COVID-19: This AI network could help detect cases using chest X-ray images

By: Tech Desk | New Delhi |

Published: March 27, 2020 2:00:40 pm

This AI tool could help in detecting cases of COVID-19, using chest X-Ray images. (Image credit: arXiv)

As the COVID-19 pandemic continues and more cases come to light globally, the use of artificial intelligence or AI-based tools to help detect the disease is being explored more extensively. The latest tool that wants to help in the early detection of potential coronavirus cases is COVID-Net, which is a deep neural network that will look at chest radiography images to determine whether a patient is infected.

Researchers part of the Canadian startup DarwinAI Corp along with researchers Linda Wang and Alexander Wong at the University of Waterloo, Canada have designed this AI-system to help with the detection of COVID-19 cases, and published details in a research paper.

Read more: From Kerala, a robot to take care of coronavirus patients | Cold, flu or Covid-19? This AI-enabled risk assessment tool can help you differentiate

One difference is that the system is open source and available to the general public. In China too, researchers at Alibaba’s DAMO academy have been working on AI tools which rely on the chest x-rays and CT-scans to help detect and diagnose COVID-19 positive cases.

One reason why AI tools are being proposed is the spread of the pandemic and how it is quickly overwhelming doctors and healthcare systems across the world. With time being, crucial AI could help shorten the period for detection and positive confirmation of a COVID-19 case and provide assistance to doctors.

According to the paper submitted by the Canadian researchers, “effective screening of infected patients” could help. One way is to look at radiological images for the chest as other studies have shown that COVID-19 patients do present abnormalities in chest radiography images, which are characteristic of the infection.

Express Tech is now on Telegram. Click here to join our channel (@expresstechie) and stay updated with the latest tech news

The researchers have used an open source chest radiography dataset for training their COVID-Net. The researchers claim the COVID-Net network had a strong 80 per cent prediction for detecting the COVID-19 cases in the data sets, though they did warn the number of positive cases in this data set was limited. The COVIDx database being used by the researchers only has “68 radiography images from 45 COVID-19 patient cases,” according to the paper.

“A better view of effectiveness will improve as more COVID-19 patient cases become available,” note the researchers. There were also false positives by the system.

The researchers do caution that the network is not yet a full ready solution and more work is needed. They note in the paper, “By no means a production-ready solution, the hope is that the open access COVID-Net, along with the description on constructing the open source COVIDx dataset, will be leveraged and build upon by both researchers and citizen data scientists alike to accelerate the development of highly accurate yet practical deep learning solutions for detecting COVID-19 cases and accelerate treatment of those who need it the most.”

📣 The Indian Express is now on Telegram. Click here to join our channel (@indianexpress) and stay updated with the latest headlines

For all the latest Technology News, download Indian Express App.

© IE Online Media Services Pvt Ltd

Leave a Comment