. 633750 CUMD - Enseignement de chirurgie orale et implantologie (ECOIM) author Spechbach Hervé Division of Primary Care Medicine, Department of Community Medicine, Geneva University Hospitals, Geneva, Switzerland. 516540 author Trombert Véronique Florence Department of Internal Medicine and Rehabilitation, Geneva University Hospitals, Geneva, Switzerland. 34823 dept de réadaptation et gériatrie (MIGRG) author Bourquin Stéphane Department of Micro-Engineering, Geneva School of Engineering, Architecture and Landscape (HEPIA), Geneva, Switzerland. author Jaggi Martin Intelligent Global Health, Machine Learning and Optimization (MLO) Laboratory, Swiss Federal Institute of Technology (EPFL), Lausanne, Switzerland. author Barazzone Constance Paediatric Pulmonology Unit, Department of Women, Child and Adolescent, University Hospitals of Geneva, Geneva, Switzerland. 29124 Département de Pédiatrie, Gynécologie et Obstétrique (PETRI) author Gervaix Alain Division of Paediatric Emergency Medicine, Department of Women, Child and Adolescent, Geneva University Hospitals, 47 Avenue de la Roseraie, 1205, Geneva, Switzerland. 31715 Département de Pédiatrie, Gynécologie et Obstétrique (PETRI) author Siebert Johan 0000-0002-2871-3382 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. 507348 103 10.1186/s12890-021-01467-w 33761909 1471-2466 PMC7988633 https://bmcpulmmed.biomedcentral.com/articles/10.1186/s12890-021-01467-w 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. Artificial intelligence Auscultation COVID-19 Deep learning Pneumonia Respiratory sounds Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Adolescent Adult Aged Aged, 80 and over Algorithms Auscultation / methods COVID-19 / diagnosis COVID-19 Testing / methods Case-Control Studies Clinical Decision Rules Clinical Protocols Deep Learning Female Humans Male Middle Aged Prognosis Prospective Studies Risk Assessment Triage Young Adult Dewey 613 Dewey 618 Dewey 618.97 Hôpitaux Universitaires de Genève 135-SIA-SARS-CoV-2 COVID Square Point Capital Georg Waechter Memorial Foundation 58 Département de pédiatrie, gynécologie et obstétrique 223 64 Département de santé et médecine communautaires 22A 65 Département de réadaptation et gériatrie 22B false 149 Etiologie des pneumonies et marqueurs inflammatoires chez l'enfant fébrile 183 148 Facteurs influençants le développement pulmonaire: étude translationnelle chez l'animal et l'homme 182 BMC pulmonary medicine 21 1 736e7323-b616-440a-bea7-62124f325b72 Article (Published version) Public Article-scientifique-from-unpaywall.pdf 919186 application/pdf CC BY-4.0 3ed6b0ac6bcc49dd4766dd18c9f5bb843814033aa40497c03a94d489cf8789c9 045158c0-2cd8-474a-b17c-8eee3160348e Appendix Public 12890_2021_1467_MOESM1_ESM.docx 32655 application/vnd.openxmlformats-officedocument.wordprocessingml.document STROBE Statement checklist. da19891d3c9b3f5c5bc4fb432292f3f00f880e2b087a33aefa28268587705178 openAccess Faculté de médecine / Section de médecine clinique / Département de pédiatrie, gynécologie et obstétrique Faculté de médecine / Section de médecine clinique / Département de santé et médecine communautaires Faculté de médecine / Section de médecine clinique / Département de réadaptation et gériatrie

Title	Published in	Year	Views	Downloads

Deep learning diagnostic and risk-stratification pattern detection for COVID-19 in digital lung auscultations: clinical protocol for a case-control and prospective cohort study	BMC pulmonary medicine	2021	137	147
Pediatric procedural sedation and analgesia in the emergency department: surveying the current European practice	European journal of pediatrics	2021	273	595

Archive ouverte UNIGE

Cantais, Aymeric