This AI tool can help doctors treat brain tumours quickly and accurately, study finds

Artificial intelligence has been the buzzword across the globe for the last year and a half. AI technology is breaking new barriers by the minute and a new study released this week by Harvard Medical School, promises another breakthrough, AI-assisted brain surgery. 

For decades, neurosurgeons gave struggled to understand the tumours (or glioma, the most common type of brain tumour). To perform surgery to remove the tumour, the doctors need a host of information, so as not to damage the surrounding brain tissue. 

They take a piece of sample and send it to a pathology lab to get real-time, immediate feedback. The pathologist runs the tests and informs them whether they are cutting the correct issue or what kind of cancer the patient has.

Usually, the entire process takes 15 minutes in state-of-the-art medical facilities. However, the analysis is not always accurate and it happens when the patient’s skull is wide open on the surgical table. 

“This process is not error proof,” Dr. Kun-Hsing Yu, a professor at Harvard Medical School, told the Guardian.

“People are under stress, and the quality of the slide is sometimes not great, so occasionally we will have misdiagnosis arising from this fast process,” he added.

Developing the AI tool 

To minimise the error and time wastage, Yu and his team turned their attention towards machine learning in which AI technology learns patterns without explicit instructions from a programmer and can help make the analysis easier. 

The team developed an AI tool named Cryosection Histopathology Assessment and Review Machine, also called CHARM. The tech was developed using more than 2,300 brain tumour samples from 1,524 people with glioma. These samples “taught” CHARM what to look for when analysing tumour samples.

Most importantly, CHARM is fast as it provided accurate analysis of some tumour cells in less than one second. 

Might take some time for clinical use

However, the technology in the study might not be ready for clinical use for several years as it needs approvals from concerned authorities, according to Yu. 

The study is being funded by the National Institute of General Medical Sciences, Google Research Scholar Award, the Blavatnik Center for Computational Biomedicine Award, the Partners’ Innovation Discovery Grant, and the Schlager Family Award for Early Stage Digital Health Innovations.

“Just like human clinicians who must engage in ongoing education and training. AI tools must keep up with the latest knowledge to remain at peak performance,” said Yu. 

(With inputs from agencies)

No Byline Policy

Editorial Guidelines

Corrections Policy

Source