Current limitations to identify COVID-19 using artificial intelligence with chest X-ray imaging(Part II). The shortcut learning problem

Vista sencilla de ítem
dc.contributor.authorLópez Cabrera, José Daniel
dc.contributor.authorOrozco Morales, Rubén
dc.contributor.authorPortal Díaz, Jorge Armando
dc.contributor.authorPerez Diaz, Marlen
dc.contributor.authorLovelle Enríquez, Orlando
dc.contributor.departmentUniversidad Central "Marta Abreu" de Las Villas. Dpto de Automáticaen_US
dc.contributor.otherDepartamento Imagenología Hospital Manuel Fajardo Riveroen_US
dc.date.accessioned2022-02-17T15:52:53Z
dc.date.available2022-02-17T15:52:53Z
dc.date.issued2021-10-10
dc.description.abstractSince the outbreak of the COVID-19 pandemic, computer vision researchers have been working on automatic identification of this disease using radiological images. The results achieved by automatic classification methods far exceed those of human specialists, with sensitivity as high as 100% being reported. However, prestigious radiology societies have stated that the use of this type of imaging alone is not recommended as a diagnostic method. According to some experts the patterns presented in these images are unspecific and subtle, overlapping with other viral pneumonias. This report seeks to evaluate the analysis the robustness and generalizability of different approaches using artificial intelligence, deep learning and computer vision to identify COVID-19 using chest X-rays images. We also seek to alert researchers and reviewers to the issue of “shortcut learning”. Recommendations are presented to identify whether COVID-19 automatic classification models are being affected by shortcut learning. Firstly, papers using explainable artificial intelligence methods are reviewed. The results of applying external validation sets are evaluated to determine the generalizability of these methods. Finally, studies that apply traditional computer vision methods to perform the same task are considered. It is evident that using the whole chest X-Ray image or the bounding box of the lungs, the image regions that contribute most to the classification appear outside of the lung region, something that is not likely possible. In addition, although the investigations that evaluated their models on data sets external to the training set, the effectiveness of these models decreased significantly, it may provide a more realistic representation as how the model will perform in the clinic. The results indicate that, so far, the existing models often involve shortcut learning, which makes their use less appropriate in the clinical setting.en_US
dc.identifier.doi10.1007/s12553-021-00609-8en_US
dc.identifier.urihttps://dspace.uclv.edu.cu/handle/123456789/13432
dc.language.isoen_USen_US
dc.relation.journalHealth and Technologyen_US
dc.source.endpage424en_US
dc.source.initialpage411en_US
dc.source.issue2en_US
dc.source.volume11en_US
dc.subjectCOVID-19en_US
dc.subjectChest X-Raysen_US
dc.subjectArtificial intelligenceen_US
dc.subjectDeep Learningen_US
dc.titleCurrent limitations to identify COVID-19 using artificial intelligence with chest X-ray imaging(Part II). The shortcut learning problemen_US
dc.typeArticleen_US
dc.type.article1en_US

Archivos

Bloque original
Mostrando 1 - 1 de 1
Miniatura
Nombre:
Lpez-Cabrera2021_Article_CurrentLimitationsToIdentifyCo.pdf
Tamaño:
1.05 MB
Formato:
Adobe Portable Document Format
Descargar
Bloque de licencias
Mostrando 1 - 1 de 1
No hay miniatura disponible
Nombre:
license.txt
Tamaño:
3.33 KB
Formato:
Item-specific license agreed upon to submission
Descripción:
Descargar

Colecciones

Artículos - Centro de Estudios de Mecánica Computacional Aplicada a Ingeniería