Scientists use AI and machine learning to detect cancer with more accuracy

UCLA scientists say they can accurately identify cancer cells in bloodstreams

Science

Scientists in California have developed a microscope that uses machine learning and photonic time stretch to locate cancer cells more efficiently.

The microscope, invented by UCLA scientists, uses artificial intelligence (AI) to analyse 36 million images per second, more accurately identifying cancer cells present in blood samples.

Currently, doctors track cancerous cells by adding biochemicals to blood samples, which then add labels to them for later detection.

However, this can damage the cells, making the samples useless. Other techniques include identifying cancer cells by their physical characteristics, but they can be inaccurate. 

The AI microscope images cells without destroying them and can identify characteristics such as size, granularity and biomass.

This is possible because of a photonic time stretch microscope invented by the UCLA scientists, which has allowed cell identification to reach 95 per cent accuracy.

The photonic time stretch works by taking pictures of flowing blood cells with laser bursts, boosting clarity and slowing them down enough to be detected and digitised.

Commenting on the technology, Ata Mahjoubfar, said: "Each frame is slowed down in time and optically amplified so it can be digitised. This lets us perform fast cell imaging that the artificial intelligence component can distinguish."

"The photonic time stretch technique allows us to identify rogue cells in a short time with low-level illumination," added Chen.

Featured Resources

Seven steps to connect and empower your frontline workers

How business leaders can improve communication with a secure platform

Free download

Create what’s next

The future of collaboration and productivity

Free Download

Leveraging the cloud without relinquishing control

Your data. Their cloud.

Free download

Re-architecting for nonstop innovation

Unlocking productivity, scalability, and lower costs for cloud natives

Free Download

Most Popular

Looking beyond the obvious: What’s best for multi-cloud?
Sponsored

Looking beyond the obvious: What’s best for multi-cloud?

8 Nov 2021
Nike to take customers into the metaverse with 'NIKELAND'
virtualisation

Nike to take customers into the metaverse with 'NIKELAND'

19 Nov 2021
How to speed up Microsoft's Windows 11
Microsoft Windows

How to speed up Microsoft's Windows 11

9 Nov 2021