think thank thunk

After its Jeopardy! fame fades, Watson is going to get down to serious work. The IBM team led by computer scientist  Dave Ferrucci is already deploying Watson in health care. The same way IBM fed Watson Wikipedia, the Bible, a geospatial database–the equivalent of a million pages of documents–it has begun to feed Watson electronic medical records, doctors’ notes, patient histories, symptoms, the USP Pharmacopeia. Here’s the amazing thing: The machine is getting faster at learning. Teaching it to play Jeopardy at a championship level took four years. Teaching it to deliver reasonably accurate answers to diagnostic questions took only four months. I can see IBM selling Watson as a Web-delivered service to doctors and hospitals seeking answers to a patient presenting with problems. Watson considers everything and creates evidence profiles (the types of information it relies on, weighted based on their reliability and utility) that feed into diagnoses graded on varying levels of confidence.  These can be offered up as charts on an iPad  showing a doctor Watson’s thought process. It’s like peering into the mind of a House, M.D. The doctors make the final call but they can assess possibilites they may not have seen and can click right to source material used to compile Watson’s answers. This is powerful stuff.

The first study to link a childhood vaccine to autism was based on doctored information about the children involved, according to a new report on the widely discredited research.

The conclusions of the 1998 paper by Andrew Wakefield and colleagues were renounced by 10 of its 13 authors and later retracted by the medical journal Lancet, where it was published. Still, the suggestion the MMR shot was connected to autism spooked parents worldwide and immunization rates for measles, mumps and rubella have never fully recovered.

A new examination found, by comparing the reported diagnoses in the paper to hospital records, that Wakefield and colleagues altered facts about patients in their study.

The analysis, by British journalist Brian Deer, found that despite the claim in Wakefield's paper that the 12 children studied were normal until they had the MMR shot, five had previously documented developmental problems. Deer also found that all the cases were somehow misrepresented when he compared data from medical records and the children's parents.

Bonus: For a two line lesson on how "mainstream media" reacts when they realize they sensationalized the wrong side of a story, just look at MSN's headline versus subhead.

Long-term use of a daily low-dose aspirin dramatically cuts the risk of dying from a wide array of cancers, a new investigation reveals.

Specifically, a British research team unearthed evidence that a low-dose aspirin (75 milligrams) taken daily for at least five years brings about a 10 percent to 60 percent drop in fatalities depending on the type of cancer.

The finding stems from a fresh analysis of eight studies involving more than 25,500 patients, which had originally been conducted to examine the protective potential of a low-dose aspirin regimen on cardiovascular disease.

Much of what medical researchers conclude in their studies is misleading, exaggerated, or flat-out wrong. So why are doctors—to a striking extent—still drawing upon misinformation in their everyday practice?

The era of antibiotics is coming to a close. In just a couple of generations, what once appeared to be miracle medicines have been beaten into ineffectiveness by the bacteria they were designed to knock out. Once, scientists hailed the end of infectious diseases. Now, the post-antibiotic apocalypse is within sight.

This isn't a failure of medical research; it's a failure of statistics, and one that is becoming more common in fields ranging from genomics to astronomy. The problem is that our statistical tools for evaluating the probability of error haven't kept pace with our own successes, in the form of our ability to obtain massive data sets and perform multiple tests on them. Even given a low tolerance for error, the sheer number of tests performed ensures that some of them will produce erroneous results at random.

Don't leap to health conclusions each time a new study gets published.