COVID-19 Artificial Intelligence Diagnosis Using Only Cough Recordings

<italic>Goal:</italic> We hypothesized that COVID-19 subjects, especially including asymptomatics, could be accurately discriminated only from a forced-cough cell phone recording using Artificial Intelligence. To train our MIT Open Voice model we built a data collection pipeline of COVID...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Jordi Laguarta, Ferran Hueto, Brian Subirana
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2020
Materias:	AI diagnostics convolutional neural networks COVID-19 screening deep learning speech recognition Computer applications to medicine. Medical informatics R858-859.7 Medical technology R855-855.5
Acceso en línea:	https://doaj.org/article/ba24e57b039c4d5d8995977205267931
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:ba24e57b039c4d5d8995977205267931
record_format	dspace
spelling	oai:doaj.org-article:ba24e57b039c4d5d89959772052679312021-11-24T00:03:37ZCOVID-19 Artificial Intelligence Diagnosis Using Only Cough Recordings2644-127610.1109/OJEMB.2020.3026928https://doaj.org/article/ba24e57b039c4d5d89959772052679312020-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9208795/https://doaj.org/toc/2644-1276<italic>Goal:</italic> We hypothesized that COVID-19 subjects, especially including asymptomatics, could be accurately discriminated only from a forced-cough cell phone recording using Artificial Intelligence. To train our MIT Open Voice model we built a data collection pipeline of COVID-19 cough recordings through our website (opensigma.mit.edu) between April and May 2020 and created the largest audio COVID-19 cough balanced dataset reported to date with 5,320 subjects. <italic>Methods:</italic> We developed an AI speech processing framework that leverages acoustic biomarker feature extractors to pre-screen for COVID-19 from cough recordings, and provide a personalized patient saliency map to longitudinally monitor patients in real-time, non-invasively, and at essentially zero variable cost. Cough recordings are transformed with Mel Frequency Cepstral Coefficient and inputted into a Convolutional Neural Network (CNN) based architecture made up of one Poisson biomarker layer and 3 pre-trained ResNet50's in parallel, outputting a binary pre-screening diagnostic. Our CNN-based models have been trained on 4256 subjects and tested on the remaining 1064 subjects of our dataset. Transfer learning was used to learn biomarker features on larger datasets, previously successfully tested in our Lab on Alzheimer's, which significantly improves the COVID-19 discrimination accuracy of our architecture. <bold><italic>Results:</italic> When validated with subjects diagnosed using an official test, the model achieves COVID-19 sensitivity of 98.5% with a specificity of 94.2% (AUC: 0.97). For asymptomatic subjects it achieves sensitivity of 100% with a specificity of 83.2%</bold>. <italic>Conclusions:</italic> AI techniques can produce a free, non-invasive, real-time, any-time, instantly distributable, large-scale COVID-19 asymptomatic screening tool to augment current approaches in containing the spread of COVID-19. Practical use cases could be for daily screening of students, workers, and public as schools, jobs, and transport reopen, or for pool testing to quickly alert of outbreaks in groups. General speech biomarkers may exist that cover several disease categories, as we demonstrated using the same ones for COVID-19 and Alzheimer's.Jordi LaguartaFerran HuetoBrian SubiranaIEEEarticleAI diagnosticsconvolutional neural networksCOVID-19 screeningdeep learningspeech recognitionComputer applications to medicine. Medical informaticsR858-859.7Medical technologyR855-855.5ENIEEE Open Journal of Engineering in Medicine and Biology, Vol 1, Pp 275-281 (2020)
institution	DOAJ
collection	DOAJ
language	EN
topic	AI diagnostics convolutional neural networks COVID-19 screening deep learning speech recognition Computer applications to medicine. Medical informatics R858-859.7 Medical technology R855-855.5
spellingShingle	AI diagnostics convolutional neural networks COVID-19 screening deep learning speech recognition Computer applications to medicine. Medical informatics R858-859.7 Medical technology R855-855.5 Jordi Laguarta Ferran Hueto Brian Subirana COVID-19 Artificial Intelligence Diagnosis Using Only Cough Recordings
description	<italic>Goal:</italic> We hypothesized that COVID-19 subjects, especially including asymptomatics, could be accurately discriminated only from a forced-cough cell phone recording using Artificial Intelligence. To train our MIT Open Voice model we built a data collection pipeline of COVID-19 cough recordings through our website (opensigma.mit.edu) between April and May 2020 and created the largest audio COVID-19 cough balanced dataset reported to date with 5,320 subjects. <italic>Methods:</italic> We developed an AI speech processing framework that leverages acoustic biomarker feature extractors to pre-screen for COVID-19 from cough recordings, and provide a personalized patient saliency map to longitudinally monitor patients in real-time, non-invasively, and at essentially zero variable cost. Cough recordings are transformed with Mel Frequency Cepstral Coefficient and inputted into a Convolutional Neural Network (CNN) based architecture made up of one Poisson biomarker layer and 3 pre-trained ResNet50's in parallel, outputting a binary pre-screening diagnostic. Our CNN-based models have been trained on 4256 subjects and tested on the remaining 1064 subjects of our dataset. Transfer learning was used to learn biomarker features on larger datasets, previously successfully tested in our Lab on Alzheimer's, which significantly improves the COVID-19 discrimination accuracy of our architecture. <bold><italic>Results:</italic> When validated with subjects diagnosed using an official test, the model achieves COVID-19 sensitivity of 98.5% with a specificity of 94.2% (AUC: 0.97). For asymptomatic subjects it achieves sensitivity of 100% with a specificity of 83.2%</bold>. <italic>Conclusions:</italic> AI techniques can produce a free, non-invasive, real-time, any-time, instantly distributable, large-scale COVID-19 asymptomatic screening tool to augment current approaches in containing the spread of COVID-19. Practical use cases could be for daily screening of students, workers, and public as schools, jobs, and transport reopen, or for pool testing to quickly alert of outbreaks in groups. General speech biomarkers may exist that cover several disease categories, as we demonstrated using the same ones for COVID-19 and Alzheimer's.
format	article
author	Jordi Laguarta Ferran Hueto Brian Subirana
author_facet	Jordi Laguarta Ferran Hueto Brian Subirana
author_sort	Jordi Laguarta
title	COVID-19 Artificial Intelligence Diagnosis Using Only Cough Recordings
title_short	COVID-19 Artificial Intelligence Diagnosis Using Only Cough Recordings
title_full	COVID-19 Artificial Intelligence Diagnosis Using Only Cough Recordings
title_fullStr	COVID-19 Artificial Intelligence Diagnosis Using Only Cough Recordings
title_full_unstemmed	COVID-19 Artificial Intelligence Diagnosis Using Only Cough Recordings
title_sort	covid-19 artificial intelligence diagnosis using only cough recordings
publisher	IEEE
publishDate	2020
url	https://doaj.org/article/ba24e57b039c4d5d8995977205267931
work_keys_str_mv	AT jordilaguarta covid19artificialintelligencediagnosisusingonlycoughrecordings AT ferranhueto covid19artificialintelligencediagnosisusingonlycoughrecordings AT briansubirana covid19artificialintelligencediagnosisusingonlycoughrecordings
_version_	1718416117197701120

COVID-19 Artificial Intelligence Diagnosis Using Only Cough Recordings

Ejemplares similares