google develops augmented reality microscope for cancer detection

Machine learning is one of the fastest growing fields of study today. Google uses machine learning in almost every area of the company. From Google Translate to search rankings with RankBrain, machine learning is embedded deep within Google. Now, the tech giant is experimenting with machine learning in medicine and health.

In Google’s latest research blog post, the Google Brain Team has unveiled a prototype for an Augmented Reality microscope, which uses machine learning to detect cancer. The Google Brain Team is a division within the company responsible for AI research, and this is just the latest in the many projects that the team is working on.

Current cancer diagnosis is mostly achieved through the use of compound light microscopes. Newer deep learning methods require a digital representation of the tissue in order to determine if cancer cells are present.

This new Augmented Reality Microscope (ARM) from Google combines both methods. Jeff Dean, Google Brain Team lead says:

How does the ARM work?

The Google Brain Team outlined in their research on the ARM:

The platform consists of a modified light microscope that enables real-time image analysis and presentation of the results of machine learning algorithms directly into the field of view.

According to the research, the ARM can be fitted “into existing light microscopes found in hospitals and clinics around the world using low-cost, readily-available components, and without the need for whole slide digital versions of the tissue being analyzed.”

Just like with a regular microscope, the user looks into the eyepiece. However, with the help of an AR display and a camera to detect field of view (FoV), the ARM will use machine learning to project a real-time, superimposed image of the specimen. This image will help users to highlight different points of interest using contours, arrows, and more.

google creates new AR microscope to detect cancer

The Augmented Reality Microscope can be configured to detect breast and prostate cancer by outlining areas of interest to pathologists. It has many other different uses, including being “capable of running many types of machine learning algorithms aimed at solving different problems such as object detection, quantification, or classification.”

Read More:

Apple’s latest machine learning entry focuses on the “Hey Siri” command

The researchers believe that “the ARM has potential for a large impact on global health, particularly for the diagnosis of infectious diseases, including tuberculosis and malaria, in developing countries.”

The biggest feature of this breakthrough is that the ARM is compatible with other computational components and deep learning models that are built using TensorFlow, allowing for a cost-effective method to deliver cancer treatments to third-world countries.



Please enter your comment!
Please enter your name here