Discriminant audio properties in deep learning based respiratory insufficiency detection in Brazilian Portuguese

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Gauy, Marcelo Matheus, Berti, Larissa Cristina, Cândido Jr, Arnaldo, Neto, Augusto Camargo, Goldman, Alfredo, Levin, Anna Sara Shafferman, Martins, Marcus, de Medeiros, Beatriz Raposo, Queiroz, Marcelo, Sabino, Ester Cerdeira, Svartman, Flaviane Romani Fernandes, Finger, Marcelo
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909212069593088
author Gauy, Marcelo Matheus
Berti, Larissa Cristina
Cândido Jr, Arnaldo
Neto, Augusto Camargo
Goldman, Alfredo
Levin, Anna Sara Shafferman
Martins, Marcus
de Medeiros, Beatriz Raposo
Queiroz, Marcelo
Sabino, Ester Cerdeira
Svartman, Flaviane Romani Fernandes
Finger, Marcelo
author_facet Gauy, Marcelo Matheus
Berti, Larissa Cristina
Cândido Jr, Arnaldo
Neto, Augusto Camargo
Goldman, Alfredo
Levin, Anna Sara Shafferman
Martins, Marcus
de Medeiros, Beatriz Raposo
Queiroz, Marcelo
Sabino, Ester Cerdeira
Svartman, Flaviane Romani Fernandes
Finger, Marcelo
contents This work investigates Artificial Intelligence (AI) systems that detect respiratory insufficiency (RI) by analyzing speech audios, thus treating speech as a RI biomarker. Previous works collected RI data (P1) from COVID-19 patients during the first phase of the pandemic and trained modern AI models, such as CNNs and Transformers, which achieved $96.5\%$ accuracy, showing the feasibility of RI detection via AI. Here, we collect RI patient data (P2) with several causes besides COVID-19, aiming at extending AI-based RI detection. We also collected control data from hospital patients without RI. We show that the considered models, when trained on P1, do not generalize to P2, indicating that COVID-19 RI has features that may not be found in all RI types.
format Preprint
id arxiv_https___arxiv_org_abs_2405_17569
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Discriminant audio properties in deep learning based respiratory insufficiency detection in Brazilian Portuguese
Gauy, Marcelo Matheus
Berti, Larissa Cristina
Cândido Jr, Arnaldo
Neto, Augusto Camargo
Goldman, Alfredo
Levin, Anna Sara Shafferman
Martins, Marcus
de Medeiros, Beatriz Raposo
Queiroz, Marcelo
Sabino, Ester Cerdeira
Svartman, Flaviane Romani Fernandes
Finger, Marcelo
Machine Learning
Artificial Intelligence
Sound
Audio and Speech Processing
This work investigates Artificial Intelligence (AI) systems that detect respiratory insufficiency (RI) by analyzing speech audios, thus treating speech as a RI biomarker. Previous works collected RI data (P1) from COVID-19 patients during the first phase of the pandemic and trained modern AI models, such as CNNs and Transformers, which achieved $96.5\%$ accuracy, showing the feasibility of RI detection via AI. Here, we collect RI patient data (P2) with several causes besides COVID-19, aiming at extending AI-based RI detection. We also collected control data from hospital patients without RI. We show that the considered models, when trained on P1, do not generalize to P2, indicating that COVID-19 RI has features that may not be found in all RI types.
title Discriminant audio properties in deep learning based respiratory insufficiency detection in Brazilian Portuguese
topic Machine Learning
Artificial Intelligence
Sound
Audio and Speech Processing
url https://arxiv.org/abs/2405.17569