AI Diagnoses Lung Diseases in X-Rays

Deep convolutional neural networks can detect pneumonia, tuberculosis and COVID.

Helen Hwang

November 10, 2022

1 Min Read
X-ray of lung with heart attack

A new AI-powered X-ray method could drastically reduce the time it takes to diagnose various lung diseases.

Traditionally, tests for lung diseases require a battery of tests, including X-rays, CT scans, ultrasounds and blood tests.

Researchers at the University of the West of Scotland (UWS) have come up with a way of using AI to get a diagnosis in a matter of minutes using only an X-ray and applying a neural network.

The scientists did a comparative analysis of patient scants who had COVID, tuberculosis and pneumonia. A deep convolutional neural network ordinarily used to look at images was used to make a diagnosis of serious lung diseases.

They found the method to be 98% accurate after extensive testing.

The tool was initially used to detect COVID from X-ray images, but the minds behind it saw how beneficial the AI method could be applied to other lung ailments.

Lead researcher Naeem Ramzan, professor of computing engineering and chair of Affective and Human Computing for Smart Environment Research Centre at UWS, said AI could aid hospital staff in alleviating some of their burdens. UWS Ph.D. students Gabriel Okolo and Stamos Katsigiannis also worked on the project.

“X-ray imaging is a relatively cheap and accessible diagnostic tool that already assists in the diagnosis of various conditions, including pneumonia, tuberculosis and COVID. Recent advances in AI have made automated diagnosis using chest X-ray scans a very real prospect in medical settings,” said Professor Ramzan.

In the future, the AI model is being applied to help detect cancer tumors.

About the Authors

Helen Hwang

Freelance reporter, AI Business

Get the newsletter
From automation advancements to policy announcements, stay ahead of the curve with the bi-weekly AI Business newsletter.