VinBrain uses data from Stanford University to improve artificial intelligence in radiology

Vietnamese AI healthcare developer VinBrain has launched a new partnership with Stanford University to improve medical imaging through AI.

Under the latest data agreement, Stanford will initially provide VinBrain with access to approximately 240,000 anonymous medical images and reports for AI training and testing on its DrAid clinical decision support platform, and will help improve the accuracy of the information extraction method for structuring radiological reports. called RadGraph.

WHY IS IT IMPORTANT

According to VinBrain, its latest partnership with Stanford aims to improve the efficiency of radiological reports while “addressing the complexity, ambiguity and limitations in the interpretation of medical images.”

With Stanford’s support, the company will be able to advance AI-assisted healthcare research as well as build a massive data source in Vietnam that underpins its DrAid platform, which has been used to analyze more than 2.3 million x-rays to date. “In order to develop a state-of-the-art AI-enabled platform for more accurate diagnosis and treatment, a huge amount of geographically dispersed data, medical images, and patient records are required to train the system,” VinBrain emphasized.

BIG TREND

At the end of January, VinBrain announced its a partnership with Microsoft — a first in Vietnam — focusing on three areas of healthcare AI: data sharing, multi-product validation, and research and development. As part of their collaboration, Microsoft Azure will be used to facilitate the exchange of millions of data on DrAid while maintaining its privacy and security, data governance and compliance. They will also use Azure Cognitive Services for Computer Vision to develop VinBrain’s AI capabilities for processing and returning images and building image classifiers.