By Janet Fang
Posting in Cities
IBM's supercomputer defeated Jeopardy's finest. Easy. Now, onto healthcare informatics. There're a lot of data out there, and Watson's Jeopardy-winning skills could help.
IBM’s supercomputer Watson conquered its human competition, scoring $77,147 during the 3-day Jeopardy! challenge that ended Wednesday.
(It even won the $1 million grand prize after an incorrect ‘U.S. Cities’ final Jeopardy response of “What is Toronto?????”)
Now what? Well, there’re a lot of data out there, and Watson’s Jeopardy-winning skillz could be perfectly tasked.
During round 2, Watson project leader David Ferrucci talked a little about future applications for health. Indeed, in the long list of IBM research projects, Advanced Healthcare Informatics is notably bulleted: “Work on extracting important HealthCare entities (e.g., drugs, diseases, procedures) from Clinical documents (e.g., textual progress notes).”
That is, information sifting and retrieval. Some ways to apply Watson’s tools in medicine:
- If a doctor doesn’t know a symptom or disease, Watson could help diagnose patients by scouring libraries-worth of medical literature that doctors don’t have time to keep up with. (Not that Watson would be the final diagnostician though.)
- Keep track of mounds of patient records
- Detect anomalies in medical images missed by the human eye
- In cardiology, Watson would analyze data to alert doctors to problems like too much digitalis in blood tests or the overuse of diuretics.
And Watson, adds IBM’s Katharine Frase, continues to learn and refine its abilities.
“There's massive production of knowledge, particularly in the biological community, and researchers can't keep up with it," says Oren Etzioni of University of Washington’s Turing Center for investigating language processing, data mining, and the web.
More concretely, IBM has reached a research agreement with Nuance Communications Inc., a provider of speech-recognition technology, to "explore, develop and commercialize" Watson’s capabilities in the healthcare industry. WSJ reports:
IBM researchers are working to apply the system to business uses, such as helping physicians and nurses find answers within huge volumes of information. A doctor considering a patient's diagnosis could use Watson's analytics technology along with Nuance's voice and clinical language understanding offerings to rapidly consider all the related texts, reference materials, prior cases and latest knowledge in medical journals to gain information from more potential sources then previously possible, making the physician more confident in the patient's diagnosis…
"We can transform the way that healthcare professionals accomplish everyday tasks by enabling them to work smarter and more efficiently," says IBM’s John E. Kelly III.
So far, two hospitals have signed up. Columbia University Medical Center and University of Maryland School of Medicine will be the sites of Watson’s first real-world tests outside of IBM labs and Jeopardy.
IBM and Nuance expect their first commercial offerings to be available in 18 to 24 months.
Feb 16, 2011
Also- Here's an article explaining Watson's reasoning behind the answer: http://asmarterplanet.com/blog/2011/02/watson-on-jeopardy-day- two-the-confusion-over-an-airport-clue.html
There are EIGHT Torontos in the United States. Actually, based on the data, there are more Torontos in the United States than in Canada! http://en.wikipedia.org/wiki/Toronto_(disambiguation) Looks like Watson isn't so dumb after all.
What good will Watson do for those Millions of hard working Americans without health coverage? The USA is the only Western NAtion that does not cover ALL the citizens. Private comapanies are in businees for profit not people. Only in the USA, People have started turning to natural healing due to lack of coverage Thousands of diabetics lose their feet due to lack of coverage http://www.ourmidland.com/voices/health/article_48bf350a-3b64-11e0-b019-001871e39ea8.html Watson is all well and good but until we go back to caring for PEOPLE all the Robots in the world will not make a difference
I would love to have Watson at my doctor's office listening to all my symptoms and suggesting possible tests and even diagnosis that matched my problems. Medicine is getting so complicated that it is really asking a lot for doctors to keep up and also remember all the problems associated with a list of symptoms to narrow down a diagnosis. I think this is an excellent place for Watson to step in and at least narrow down the information so the doctor can make an informed decision without leaving out any of the possibilities. Bases on what symptoms I describe Watson could also tell the doctor what questions to ask to narrow down what is wrong with me. Once the doctor decides what my problem is, Watson could suggest the best therapy based on the latest medical research.
Not true. 1 DISEASE Watson will NEVER AREST/CURE is called sue you ites. THE bigger POTENAL PAYOFF THE MORE LIDAGATION. Promoed by lawers,DR,INSERANCE &underwriters & /or any GREEDY ROYheM. EVERY WANTS EASY peace MILONS$$$ PIE. THIS is major screwup or helth care now with no end possable. Me/you miss prepay/copay get no service or 1/2 way/bandaid. not treatment. ? Watson & ALL HOW much IS my carcus/body worth?? in $& C's ??
Might be a good idea as long as they remember GIGO. The efficacy of such a plan will depend on the database that is constructed. There have been as many contradictory papers on diagnosis and treatment as there have been on nutrition (remember eggs are bad then eggs are good, coffee is bad and coffee is good..same for choclate). How about fibromyalgia exists and fibromyalgia does not exist. In analyzing recent data will Watson ignore past data..does it have the capability to weigh one against the other? And does a study from 'across the pond' have meaning in the US?I congratulate IBM on forward thinking development but we must proceed with caution because GIGO exists. Just a thought
I just want to know why Watson picked Toronto as an answer to a question in a category on U.S. Cities
The rules of the Jeopardy competition were the rules. And disembodied human brains would have had a pretty tough time competing. It's not about "conquering" people, it's about bringing to bear the ability of supercomputers to analyze large quantities of info that would take large numbers of people a long time to do. The Jeopardy competition demonstrated that supercomputers can be used to analyze info beyond the traditional numerical info. With modern telecom, they can be used anywhere that has access to satellite communications.
"Watson conquered its human competition" hmmm... in order to agree with that, I would need to see it done by a computer the size of a human brain, not one that takes up an entire room. That computer they showed on TV was just an avatar for the real computer backstage.
At this point, doctors retain volumes of knowledge to successfully complete their medical training. Remember all the things you remembered until you realized you could just find them at anytime through the internet? As a doctor retains less knowledge by relying on Watson, will his diagnostic ability erode to the point that the computer and a technician replace the doctor? Will this team have enough confidence to override the computer's diagnosis and treatment plan, if they are even capable of recognizing a discrepancy? All I am saying is that, what starts as a tool becomes a crutch without which one can not walk. We do not have primary care "providers" for millions of people. Of the diminishing few smart and empathetic young people that pursue medicine, a minority will practice in primary care. I worry that, IF you have insurance, this is what you will be offered as a cost effective means to direct your care as the number of doctors dwindles over the next generation. Doctors are people that are already good at asking a lot from themselves and you will miss them when they are gone. But this new era of heath care will probably save the country a lot in health care costs. Why would an insurance company pay a doctor when the computer is so efficient? Just a thought.