VinBrain harnesses Stanford data to improve radiology AI
It seeks to use AI to improve the accuracy of an information-extracting method for structuring radiology reports.
Vietnamese healthcare AI developer VinBrain has partnered with Stanford University to enhance AI medical imaging.
Based on their latest data use agreement, Stanford will initially provide VinBrain access to about 240,000 anonymized medical images and reports to train and test the AI on its DrAid clinical decision support platform while also helping improve the accuracy of an information extracting method for structuring radiology reports called RadGraph.
WHY IT MATTERS
According to VinBrain, its latest partnership with Stanford aims to enhance efficiency in radiology reports while “tackling complexity, ambiguity, and limitations in medical image interpretation.”
Through Stanford’s support, the company will be able to advance the research in AI healthcare while also building a huge data source in Vietnam that underpins its DrAid platform, which has so far been used to analyze over 2.3 million x-ray images. “To design a state-of-the-art AI-enabled platform for more accurate diagnosis and treatments, a tremendous amount of geographically distributed data, medical images, and patient medical records are required for system training,” VinBrain stressed.
THE LARGER TREND
In late January, VinBrain announced its partnership with Microsoft – a first in Vietnam – focusing on three areas of healthcare AI: data sharing, cross-product validation, and R&D. As part of their collaboration, Microsoft Azure will be used to facilitate the sharing of millions of data on DrAid while ensuring its privacy and security, data governance, and regulatory compliance. They will also use Azure Cognitive Services for Computer Vision to develop VinBrain’s AI capability for processing and returning images and building image classifiers.