Using AI, X-rays are now better at detecting explosives

Using AI, X-rays could also find cancerous tumors and cracks in aircraft.

Ben Wodecki

September 29, 2022

1 Min Read

Using AI, X-rays could also find cancerous tumors and cracks in aircraft.

A team of U.K.-based scientists has created a deep learning AI model that can allow X-rays to scan luggage to find small traces of explosives.

Researchers from the University College London collaborated with Nylers and XPCI to develop the deep learning system.

Prior research found that X-rays produce bends when they strike an object that differs depending on the material.

Taking note, the researchers taught a deep learning AI model to recognize the bends produced when scanning explosive materials.

Upon completion of training, the model was tasked was scanning various objects, some embedded with explosives. The AI-augmented X-ray machine was able to identify 100% of the hidden explosives.

The X-ray device identified bends as tiny as a single microradian, the equivalent of one-20,000th of a degree.

Upon publishing the results of the tests in the journal Nature, the researchers suggested it could be applied to other, less explosive industries.

They opined that the AI-powered X-ray could be used in health care settings to detect cancerous tumors. Or in aeronautics to identify cracks on airplane surfaces.

“While the small scale of these studies means further research is required, results indicate potential for a combined use of dark-field and deep neural networks in security applications and beyond,” the findings read.

Credit: Chalabala/ Getty

About the Authors

Ben Wodecki

Assistant Editor

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