Vietnam-based VinBrain, which specialises in combining IoT and AI in healthcare applications, recently signed a data use agreement (DUA) with the Stanford Center for Artificial Intelligence for Medical and Imaging (AIMI) as part of a collaboration to create an AI platform that would produce a more accurate medical diagnosis and treatments.
Developing the AI platform will entail a tremendous amount of geographically distributed data, medical images, and patient medical records. The DUA enables Stanford University and VinBrain to improve efficiency in radiology reports and advance AI in healthcare by sharing relevant data for model training and testing.
Stanford AIMI will provide 240,000 anonymised medical images and full-text reports for the joint effort to improve the accuracy of the RadGraph method for interpreting medical images by taking into account their complexity, ambiguity and limitations. On its side, VinBrain will contribute to these data sets with more than 2.3 million medical images that it had analysed to date.
"VinBrain always strictly adheres to data security regulations according to US standards,” said company CEO Steven Truong, pointing out Stanford sharing its data shows absolute trust in VinBrain's commitment to data privacy and security.
VinBrain has been working with the US university for more than three years on many projects that have achieved desirable clinical outcomes.
“We appreciate VinBrain's leadership in research and the capabilities that they have. The speed of work is incredible, and I also appreciate the company’s collaborative spirit,” said Professor Curtis Langlotz, director at Standford AIMI.
“We have done a lot, but there is still a lot more to do. I look forward to our continued collaboration.”
Part of Vingroup, VinBrain has developed AI-based DrAid, a comprehensive doctor assistant platform that helps doctors diagnose heart, lung, liver, and bone diseases based on digital medical images. The platform consists of two parts:
- DrAid Radiology V1: An AI-enabled triage and notification software designed to aid the clinical assessment of adult Chest X-ray cases with features suggestive of pneumothorax. This software is FDA 510(K) cleared and ready for the US market.
- DrAid Appliance: a system classify an adult Chest X-ray between normal and nearly 60 abnormal findings with higher-than-average quality standards of primary care physicians