Stanford University, VinBrain partner to improve diagnostic imaging efficiency

Stanford University and Vietnamese AI healthcare products firm VinBrain have signed a deal to design a state-of-the-art AI-enabled platform to advance medical diagnosis and treatments.

Under the data use agreement, or DUA, the American renowned university and the AI unit of Vietnamese conglomerate Vingroup will together improve the precision of radiology interpretation using a multi-modal “RadGraph” method, VinBrain announced Wednesday.

To design the artificial intelligence-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.

The DUA grants Stanford University and VinBrain collaborative power by sharing relevant de-identified data for model training and testing, while tackling complexity, ambiguity, and limitations in medical image interpretation, according to the announcement.

Curtis Langlotz, professor of radiology and biomedical informatics, Stanford University; director, Center for Artificial Intelligence in Medicine & Imaging, and VinBrain CEO Truong Quoc Hung signed the DUA.

“We appreciate VinBrain’s leadership in research and the capabilities that they have. The speed of work is incredible, and I also appreciate the collaborative spirit of VinBrain,” Prof. Langlotz said.

“I know that we’ve done a lot, but there’s still a lot more to do. I look forward to continued collaborations. I’m glad we sat down together to sign this agreement,” he added.

VinBrain’s Hung said, “The agreement signed with Stanford University is not an overnight result but a long journey of more than three years of collaboration. This collaboration has produced many research projects together that have achieved desirable clinical outcomes.”

RadGraph is a novel method introduced by Prof. Pranav Rajpurkar at Harvard University and Prof. Langlotz at Stanford, according to the statement.

Starting with 240,000 anonymized medical images and report pairs provided by Stanford through this DUA, VinBrain will work to improve the accuracy of the RadGraph method through a research collaboration with Stanford to extract clinical entities and relation annotations from a large dataset of full-text radiology reports.

The agreement promises to comprehensively expand cooperation in the next areas of development, strengthen the relationship between VinBrain and Stanford, and take VinBrain further in the research of AI-developed solutions for patients.

In late January, American tech giant Microsoft and VinBrain embarked on a healhtech partnership that will see the duo work on data sharing, product validation, and R&D. The partnership is Microsoft’s first healthtech collaboration in Vietnam.