“Building Artificial Intelligence from Human Intelligence: A Case Study in Breast Cancer Detection” Speaker: Assistant Professor Feng Mengling ‘Mornin’, SSHSPH & Senior Assistant Director, NUHS AIO

The Digital Mammography DREAM Challenge was nothing short of a data scientist’s dream come true. Organised by Sage Bionetworks and DREAM, the competition offered 640,000 de-identified mammograms and almost unlimited computational resources sponsored by IBM and Amazon to use Artificial Intelligence (AI) to improve breast cancer detection. With so much data at our disposal, we thought we could crack the problem easily, and got straight to work. The network we designed was so complicated, it took us nearly two weeks to train it. But to our great disappointment, our AI agent turned out to be barely better than a coin flip. So what went wrong?

“At the Edge of Discovery” Speaker: Assistant Professor Ngiam Kee Yuan, Group Chief Technology Officer, Consultant, Division of Thyroid and Endocrine Surgery, Department of Surgery, NUH

AI is becoming a ubiquitous technology that allows patients to access services or things that were not previously possible. To enable such technologies to become mainstream, NUHS, in partnership with NUS, developed the DISCOVERY AI platform, a production level secure sandbox for building AI tools from huge multi-domain medical databases. It is the first platform that is able to anonymously link multi-institution datasets yet share data securely and equitably between clinicians, researchers and data scientists. The security of the platform is guaranteed by a proprietary blockchain technology coupled iwth the highest enterprise-grade security. Access to this platform is unified through a cluster-wide governance policy. Multiple AI tools have been built on this platform and is integrated with existing and future EMR systems to alert clinicians directly. The speaker will be detailing the features and AI capabilities that would support healthcare practitioners in their daily work.

Registration & breakfast at 7.10am, talk begins at 7.40am.