A couple of weeks ago I wrote a blog post about how IBM's Watson might be used in medicine. Seems that's now happening, according to this Time article. Getting Watson through medical school so he understands medical records, jargon, shorthand, and journal articles will tak a while, but it seems very do-able.
No word yet about Watson's use in radiology, though--a place where I believe substantial savings could come quickly.
Showing posts with label medicine. Show all posts
Showing posts with label medicine. Show all posts
Wednesday, March 2, 2011
Friday, February 25, 2011
The Mystery of Dr. Mel Levine
Years ago I went to a talk at the Harvard Ed School given by Dr. Mel Levine, the noted pediatrician and advocate for children with learning differences. I remember him as a dynamic speaker, convincing and provocative in his description of the way smart, hard-working kids with minor brain-wiring differences could suffer in school. Some kids need to fidget, he said: fidgeting is what allows them to concentrate. Writing, he said, was one of the most difficult tasks we could ask a kid to undertake. To pull together at one time the big picture and the small picture needed for organization, the auditory and visual systems needed for spelling, the fine motor skills and the timing needed for pushing the pencil - that was a huge task.
A few years later, when one of our sons struggled to produce anything in writing, I thought back to that lecture and was grateful to Dr. Levine. He helped me see that our son, who looked as if he was doing nothing, was really making a heroic effort. Dr. Levine has probably helped a lot of children, I thought.
Last week, Dr. Mel Levine, now 71, committed suicide by shotgun to the head. He had been a Rhodes scholar, lauded author, and at one time chief of ambulatory pediatrics at Children's Hospital in Boston. But in recent years complaints had surfaced that when the door to the examining room closed, he improperly examined and fondled the genitals of young boys. In 2008, a case involving seven boys was settled in Massachusetts. In 2009, Levine voluntarily gave up his medical license in North Carolina. Then last week came the news that another lawsuit had been filed in Massachusetts accusing Dr. Levine of sexually abusing thousands of patients.
Thousands. It's a stunning accusation. When I heard earlier reports, I had hoped they weren't true. After all, doctors examine kids. They check for testicular masses. Some child might feel uncomfortable and misinterpret that.
But nobody needs to re-examine for testicular masses at every follow-up visit for a learning problem. And although these accusations never got their chance to air in court, they certainly carry the air of authenticity - the pain, shame, and rage. How could such abuse have gone on so long?
Like a priest, a pediatrician for troubled kids holds a position of trust, power, and influence. For a doctor, though, boundaries of nudity and touch lie in a far different place than they do with a priest; it may be hard for a child to be certain what is within bounds and what is not. Should the child complain? Even parents may have difficulty knowing whether what their kids describe is really not okay. I don't know what parents said over the years to Children's Hospital, but a lot of information on this topic may come out in the lawsuit, which is slated to proceed.
Still, the mystery that disturbs me more is the mystery of Dr. Levine's own mind and dark heart. How did he justify what he was doing to the very kids he was advocating for, kids he probably did care about and identify with in some way? I'm always frustrated that even when a crime is proved, there is a great, impenetrable silence at the core of it. How often does a criminal try to explain what he did and why, how he came to this place? I can imagine the shame and fear that led to Mel Levine's suicide. If only he had found the courage to unburden his heart. Few would have forgiven him, but if only we could start to understand how such deviance happens, we might get better at preventing the terrible harm it does.
A few years later, when one of our sons struggled to produce anything in writing, I thought back to that lecture and was grateful to Dr. Levine. He helped me see that our son, who looked as if he was doing nothing, was really making a heroic effort. Dr. Levine has probably helped a lot of children, I thought.
Last week, Dr. Mel Levine, now 71, committed suicide by shotgun to the head. He had been a Rhodes scholar, lauded author, and at one time chief of ambulatory pediatrics at Children's Hospital in Boston. But in recent years complaints had surfaced that when the door to the examining room closed, he improperly examined and fondled the genitals of young boys. In 2008, a case involving seven boys was settled in Massachusetts. In 2009, Levine voluntarily gave up his medical license in North Carolina. Then last week came the news that another lawsuit had been filed in Massachusetts accusing Dr. Levine of sexually abusing thousands of patients.
Thousands. It's a stunning accusation. When I heard earlier reports, I had hoped they weren't true. After all, doctors examine kids. They check for testicular masses. Some child might feel uncomfortable and misinterpret that.
But nobody needs to re-examine for testicular masses at every follow-up visit for a learning problem. And although these accusations never got their chance to air in court, they certainly carry the air of authenticity - the pain, shame, and rage. How could such abuse have gone on so long?
Like a priest, a pediatrician for troubled kids holds a position of trust, power, and influence. For a doctor, though, boundaries of nudity and touch lie in a far different place than they do with a priest; it may be hard for a child to be certain what is within bounds and what is not. Should the child complain? Even parents may have difficulty knowing whether what their kids describe is really not okay. I don't know what parents said over the years to Children's Hospital, but a lot of information on this topic may come out in the lawsuit, which is slated to proceed.
Still, the mystery that disturbs me more is the mystery of Dr. Levine's own mind and dark heart. How did he justify what he was doing to the very kids he was advocating for, kids he probably did care about and identify with in some way? I'm always frustrated that even when a crime is proved, there is a great, impenetrable silence at the core of it. How often does a criminal try to explain what he did and why, how he came to this place? I can imagine the shame and fear that led to Mel Levine's suicide. If only he had found the courage to unburden his heart. Few would have forgiven him, but if only we could start to understand how such deviance happens, we might get better at preventing the terrible harm it does.
Saturday, February 19, 2011
Could IBM's Watson transform medicine?
Last week, we saw a machine trounce the two best humans who have ever played the game of Jeopardy, a wide-ranging quiz game filled with puns and wordplay. The computer, named Watson by the IBM engineers who spent four years and tens of millions of dollars developing it, tapped into its huge database of facts and rules to come up with rank-ordered possibilities for potential answers to each question. It also assigned probabilities to these possible answers, and if the top probability was above its risk threshold, it buzzed in with lightning speed.
According to a NOVA show about Watson's development, the most important innovation in Watson's development was "his" ability to use machine learning. It would take forever to enter all the rules and relationships Watson needed to have a shot at winning Jeopardy. Instead, the engineers fed him huge amounts of data and let him learn from it by giving him tens of thousands of examples--Jeopardy questions with correct answers. Watson learned from his mistakes. Presumably he learned which kinds of evidence (mostly word associations) from his huge memory banks were more important in which circumstances. Very smart.
It's these two features--the ability to learn from thousands of examples and the ability to assign probabilities to different possible decisions--that make Watson's intelligence so relevant to other fields. Consider medicine. The process of diagnosis is, at its heart, gathering a lot of data, considering it, and then ranking possible causes of the patient's symptoms according to an intuitive sense of probabilities.
Doctors know the potential sources of error in this process. One, of course, would be gathering the wrong data--an inaccurate history or physical exam, lab error or mixed lab slips. I'm not sure Watson could help with physical exam, but he would be better than humans at taking into account the likelihood of lab error. The next source of error is the human tendency, nearly impossible to train away, of weighing anecdotal experience, especially recent experience, too highly. The diagnosis we've most recently encountered is the one that jumps to mind. Which brings us to a third source of error--not thinking of a certain diagnosis at all.
Now imagine Watson, fed all the epidemiological data we can gather, all the journal articles published in the last sixty years, all the case studies and Grand Rounds. He can perform statistical analysis and make correlations. Within microseconds of being turned loose on the data of a certain case, he could spit out a comprehensive list of potential diagnoses with probabilities assigned.
Of course, he could still make silly errors, as when he seemed to believe Dorothy Parker was a book or Toronto an American city, but humans can weed out such answers. Doctors would have their memories jogged about possibilities they'd neglected, and the probability rankings would help them decide which tests to pursue next. Doctors would still take the history, interact with the patient, advise the family, doing the things humans are best at.
Radiologists, however, might find themselves out of a job. Radiology is essentially a matter of pattern detection. A radiologist looking at a chest X-ray may have one line of information about the patient: "60-yr. old smoker w. cough x 3 wks." After that, there's only the X-ray, which the radiologist compares in memory with all the other chest X-rays she's read. She compares heart width to chest width, looks for flattening of the diaphragm, and searches systematically for anything that breaks the expected pattern, such as unusual spots or shadows.
Watson could do all that. Feed him ten thousand chest X-ray images and ten thousand diagnoses, and he would be better at searching his memory than any human. He would never get bored by the repetitiveness of his task. He would be diligent and systematic. He would continue to learn, adding CT scans, MRIs, bone scans, thallium heart scans, and whatever else we ask to his repertoire. He would work day and night without vacation, and he would spit out his answers in microseconds. And once the capital cost was covered, taking one more highly-paid human out of the chain of medical diagnosis and treatment would save money and help to keep health care costs down.
According to a NOVA show about Watson's development, the most important innovation in Watson's development was "his" ability to use machine learning. It would take forever to enter all the rules and relationships Watson needed to have a shot at winning Jeopardy. Instead, the engineers fed him huge amounts of data and let him learn from it by giving him tens of thousands of examples--Jeopardy questions with correct answers. Watson learned from his mistakes. Presumably he learned which kinds of evidence (mostly word associations) from his huge memory banks were more important in which circumstances. Very smart.
It's these two features--the ability to learn from thousands of examples and the ability to assign probabilities to different possible decisions--that make Watson's intelligence so relevant to other fields. Consider medicine. The process of diagnosis is, at its heart, gathering a lot of data, considering it, and then ranking possible causes of the patient's symptoms according to an intuitive sense of probabilities.
Doctors know the potential sources of error in this process. One, of course, would be gathering the wrong data--an inaccurate history or physical exam, lab error or mixed lab slips. I'm not sure Watson could help with physical exam, but he would be better than humans at taking into account the likelihood of lab error. The next source of error is the human tendency, nearly impossible to train away, of weighing anecdotal experience, especially recent experience, too highly. The diagnosis we've most recently encountered is the one that jumps to mind. Which brings us to a third source of error--not thinking of a certain diagnosis at all.
Now imagine Watson, fed all the epidemiological data we can gather, all the journal articles published in the last sixty years, all the case studies and Grand Rounds. He can perform statistical analysis and make correlations. Within microseconds of being turned loose on the data of a certain case, he could spit out a comprehensive list of potential diagnoses with probabilities assigned.
Of course, he could still make silly errors, as when he seemed to believe Dorothy Parker was a book or Toronto an American city, but humans can weed out such answers. Doctors would have their memories jogged about possibilities they'd neglected, and the probability rankings would help them decide which tests to pursue next. Doctors would still take the history, interact with the patient, advise the family, doing the things humans are best at.
Radiologists, however, might find themselves out of a job. Radiology is essentially a matter of pattern detection. A radiologist looking at a chest X-ray may have one line of information about the patient: "60-yr. old smoker w. cough x 3 wks." After that, there's only the X-ray, which the radiologist compares in memory with all the other chest X-rays she's read. She compares heart width to chest width, looks for flattening of the diaphragm, and searches systematically for anything that breaks the expected pattern, such as unusual spots or shadows.
Watson could do all that. Feed him ten thousand chest X-ray images and ten thousand diagnoses, and he would be better at searching his memory than any human. He would never get bored by the repetitiveness of his task. He would be diligent and systematic. He would continue to learn, adding CT scans, MRIs, bone scans, thallium heart scans, and whatever else we ask to his repertoire. He would work day and night without vacation, and he would spit out his answers in microseconds. And once the capital cost was covered, taking one more highly-paid human out of the chain of medical diagnosis and treatment would save money and help to keep health care costs down.
Thursday, January 6, 2011
The "Vaccines Cause Autism" Scam
Don't vaccines cause autism? Wasn't there mercury in the MMR vaccine that caused neurological symptoms? Well, no.
For years, medical researchers have published careful studies failing to support a purported link between childhood vaccinations - particularly the MMR, or measles/mumps/rubella vaccine - and autism. However, many anxious parents and autism advocates have remained wary of vaccines, mostly based on the work of British doctor Andrew Wakefield. Even after the medical journal Lancet retracted Wakefield's original 1998 article on the subject as fatally flawed - even after Wakefield and a co-author were "stuck off the medical register" (i.e. had their licenses revoked), true believers maintained that Wakefield was a martyr to the truth, hounded by greedy vaccine makers intent on filling their pockets.
Yesterday the British Medical Journal published an article by Brian Deer, a journalist who has gone back and carefully examined the medical records and interviewed the parents of the original 12 patients discussed in the Wakefield paper. BMJ's conclusion: the retracted Lancet paper did not reflect merely carelessness or scientific misconduct. It was barefaced fraud for financial gain.
First, the study. Wakefield's team examined 12 children with neurological and gastrointestinal complaints. Most of the parents suspected the MMR vaccine as the cause of their child's symptoms. Wakefield "found" colitis in most of the children, and also "found" that on average, onset of symptoms occurred 6.3 days after the vaccine.
What Wakefield did not disclose in the article was that for two years before and during the research he was being lavishly paid by a tort lawyer hoping to build a case against vaccine manufacturers. Wakefield's take was $668,000 plus expenses. In a confidential grant application to the UK government's Legal Aid board, Wakefield and his lawyer sponsor voiced their conviction that a vaccine-autism link existed. Their grant application stated, "The objective is to seek evidence which will be acceptable in a court of law of the causative connection between either the mumps, measles and rubella vaccine or the measles/rubella vaccine and certain conditions which have been reported with considerable frequency by families of children who are seeking compensation." (The italics are mine.) In other words, give us the money and we'll give you the evidence to make you rich.
Wakefield knew what he had to show, and he proceeded to manipulate his data to show it. Almost nothing about the data presented in the Lancet paper was fairly represented.
- Patients were recruited through an anti-vaccine group called JABS. Parents were predisposed to blame their children's symptoms on the MMR - in fact, the lawyer had sent out a letter to his clients telling them this doctor believed there was a link. Parents were then interviewed and gently prodded to remember symptoms starting shortly after the vaccine. In one case, a parent remembered that a child was "pale" one week after the vaccine. Prompted further by the interviewer, the parent admitted that the child may have had fever or delirium. In the paper, the fever and delirium are reported as fact.
- The paper reported that 9 of 12 patients developed normally and then regressed, a form of autism called "regressive autism." In fact, no more than 6, and perhaps only one, of the 12 children had this form.
- The children's outside medical records show that only 2 of the 12 patients exhibited their first symptoms within days of their MMR vaccination. The paper changed this to 9 of 12 patients, and reported a fabricated "average" time of onset as 6 days. (Maybe they averaged the patients whose symptoms started many months after vaccination with those whose symptoms started many months before!)
- The hospital pathology department found non-specific colitis (inflamed bowel) in 3 of the 12 patients. Wakefield "re-examined" the specimens and raised this number to 11.
- All of the children suffered from severe constipation, a symptom not associated with colitis and not reported in the paper. Wakefield, a gastroenterologist, did not treat them for this symptom.
-Review by the medical board found that ten children were subjected to unnecessary invasive procedures without ethical approval.
All in all, the British Medical Journal article provides a flat but breath-taking indictment of Wakefield, his partner John Walker-Smith, and their eleven careless but willing co-authors. Does this mean parents and celebrities will finally abandon the mercenary Wakefield and his hypothesis? Probably not. Autism is a devastating disease, and one of parents' strongest impulses when faced with such a diagnosis is to identify an external cause or culprit. Although some parents, when faced with lack of evidence, can accept that the cause of a terrible event is unknown, others cannot. They need someone to blame.
We're creatures who look for patterns to explain our world. When I was a medical student, a young Mexican man came into the hospital with leukemia, which first manifested itself as prolonged bleeding after a tooth extraction. Nothing would shake the belief of this man's family that the dental procedure had caused his leukemia.
So what's the lesson here? Unfortunately, it seems to be that the scientific community has to be yet more vigilant against fraud. When results are truly unexpected, it may be worth asking to look at the raw data. Volunteer peer reviewers can't take that on; it's a responsibility that will need to be shouldered by journal editors in selected (or random) cases. Beyond that, all of us have to guard against our human tendency to reason from anecdote instead of evidence, and our capacity to theorize about vast conspiracies when one person's lies are a lot more likely.
For years, medical researchers have published careful studies failing to support a purported link between childhood vaccinations - particularly the MMR, or measles/mumps/rubella vaccine - and autism. However, many anxious parents and autism advocates have remained wary of vaccines, mostly based on the work of British doctor Andrew Wakefield. Even after the medical journal Lancet retracted Wakefield's original 1998 article on the subject as fatally flawed - even after Wakefield and a co-author were "stuck off the medical register" (i.e. had their licenses revoked), true believers maintained that Wakefield was a martyr to the truth, hounded by greedy vaccine makers intent on filling their pockets.
Yesterday the British Medical Journal published an article by Brian Deer, a journalist who has gone back and carefully examined the medical records and interviewed the parents of the original 12 patients discussed in the Wakefield paper. BMJ's conclusion: the retracted Lancet paper did not reflect merely carelessness or scientific misconduct. It was barefaced fraud for financial gain.
First, the study. Wakefield's team examined 12 children with neurological and gastrointestinal complaints. Most of the parents suspected the MMR vaccine as the cause of their child's symptoms. Wakefield "found" colitis in most of the children, and also "found" that on average, onset of symptoms occurred 6.3 days after the vaccine.
What Wakefield did not disclose in the article was that for two years before and during the research he was being lavishly paid by a tort lawyer hoping to build a case against vaccine manufacturers. Wakefield's take was $668,000 plus expenses. In a confidential grant application to the UK government's Legal Aid board, Wakefield and his lawyer sponsor voiced their conviction that a vaccine-autism link existed. Their grant application stated, "The objective is to seek evidence which will be acceptable in a court of law of the causative connection between either the mumps, measles and rubella vaccine or the measles/rubella vaccine and certain conditions which have been reported with considerable frequency by families of children who are seeking compensation." (The italics are mine.) In other words, give us the money and we'll give you the evidence to make you rich.
Wakefield knew what he had to show, and he proceeded to manipulate his data to show it. Almost nothing about the data presented in the Lancet paper was fairly represented.
- Patients were recruited through an anti-vaccine group called JABS. Parents were predisposed to blame their children's symptoms on the MMR - in fact, the lawyer had sent out a letter to his clients telling them this doctor believed there was a link. Parents were then interviewed and gently prodded to remember symptoms starting shortly after the vaccine. In one case, a parent remembered that a child was "pale" one week after the vaccine. Prompted further by the interviewer, the parent admitted that the child may have had fever or delirium. In the paper, the fever and delirium are reported as fact.
- The paper reported that 9 of 12 patients developed normally and then regressed, a form of autism called "regressive autism." In fact, no more than 6, and perhaps only one, of the 12 children had this form.
- The children's outside medical records show that only 2 of the 12 patients exhibited their first symptoms within days of their MMR vaccination. The paper changed this to 9 of 12 patients, and reported a fabricated "average" time of onset as 6 days. (Maybe they averaged the patients whose symptoms started many months after vaccination with those whose symptoms started many months before!)
- The hospital pathology department found non-specific colitis (inflamed bowel) in 3 of the 12 patients. Wakefield "re-examined" the specimens and raised this number to 11.
- All of the children suffered from severe constipation, a symptom not associated with colitis and not reported in the paper. Wakefield, a gastroenterologist, did not treat them for this symptom.
-Review by the medical board found that ten children were subjected to unnecessary invasive procedures without ethical approval.
All in all, the British Medical Journal article provides a flat but breath-taking indictment of Wakefield, his partner John Walker-Smith, and their eleven careless but willing co-authors. Does this mean parents and celebrities will finally abandon the mercenary Wakefield and his hypothesis? Probably not. Autism is a devastating disease, and one of parents' strongest impulses when faced with such a diagnosis is to identify an external cause or culprit. Although some parents, when faced with lack of evidence, can accept that the cause of a terrible event is unknown, others cannot. They need someone to blame.
We're creatures who look for patterns to explain our world. When I was a medical student, a young Mexican man came into the hospital with leukemia, which first manifested itself as prolonged bleeding after a tooth extraction. Nothing would shake the belief of this man's family that the dental procedure had caused his leukemia.
So what's the lesson here? Unfortunately, it seems to be that the scientific community has to be yet more vigilant against fraud. When results are truly unexpected, it may be worth asking to look at the raw data. Volunteer peer reviewers can't take that on; it's a responsibility that will need to be shouldered by journal editors in selected (or random) cases. Beyond that, all of us have to guard against our human tendency to reason from anecdote instead of evidence, and our capacity to theorize about vast conspiracies when one person's lies are a lot more likely.
Friday, October 15, 2010
Book touring
I'm still on book tour, in Minneapolis now. During a Noyce Foundation meeting in California, I did a reading at Kepler's, a great bookstore in Palo Alto; now I've stopped in Minneapolis for a five-day stay on the way home.
At Kepler's, extended family and foundation friends did a great job of getting the word out. I gave a short talk and readto a group of forty friendly faces (this sentence is becoming worth of Mr. Garrulity), and Kepler's sold thirty books and ordered more.
I came to Minneapolis in part to do a live interview on KFAI radio for a show called "Write On!" where I was intervewed by Charlotte Sullivan. Ironically, Charlotte ended up being busy the night of the show, so we taped a phone interview ahead of time. Maybe her engineers edited out all the times I said "um," because the interview came out sounding very smooth and comfortable. I believe you can find the interview here. The first half hour is an interview with noted science fiction and cyberpnk writer William Gibson, though he comes off as rather distracted.
My second task in Minneapolis is to work with editor and consultant Ian Leask, a deep reader and astute literary tactician. He's helping me get over the last hump with my new book, a mythic coming-of-age novel called The Beechwood Flute. The work is intense and exhilarating enough to wake me at four in the morning full of ideas.
Finally, I'm doing three more readings in Minneapolis. Last night about seven people came and I sold four books, but it was fun, and I signed another pile while talking to the event organizer. I have two more events Saturday before returing home Sunday.
A highlight of the trip has been connecting with old friends, including a group of my fellow residents in internal medicine. Twenty-five years ago we spent three years together learning medicine and fighting daily to save and heal a long line of decent, brave, very sick patients. We were exhausted and stressed but living with a sense of burning purpose. One of them almost died the first year, when he fell asleep at the wheel after a particularly grueling night on call. To see them again has been wonderful.
I'm learning that it's definitely true that traveling for book signings is definitely not likely to be "worth it" financially, but it's worth it in terms of connecting with old friends and stimulating new ideas.
At Kepler's, extended family and foundation friends did a great job of getting the word out. I gave a short talk and readto a group of forty friendly faces (this sentence is becoming worth of Mr. Garrulity), and Kepler's sold thirty books and ordered more.
I came to Minneapolis in part to do a live interview on KFAI radio for a show called "Write On!" where I was intervewed by Charlotte Sullivan. Ironically, Charlotte ended up being busy the night of the show, so we taped a phone interview ahead of time. Maybe her engineers edited out all the times I said "um," because the interview came out sounding very smooth and comfortable. I believe you can find the interview here. The first half hour is an interview with noted science fiction and cyberpnk writer William Gibson, though he comes off as rather distracted.
My second task in Minneapolis is to work with editor and consultant Ian Leask, a deep reader and astute literary tactician. He's helping me get over the last hump with my new book, a mythic coming-of-age novel called The Beechwood Flute. The work is intense and exhilarating enough to wake me at four in the morning full of ideas.
Finally, I'm doing three more readings in Minneapolis. Last night about seven people came and I sold four books, but it was fun, and I signed another pile while talking to the event organizer. I have two more events Saturday before returing home Sunday.
A highlight of the trip has been connecting with old friends, including a group of my fellow residents in internal medicine. Twenty-five years ago we spent three years together learning medicine and fighting daily to save and heal a long line of decent, brave, very sick patients. We were exhausted and stressed but living with a sense of burning purpose. One of them almost died the first year, when he fell asleep at the wheel after a particularly grueling night on call. To see them again has been wonderful.
I'm learning that it's definitely true that traveling for book signings is definitely not likely to be "worth it" financially, but it's worth it in terms of connecting with old friends and stimulating new ideas.
Subscribe to:
Posts (Atom)
