Medical Discovery News
Science permeates everyday life. Yet the understanding of advances in biomedical science is limited at best. Few people make the connection that biomedical science is medicine and that biomedical scientists are working today for the medicine of tomorrow. Our weekly five-hundred-word newspaper column (http://www.illuminascicom.com/) and two-minute radio show provide insights into a broad range of biomedical science topics. Medical Discovery News is dedicated to explaining discoveries in biomedical research and their promise for the future of medicine. Each release is designed to stimulate listeners to think, question and appreciate how science affects their health as well as that of the rest of the world. We also delve into significant biomedical discoveries and portray how science (or the lack of it) has impacted health throughout history.
Medical Discovery News
Do you really Want to Know
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
1008 Do you really Want to Know
Welcome to Medical Discovery News. I’m Dr Norbert Herzog.
And I’m Dr. David Niesel
Would you want to know as much as twenty years ahead whether you’d develop something like pancreatic cancer? We may soon be confronted with this question.
Researchers have built an AI that predicts a person’s risks for over one thousand diseases up to two decades before symptoms appear. It could help people avoid or ease the effect of certain illnesses. But that kind of information can also be stressful.
The AI we’re talking about is Delphi-twoM, a Large Language Model AI like ChatGPT. In this case, it models health outcomes using past medical events, sex, body mass, lifestyle factors such as smoking and alcohol consumption and how they interact to predict future health issues.
The AI was trained using four hundred thousand UK Biobank participants and then tested using two million health records from Denmark.
Delphi-2M reads the information and predicts the illnesses and when they’ll happen if the most likely sets of events occur. It can generate a complete health trajectory, but the accuracy drops to sixty percent at about twenty years out.
It’s also not as good at predicting rare diseases and conditions that result from being in certain environments. Plus, it’s trained on the UK Biobank which is biased toward mostly white, educated and relatively healthy people.
Once the model is improved with more diverse data, one day, doctors could use it along with patients’ data to make longer-term health plans and even help them live longer lives.
We are Drs. David Niesel and Norbert Herzog, at UTMB and Quinnipiac University, where biomedical discoveries shape the future of medicine. For much more and our disclaimer go to medicaldiscoverynews.com or subscribe to our podcast.
