Searching for pneumothorax in x-ray images using autoencoded deep features

Abstract Fast diagnosis and treatment of pneumothorax, a collapsed or dropped lung, is crucial to avoid fatalities. Pneumothorax is typically detected on a chest X-ray image through visual inspection by experienced radiologists. However, the detection rate is quite low due to the complexity of visua...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Antonio Sze-To, Abtin Riasatian, H. R. Tizhoosh
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/98e93648952c4607a708cadc551e6efc
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:98e93648952c4607a708cadc551e6efc
record_format	dspace
spelling	oai:doaj.org-article:98e93648952c4607a708cadc551e6efc2021-12-02T16:51:03ZSearching for pneumothorax in x-ray images using autoencoded deep features10.1038/s41598-021-89194-42045-2322https://doaj.org/article/98e93648952c4607a708cadc551e6efc2021-05-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-89194-4https://doaj.org/toc/2045-2322Abstract Fast diagnosis and treatment of pneumothorax, a collapsed or dropped lung, is crucial to avoid fatalities. Pneumothorax is typically detected on a chest X-ray image through visual inspection by experienced radiologists. However, the detection rate is quite low due to the complexity of visual inspection for small lung collapses. Therefore, there is an urgent need for automated detection systems to assist radiologists. Although deep learning classifiers generally deliver high accuracy levels in many applications, they may not be useful in clinical practice due to the lack of high-quality and representative labeled image sets. Alternatively, searching in the archive of past cases to find matching images may serve as a “virtual second opinion” through accessing the metadata of matched evidently diagnosed cases. To use image search as a triaging or diagnosis assistant, we must first tag all chest X-ray images with expressive identifiers, i.e., deep features. Then, given a query chest X-ray image, the majority vote among the top k retrieved images can provide a more explainable output. In this study, we searched in a repository with more than 550,000 chest X-ray images. We developed the Autoencoding Thorax Net (short AutoThorax -Net) for image search in chest radiographs. Experimental results show that image search based on AutoThorax -Net features can achieve high identification performance providing a path towards real-world deployment. We achieved 92% AUC accuracy for a semi-automated search in 194,608 images (pneumothorax and normal) and 82% AUC accuracy for fully automated search in 551,383 images (normal, pneumothorax and many other chest diseases).Antonio Sze-ToAbtin RiasatianH. R. TizhooshNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-13 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Antonio Sze-To Abtin Riasatian H. R. Tizhoosh Searching for pneumothorax in x-ray images using autoencoded deep features
description	Abstract Fast diagnosis and treatment of pneumothorax, a collapsed or dropped lung, is crucial to avoid fatalities. Pneumothorax is typically detected on a chest X-ray image through visual inspection by experienced radiologists. However, the detection rate is quite low due to the complexity of visual inspection for small lung collapses. Therefore, there is an urgent need for automated detection systems to assist radiologists. Although deep learning classifiers generally deliver high accuracy levels in many applications, they may not be useful in clinical practice due to the lack of high-quality and representative labeled image sets. Alternatively, searching in the archive of past cases to find matching images may serve as a “virtual second opinion” through accessing the metadata of matched evidently diagnosed cases. To use image search as a triaging or diagnosis assistant, we must first tag all chest X-ray images with expressive identifiers, i.e., deep features. Then, given a query chest X-ray image, the majority vote among the top k retrieved images can provide a more explainable output. In this study, we searched in a repository with more than 550,000 chest X-ray images. We developed the Autoencoding Thorax Net (short AutoThorax -Net) for image search in chest radiographs. Experimental results show that image search based on AutoThorax -Net features can achieve high identification performance providing a path towards real-world deployment. We achieved 92% AUC accuracy for a semi-automated search in 194,608 images (pneumothorax and normal) and 82% AUC accuracy for fully automated search in 551,383 images (normal, pneumothorax and many other chest diseases).
format	article
author	Antonio Sze-To Abtin Riasatian H. R. Tizhoosh
author_facet	Antonio Sze-To Abtin Riasatian H. R. Tizhoosh
author_sort	Antonio Sze-To
title	Searching for pneumothorax in x-ray images using autoencoded deep features
title_short	Searching for pneumothorax in x-ray images using autoencoded deep features
title_full	Searching for pneumothorax in x-ray images using autoencoded deep features
title_fullStr	Searching for pneumothorax in x-ray images using autoencoded deep features
title_full_unstemmed	Searching for pneumothorax in x-ray images using autoencoded deep features
title_sort	searching for pneumothorax in x-ray images using autoencoded deep features
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/98e93648952c4607a708cadc551e6efc
work_keys_str_mv	AT antonioszeto searchingforpneumothoraxinxrayimagesusingautoencodeddeepfeatures AT abtinriasatian searchingforpneumothoraxinxrayimagesusingautoencodeddeepfeatures AT hrtizhoosh searchingforpneumothoraxinxrayimagesusingautoencodeddeepfeatures
_version_	1718382995176423424

Searching for pneumothorax in x-ray images using autoencoded deep features

Ejemplares similares