Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/132630
Title: Screening Chest X-rays for Covid-19 with Deep Learning
Author: Robert Gill, Eric
Tutor: Nuñez Do Rio, Joan Manuel
Others: Arnedo-Moreno, Joan  
Abstract: The novel Coronavirus has caused a global pandemic with high economic and social costs. The emergent nature of the Coronavirus' associated disease, Covid-19, found many nations ill-prepared to control its spread, leading to high rates of infection and strain on health systems. Rapid detection and the subsequent quarantine of infected people is the most effective measure against the spread of the virus outside of vaccination. Diagnosis is generally carried out by Reverse Polymer or antigens tests which can be expensive and not always readily available. These methods require specialised staff and physical contact with the patient, as well as time to process results. X-ray machines are available in hospitals across the world in countries of nearly all economic situations. Radiography has been used in many screening and diagnostic use cases and there is wide-spread investigation into its applicability in the Coronavirus pandemic. This project investigates the viability of employing Convolutional Neural Networks to screen radiographic images of lungs for symptoms of Covid-19. X-rays are massively available, economical, and non-invasive. The results show that Convolutional Neural Networks can classify x-ray images into Covid, Normal and Viral Pneumonia classes with high levels of Precision and Recall.
Keywords: COVID-19
X-rays
Convolutional Neural Network
Document type: info:eu-repo/semantics/bachelorThesis
Issue Date: 13-Jun-2021
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Bachelor thesis, research projects, etc.

Files in This Item:
File Description SizeFormat 
egill31TFG0621memory.pdfMemory of TFG2 MBAdobe PDFThumbnail
View/Open
egill31TFG0621presentation.pdfPresentation of TFG1,47 MBAdobe PDFThumbnail
View/Open