AI4COVID-19: AI enabled preliminary diagnosis for COVID-19 from cough samples via an app.
Artificial intelligence
COVID-19
Pre-screening
Preliminary medical diagnosis
Public healthcare
Journal
Informatics in medicine unlocked
ISSN: 2352-9148
Titre abrégé: Inform Med Unlocked
Pays: England
ID NLM: 101718051
Informations de publication
Date de publication:
2020
2020
Historique:
received:
04
05
2020
revised:
19
06
2020
accepted:
19
06
2020
entrez:
26
8
2020
pubmed:
26
8
2020
medline:
26
8
2020
Statut:
ppublish
Résumé
The inability to test at scale has become humanity's Achille's heel in the ongoing war against the COVID-19 pandemic. A scalable screening tool would be a game changer. Building on the prior work on cough-based diagnosis of respiratory diseases, we propose, develop and test an Artificial Intelligence (AI)-powered screening solution for COVID-19 infection that is deployable via a smartphone app. The app, named AI4COVID-19 records and sends three 3-s cough sounds to an AI engine running in the cloud, and returns a result within 2 min. Cough is a symptom of over thirty non-COVID-19 related medical conditions. This makes the diagnosis of a COVID-19 infection by cough alone an extremely challenging multidisciplinary problem. We address this problem by investigating the distinctness of pathomorphological alterations in the respiratory system induced by COVID-19 infection when compared to other respiratory infections. To overcome the COVID-19 cough training data shortage we exploit transfer learning. To reduce the misdiagnosis risk stemming from the complex dimensionality of the problem, we leverage a multi-pronged mediator centered risk-averse AI architecture. Results show AI4COVID-19 can distinguish among COVID-19 coughs and several types of non-COVID-19 coughs. The accuracy is promising enough to encourage a large-scale collection of labeled cough data to gauge the generalization capability of AI4COVID-19. AI4COVID-19 is not a clinical grade testing tool. Instead, it offers a screening tool deployable anytime, anywhere, by anyone. It can also be a clinical decision assistance tool used to channel clinical-testing and treatment to those who need it the most, thereby saving more lives.
Sections du résumé
BACKGROUND
BACKGROUND
The inability to test at scale has become humanity's Achille's heel in the ongoing war against the COVID-19 pandemic. A scalable screening tool would be a game changer. Building on the prior work on cough-based diagnosis of respiratory diseases, we propose, develop and test an Artificial Intelligence (AI)-powered screening solution for COVID-19 infection that is deployable via a smartphone app. The app, named AI4COVID-19 records and sends three 3-s cough sounds to an AI engine running in the cloud, and returns a result within 2 min.
METHODS
METHODS
Cough is a symptom of over thirty non-COVID-19 related medical conditions. This makes the diagnosis of a COVID-19 infection by cough alone an extremely challenging multidisciplinary problem. We address this problem by investigating the distinctness of pathomorphological alterations in the respiratory system induced by COVID-19 infection when compared to other respiratory infections. To overcome the COVID-19 cough training data shortage we exploit transfer learning. To reduce the misdiagnosis risk stemming from the complex dimensionality of the problem, we leverage a multi-pronged mediator centered risk-averse AI architecture.
RESULTS
RESULTS
Results show AI4COVID-19 can distinguish among COVID-19 coughs and several types of non-COVID-19 coughs. The accuracy is promising enough to encourage a large-scale collection of labeled cough data to gauge the generalization capability of AI4COVID-19. AI4COVID-19 is not a clinical grade testing tool. Instead, it offers a screening tool deployable anytime, anywhere, by anyone. It can also be a clinical decision assistance tool used to channel clinical-testing and treatment to those who need it the most, thereby saving more lives.
Identifiants
pubmed: 32839734
doi: 10.1016/j.imu.2020.100378
pii: S2352-9148(20)30302-6
pii: 100378
pmc: PMC7318970
doi:
Types de publication
Journal Article
Langues
eng
Pagination
100378Informations de copyright
© 2020 The Authors.
Déclaration de conflit d'intérêts
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Références
Ann Intern Med. 1990 Nov 1;113(9):664-70
pubmed: 2221647
J Eur Acad Dermatol Venereol. 2020 May;34(5):e212-e213
pubmed: 32215952
Chest. 2018 Jan;153(1):196-209
pubmed: 29080708
Annu Int Conf IEEE Eng Med Biol Soc. 2019 Jul;2019:2601-2605
pubmed: 31946429
IEEE Access. 2020 Jul 15;8:130820-130839
pubmed: 34812339
AJR Am J Roentgenol. 2020 Jun;214(6):1280-1286
pubmed: 32130038
Eur Radiol. 2020 Aug;30(8):4381-4389
pubmed: 32193638
Annu Int Conf IEEE Eng Med Biol Soc. 2019 Jul;2019:7153-7156
pubmed: 31947484
IEEE Trans Med Imaging. 2020 Aug;39(8):2584-2594
pubmed: 32730211
PLoS One. 2020 Jul 9;15(7):e0235844
pubmed: 32645053
Chest. 2006 Jan;129(1 Suppl):1S-23S
pubmed: 16428686
Exp Ther Med. 2020 Aug;20(2):727-735
pubmed: 32742318
Can Assoc Radiol J. 2020 May;71(2):195-200
pubmed: 32129670
N Engl J Med. 2000 Dec 7;343(23):1715-21
pubmed: 11106722
PLoS One. 2016 Sep 01;11(9):e0162128
pubmed: 27583523
Phys Eng Sci Med. 2020 Jun;43(2):635-640
pubmed: 32524445
Int J Med Sci. 2020 Jun 6;17(10):1439-1448
pubmed: 32624700
Radiology. 2020 Aug;296(2):E72-E78
pubmed: 32216717
Radiology. 2020 Aug;296(2):E46-E54
pubmed: 32155105
Pattern Anal Appl. 2021;24(3):1207-1220
pubmed: 33994847
Engineering (Beijing). 2020 Oct;6(10):1122-1129
pubmed: 32837749
Radiology. 2020 Aug;296(2):E65-E71
pubmed: 32191588
Sci Rep. 2020 Nov 11;10(1):19549
pubmed: 33177550
Pattern Recognit Lett. 2020 Oct;138:638-643
pubmed: 32958971
Med J Aust. 2006 Apr 17;184(8):398-403
pubmed: 16618239
Med J Aust. 2010 Mar 1;192(5):265-71
pubmed: 20201760
J Chronic Dis. 1984;37(3):215-25
pubmed: 6699126
Respir Res. 2019 Jun 6;20(1):81
pubmed: 31167662
Pulm Pharmacol. 1996 Oct-Dec;9(5-6):261-8
pubmed: 9232662
AJR Am J Roentgenol. 2020 May;214(5):1072-1077
pubmed: 32125873
Radiology. 2020 Oct;297(1):E236-E237
pubmed: 32339083
N Engl J Med. 2020 Apr 16;382(16):1564-1567
pubmed: 32182409
J Thorac Oncol. 2020 May;15(5):700-704
pubmed: 32114094
J Med Eng Technol. 2013 Apr;37(3):165-71
pubmed: 23631519