Classification of Pneumonia images on mobile devices with Quantized Neural Network

Detalhes bibliográficos
Autor(a) principal: Sousa, Jose Vigno Moura
Data de Publicação: 2020
Outros Autores: Almeida, Vilson Rosa de, Saraiva, Aratã Andrade, Santos, Domingos Bruno Sousa, Pimentel, Pedro Mateus Cunha, Sousa, Luciano Lopes de
Tipo de documento: Artigo
Idioma: eng
Título da fonte: Research, Society and Development
Texto Completo: https://rsdjournal.org/index.php/rsd/article/view/8382
Resumo: This paper presents an approach for the classification of child chest X-ray images into two classes: pneumonia and normal. We employ Convolutional Neural Networks, from pre-trained networks together with a quantization process, using the platform TensorFlow Lite method. This reduces the processing requirement and computational cost. Results have shown accuracy up to 95.4% and 94.2% for MobileNetV1 and MobileNetV2, respectively. The resulting mobile app also presents a simple and intuitive user interface.
id UNIFEI_6fd758901b7107bf917316217ce4b778
oai_identifier_str oai:ojs.pkp.sfu.ca:article/8382
network_acronym_str UNIFEI
network_name_str Research, Society and Development
repository_id_str
spelling Classification of Pneumonia images on mobile devices with Quantized Neural NetworkClasificación de imágenes de neumonía en dispositivos móviles con red neuronal cuantificadaClassificação de imagens de Pneumonia em dispositivos móveis com Rede Neural QuantizadaClasificaciónImágenesCuantizaciónDispositivos móvilesNeumonía.ClassificaçãoImagensQuantizaçãoDispositivos MóveisPneumonia.ClassificationImagesQuantizationMobile DevicesPneumonia.This paper presents an approach for the classification of child chest X-ray images into two classes: pneumonia and normal. We employ Convolutional Neural Networks, from pre-trained networks together with a quantization process, using the platform TensorFlow Lite method. This reduces the processing requirement and computational cost. Results have shown accuracy up to 95.4% and 94.2% for MobileNetV1 and MobileNetV2, respectively. The resulting mobile app also presents a simple and intuitive user interface.Este artículo presenta un enfoque para clasificar las imágenes de rayos X de tórax de los niños en dos clases: neumonía y normal. Usamos redes neuronales convolucionales, de redes pre-entrenadas junto con un proceso de cuantificación, utilizando el método de la plataforma TensorFlow Lite. Esto reduce los requisitos de procesamiento y el costo computacional. Los resultados mostraron una precisión de hasta 95,4% y 94,2% para MobileNetV1 y MobileNetV2, respectivamente. La aplicación móvil resultante también cuenta con una interfaz de usuario sencilla e intuitiva.Este artigo apresenta uma abordagem para a classificação de imagens de radiografias de tórax de crianças em duas classes: pneumonia e normal. Empregamos Redes Neurais Convolucionais, a partir de redes pré-treinadas em conjunto com um processo de quantização, utilizando o método da plataforma TensorFlow Lite. Isso reduz a necessidade de processamento e o custo computacional. Os resultados mostraram precisão de até 95,4% e 94,2% para MobileNetV1 e MobileNetV2, respectivamente. O aplicativo móvel resultante também apresenta uma interface de usuário simples e intuitiva.Research, Society and Development2020-09-19info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfhttps://rsdjournal.org/index.php/rsd/article/view/838210.33448/rsd-v9i10.8382Research, Society and Development; Vol. 9 No. 10; e889108382Research, Society and Development; Vol. 9 Núm. 10; e889108382Research, Society and Development; v. 9 n. 10; e8891083822525-3409reponame:Research, Society and Developmentinstname:Universidade Federal de Itajubá (UNIFEI)instacron:UNIFEIenghttps://rsdjournal.org/index.php/rsd/article/view/8382/7392Copyright (c) 2020 Jose Vigno Moura Sousa; Vilson Rosa de Almeida; Aratã Andrade Saraiva; Domingos Bruno Sousa Santos; Pedro Mateus Cunha Pimentel; Luciano Lopes de Sousahttps://creativecommons.org/licenses/by/4.0info:eu-repo/semantics/openAccessSousa, Jose Vigno Moura Almeida, Vilson Rosa de Saraiva, Aratã Andrade Santos, Domingos Bruno Sousa Pimentel, Pedro Mateus CunhaSousa, Luciano Lopes de 2020-10-31T12:03:23Zoai:ojs.pkp.sfu.ca:article/8382Revistahttps://rsdjournal.org/index.php/rsd/indexPUBhttps://rsdjournal.org/index.php/rsd/oairsd.articles@gmail.com2525-34092525-3409opendoar:2024-01-17T09:30:52.209364Research, Society and Development - Universidade Federal de Itajubá (UNIFEI)false
dc.title.none.fl_str_mv Classification of Pneumonia images on mobile devices with Quantized Neural Network
Clasificación de imágenes de neumonía en dispositivos móviles con red neuronal cuantificada
Classificação de imagens de Pneumonia em dispositivos móveis com Rede Neural Quantizada
title Classification of Pneumonia images on mobile devices with Quantized Neural Network
spellingShingle Classification of Pneumonia images on mobile devices with Quantized Neural Network
Sousa, Jose Vigno Moura
Clasificación
Imágenes
Cuantización
Dispositivos móviles
Neumonía.
Classificação
Imagens
Quantização
Dispositivos Móveis
Pneumonia.
Classification
Images
Quantization
Mobile Devices
Pneumonia.
title_short Classification of Pneumonia images on mobile devices with Quantized Neural Network
title_full Classification of Pneumonia images on mobile devices with Quantized Neural Network
title_fullStr Classification of Pneumonia images on mobile devices with Quantized Neural Network
title_full_unstemmed Classification of Pneumonia images on mobile devices with Quantized Neural Network
title_sort Classification of Pneumonia images on mobile devices with Quantized Neural Network
author Sousa, Jose Vigno Moura
author_facet Sousa, Jose Vigno Moura
Almeida, Vilson Rosa de
Saraiva, Aratã Andrade
Santos, Domingos Bruno Sousa
Pimentel, Pedro Mateus Cunha
Sousa, Luciano Lopes de
author_role author
author2 Almeida, Vilson Rosa de
Saraiva, Aratã Andrade
Santos, Domingos Bruno Sousa
Pimentel, Pedro Mateus Cunha
Sousa, Luciano Lopes de
author2_role author
author
author
author
author
dc.contributor.author.fl_str_mv Sousa, Jose Vigno Moura
Almeida, Vilson Rosa de
Saraiva, Aratã Andrade
Santos, Domingos Bruno Sousa
Pimentel, Pedro Mateus Cunha
Sousa, Luciano Lopes de
dc.subject.por.fl_str_mv Clasificación
Imágenes
Cuantización
Dispositivos móviles
Neumonía.
Classificação
Imagens
Quantização
Dispositivos Móveis
Pneumonia.
Classification
Images
Quantization
Mobile Devices
Pneumonia.
topic Clasificación
Imágenes
Cuantización
Dispositivos móviles
Neumonía.
Classificação
Imagens
Quantização
Dispositivos Móveis
Pneumonia.
Classification
Images
Quantization
Mobile Devices
Pneumonia.
description This paper presents an approach for the classification of child chest X-ray images into two classes: pneumonia and normal. We employ Convolutional Neural Networks, from pre-trained networks together with a quantization process, using the platform TensorFlow Lite method. This reduces the processing requirement and computational cost. Results have shown accuracy up to 95.4% and 94.2% for MobileNetV1 and MobileNetV2, respectively. The resulting mobile app also presents a simple and intuitive user interface.
publishDate 2020
dc.date.none.fl_str_mv 2020-09-19
dc.type.driver.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.uri.fl_str_mv https://rsdjournal.org/index.php/rsd/article/view/8382
10.33448/rsd-v9i10.8382
url https://rsdjournal.org/index.php/rsd/article/view/8382
identifier_str_mv 10.33448/rsd-v9i10.8382
dc.language.iso.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv https://rsdjournal.org/index.php/rsd/article/view/8382/7392
dc.rights.driver.fl_str_mv https://creativecommons.org/licenses/by/4.0
info:eu-repo/semantics/openAccess
rights_invalid_str_mv https://creativecommons.org/licenses/by/4.0
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Research, Society and Development
publisher.none.fl_str_mv Research, Society and Development
dc.source.none.fl_str_mv Research, Society and Development; Vol. 9 No. 10; e889108382
Research, Society and Development; Vol. 9 Núm. 10; e889108382
Research, Society and Development; v. 9 n. 10; e889108382
2525-3409
reponame:Research, Society and Development
instname:Universidade Federal de Itajubá (UNIFEI)
instacron:UNIFEI
instname_str Universidade Federal de Itajubá (UNIFEI)
instacron_str UNIFEI
institution UNIFEI
reponame_str Research, Society and Development
collection Research, Society and Development
repository.name.fl_str_mv Research, Society and Development - Universidade Federal de Itajubá (UNIFEI)
repository.mail.fl_str_mv rsd.articles@gmail.com
_version_ 1797052740601380864

Registros relacionados