Shulou(Shulou.com)11/24 Report--

Thanks to CTOnews.com netizens Wu Yanzu in South China for the delivery of clues! CTOnews.com, September 10 (Xinhua)-- Microsoft has announced that it will work with digital pathology provider Paige to build the world's largest image-based artificial intelligence model for cancer recognition.

According to the press release, the artificial intelligence model is being trained for an "unprecedented amount of data", including billions of images.

According to reports, it can identify common cancers and rare cancers that are difficult to diagnose, and researchers hope it will eventually help doctors cope with understaffing and a growing number of patients.

Paige has developed digital and AI-driven solutions for pathologists who conduct laboratory tests on body fluids and tissues to make a diagnosis.

"you don't have cancer until the pathologist says there is cancer. This is a critical step in the entire medical system," Thomas Fuchs, co-founder and chief scientist of Paige, told CNBC.

Fuchs says the workflow of pathologists has barely changed in the past 150 years. For example, to diagnose cancer, they look at a piece of tissue on a slide under a microscope. Although this method has been proven and mature, it can have serious consequences for patients if something is left out during the period.

As a result, Paige has been working to digitize the workflows of pathologists to improve accuracy and efficiency in this area of expertise.

The CTOnews.com query found that Paige's FullFocus has been approved by the U.S. Food and Drug Administration (FDA) to help pathologists view digital slides directly on the screen without relying on microscopes.

In addition, Paige has created an AI model that helps pathologists identify breast, colon and prostate cancers on the screen.

In fact, Paige is the only company approved by FDA to allow pathologists to use AI as an auxiliary tool to identify prostate cancer. Andy Moy, its CEO, says this may be due in part to obstacles related to storage costs and data collection.

Digitizing a single slide may require more storage space than 1GB, so the infrastructure and costs associated with large-scale data collection can increase rapidly. Storage costs can be a big hindrance for smaller health systems, which is why only wealthy academic centers have traditionally been able to invest in digital pathology, Foss says.

To illustrate its scale, Paige says it has 10 times more data than Netflix, including all the shows and movies that exist on the platform. But in order to expand its operations and build an AI tool that can identify more types of cancer, Paige turned to Microsoft.

For nearly a year and a half, Paige has been using Microsoft's cloud storage and supercomputing infrastructure to build an advanced new AI model.

Paige's original AI model is said to have used more than 1 billion images from 500000 pathological sections, but Foss said the model developed by the company in collaboration with Microsoft was "orders of magnitude larger than anything available". The model is being trained using 4 million slides to identify common and rare cancers that are often difficult to diagnose. Paige says this is the largest computer vision model ever announced.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.