Loading...
Loading...
Browse all stories on DeepNewz
VisitWhat will be the market share of the Virchow2G model in the AI pathology tools market by December 31, 2024?
0-10% • 25%
11-20% • 25%
21-30% • 25%
31%+ • 25%
Market research reports from firms like Gartner, IDC, etc.
Microsoft and Paige Launch Virchow2G AI Models for Cancer Diagnosis, Scaling to 1.85 Billion Parameters
Aug 15, 2024, 01:31 PM
Microsoft and Paige have collaborated to develop Virchow2 and Virchow2G, second-generation foundation models for computational pathology. These models are designed to enhance cancer diagnosis by leveraging AI and self-supervised learning techniques. The Virchow2G model significantly increases the scale of previous work, reaching 1.85 billion parameters. A paper published in Nature Medicine highlights the original Virchow model, trained on 1.5 million whole-slide images from 10,000 patients, which has improved the detection of rare cancers. This advancement represents a strategic pivot for Paige, aiming to further the role of AI in cancer care. Virchow2 and 2G are now available on Hugging Face.
View original story
Less than 10% • 33%
10% to 25% • 33%
More than 25% • 33%
0-10% • 25%
11-25% • 25%
26-50% • 25%
51%+ • 25%
North America • 25%
Europe • 25%
Asia • 25%
Other • 25%
Market share increases • 25%
Market share remains the same • 25%
Market share decreases • 25%
Foresight exits cancer-detection market • 25%
Mayo Clinic • 25%
Cleveland Clinic • 25%
Johns Hopkins Hospital • 25%
Other • 25%
Less than 10% • 25%
10% to 20% • 25%
20% to 30% • 25%
More than 30% • 25%
Less than 10% • 25%
10% - 20% • 25%
20% - 30% • 25%
More than 30% • 25%
North America • 25%
Europe • 25%
Asia • 25%
Other • 25%
0-5% • 25%
5-10% • 25%
10-15% • 25%
15% or more • 25%
Google Health • 25%
IBM Watson Health • 25%
GE Healthcare • 25%
Philips Healthcare • 25%
Less than 10 • 25%
10 to 20 • 25%
21 to 50 • 25%
More than 50 • 25%
Increase by more than 5% • 25%
Increase by 2-5% • 25%
Increase by less than 2% • 25%
No significant increase • 25%
No • 50%
Yes • 50%
No • 50%
Yes • 50%
21-30 papers • 25%
31+ papers • 25%
0-10 papers • 25%
11-20 papers • 25%