IBM’s Watson Supercomputer Becoming The Best Doctor?

Coincidentally we had a few IBM researchers, including elements from the Watson team, visiting our department (Integrated Mathematical Oncology) today. Good stuff but to be a doctor you need more than data, you also need a biological understanding. Which is why I think they came to visit and why I think a collaboration could work.

ScienceRoll

An interesting article was published on Business Insider. I’m not saying it’s technically impossible for an algorithm to become better at making diagnoses than a human, but it certainly should not be the ultimate goal in medicine. This is why I’m writing now my new book, The Guide to the Future of Medicine, to underscore this notion with stories and practical examples.

A quote from the article:

“Watson, the supercomputer that is now the world Jeopardy champion, basically went to med school after it won Jeopardy,” MIT’s Andrew McAfee, coauthor of The Second Machine Age, said recently in an interview with Smart Planet. “I’m convinced that if it’s not already the world’s best diagnostician, it will be soon.”

watson

Read similar news on Medicalfuturist.com!

View original post

Credit where is due?

Credit where is due? Detailing the roles of each author in a paper is better than merely assuming that only the first and last person in an author list matter. This is particularly important in interdisciplinary research where every project will involve a number of people performing the actual work (AKA postdocs) and a number […]

Modelling aneuploidy in colorectal cancer

Image

Last week we were fortunate to have Albert Rubben, an oncologist from RTWH Aachen in Germany. Dr Rubben is mainly a clinician but is knowledgeable about the biology of cancer and very keen on the use of mathematical and computational models to understand how tumours evolve. He gave a great talk about tumour heterogeneity in early and late stages of melanoma (his specialty as an oncologist) and you can have a sense of it if you check my twitter posts for Thursday last week.

But the picture at the top of this post is what shows what he really came to Tampa for. When he found that CancerEvo’s Arturo Araujo was the leading author behind a computational model of aneuploidy he inmediately contacted him to see if they both could work on a new one that could study that in the context of a specific cancer: colo-rectal cancer. I am certainly looking forward to see what they find!

Medical hackathons

  Medical hackathons An interesting piece in Slate. Traditionally hackathons have been one way to drive innovation in the technology and internet sectors. Turns out that they also have a place in the medical sector as long as technology has something to do with it. Slate’s article reminds me of the workshops that Sandy Anderson […]