您现在的位置是:Google's AI microscope ARM to detect cancer cells in seconds >>正文

Google's AI microscope ARM to detect cancer cells in seconds

上海品茶网 - 夜上海最新论坛社区 - 上海千花论坛8633人已围观

简介By subscribing, you agree to our Terms of Use and Policies You may unsubscribe at any time.Google an...

By subscribing, you agree to our Terms of Use and Policies You may unsubscribe at any time.

Google and the US Department of Defense (DoD) have teamed up to build an artificial intelligence-backed microscope that can help in cancer pathology. They have been working on the microscope for years.

According to a report by CNBC, the AI tool is called Augmented Reality Microscope (ARM), and given its cost estimated between $90,000 to $100,000, it is a hopeful tool for small hospitals and facilities for early detection and treatment of cancer.

Google has also applied for a patent for its latest AI tool.

Google's AI microscope ARM to detect cancer cells in seconds
Google's ARM

CNBC 

Cancer is a leading cause of death worldwide, accounting for nearly 10 million deaths in 2020. The most common cancer cases are related to the breast, lung, colon, prostate, skin, and stomach. Each year, approximately 400,000 children develop cancer. While the most common type of cancer varies between countries, cervical cancer is the most common in 23 countries.

See Also Related
  • Google DeepMind's new AI tool can predict genetic diseases 
  • Google's android alert failed to notify people of Turkey earthquakes: BBC investigation 
  • Google is secretly pitching its AI tool to news organizations 

Google’s ARM can flag cancer in seconds 

How this microscope works is that it enhances the ‘heat map’ in the field of view when likely cancer tissue is placed in the sample space and is observed under it. The heat map helps a viewer, a doctor, characterize the sample as it directs their attention to areas of interest that are particularly likely to be cancerous.

What’s more, if the pathologist wants to zoom in on the heat map area to get a new field of view of the sample, a new image would be captured in real time, and a new heat map would be overlaid on the field of view. This would help the doctor/pathologist further investigate the sample.

The AI factor comes in when the digital images captured by the microscope’s camera are supplied to a computing unit. The compute unit includes a machine learning pattern recognizer trained on large data sets - in this case, microscope slide images of cancerous specimens. The pattern recognizer is trained to identify regions of interest in an image, like cancerous cells or tissue, viruses or bacteria, eggs from parasites, etc.

A couple of years before we see ARM in local hospitals

The CNBC report says that ARM has not been fully developed and is still in its early days. But, initial research is promising and could prove to be a valuable tool for pathologists without easy access to a second opinion.

A fundamental problem with cancer is that it is often diagnosed late. Even in developed countries with state-of-the-art health systems and machines in place, cancer is often detected at the last stage. This makes the doctor's job much more difficult as cancer is more easily treated when detected early.

There exist 13 ARMs in the world, one operational at a Mitre facility just outside of Washington, DC. Mitre is a nonprofit that works with government agencies to tackle big problems involving technology, said the CNBC report.

Intended to serve as a second line of defense for pathologists, Dr Niels Olson told CNBC, “It’s not just your job to say ‘This is cancer, it’s this kind of cancer.’ Part of the job is saying, ‘It’s absolutely not cancer,’ which can be nerve-wracking when you’re alone. I would have loved to have an Augmented Reality Microscope in Guam, just so there’d be somebody else helping.”

Dr. Olson is the chief medical officer of the DoD’s Defense Innovation Unit, or DIU, created in 2015 to help the military integrate and bring up-to-speed commercial cutting-edge technology to the public.

Tags:

相关文章



友情链接