Researchers from Optima Partners, Biogen, and the University of Edinburgh have made a breakthrough in predicting disease risk using machine learning and analyzing medical data. By identifying protein patterns linked to diseases, they could accurately predict disease risk up to 10 years before diagnosis. The study also showed the potential for comparing patient blood test results with new protein patterns to detect disease possibilities earlier.
Microsoft’s Copilot AI in OneNote can now read and analyze handwritten notes, converting them into text for easy editing and sharing. The feature is currently beta testing and will be available to existing Copilot subscribers once it rolls out more broadly. Users can also ask Copilot to summarize notes, generate to-do lists, and more based on the handwritten content.
Microsoft has developed an AI speech generator called VALL-E 2 that can accurately reproduce human speech. However, due to concerns about potential misuse, Microsoft has decided not to release it to the public. The AI engine achieves human parity in speech quality by using features like Repetition Aware Sampling and Grouped Code Modeling. While VALL-E 2 has potential applications in various fields, Microsoft emphasizes the need for protocols to ensure speaker approval and detect synthesized speech.
Why do we care?
The disease investigation discovery is exactly the kind of AI use case I’m most interested in – using data to improve business outcomes. The benefits are obvious. This discovery has significant implications for early intervention and prevention in healthcare, which is improved patient outcomes.
We’re looking for ways to leverage the technology in this way.