Wednesday 11th June. Jump to:
Some pre-conference musings
William Haseltine (medicine that’ll change the world)
Josef Penninger (genetics)
Kevin Kamler (extreme telemedicine and Everest)
Laurie Garrett (SARS in China)
Conversation Break (I got a tour around the Kimmel Center)
Marvin Minsky (crazy-ass future and AI talk)
Sherwin Nuland (‘why do we want to live forever?’)
Keith Black (getting rid of invasive surgery)
David Macauley (writer and illustrator extraordinaire, and I got to talk to him!)
Michael Jackson (biodata-mining)
IBM Life Sciences Reception
(You can skip this bit if you only want to read my notes on the speakers)
As invariably happens these days, it all began with an email. Back in 2001, I managed to get myself invited to speak at the 11th Technology, Entertainment and Design conference (TED11) in Monterey, California; you can read my report of that conference here. Since then, I occasionally get sent mass mailings from the TED conference organiser, Richard Saul Wurman, usually on how I should sign up for the next conference. As it costs around $3000 to simply register at TED conferences, it’s not too surprising to see why I tend to ignore those emails.
So when I received an email from Richard in early May, it was just luck that prevented me from sending it straight to my ‘Non-critical’ email folder (it’s not as bad as my junk mail folder, but not as good as my meticulously maintained inbox). The email turned out to be an invitation to attend the TEDMED3 conference in Philadelphia in June, TEDMED3 standing for the third TED Medical conference.
It didn’t take long for me to make up my mind about attending and book the flight over to Philadelphia, hopefully before they realised that they’d made a terrible mistake in inviting me. I found TED11 to be one of the most exciting and intellectually stimulating events I’ve ever been to, both in terms of the speakers present and the people I was able to talk to. Sure, I’d still have to pay for my own accommodation and flights to Philadelphia, but a TED conference is a sure thing. With the speaker list including people like Oliver Sacks, Marvin Minsky, Laurie Garrett and Dean Kamen, there wasn’t much of a dilemma.
Some pre-conference musings
I used to enjoy flying, but I think the magic disappeared when I was about 15 years old. Nowadays I spend most of my time trying to keep myself hydrated and figuring out what the least-worst movie to watch. Still, it’s not really the actual flying that bothers me, it’s everything that surrounds it – the stress of having to get to the airport in time, check in, do the baggage checks, waiting around for hours and spending ridiculous amounts of time in queues.
Since I don’t have a laptop, I brought a little paper notebook with me to the conference. One of the first things I wrote down was ‘Queue theory’. Queue theory is actually a real area of research, but in this case I’m just using it as a springboard for a digression. If you are concerned – even mildly so – about reducing the amount of time you spend in queues, you probably have some ideas about how to improve queueing systems, especially those at airport immigration. I have a lot of ideas. These include only having one main queue that splits off into subqueues for each desk, and some kind of adaptive system that routes around damage (i.e. people who are spending far too much time at an immigration desk).
Anyway. I got into Philadelphia on Tuesday (June 10th) and made my way to the Bank Street Hostel in the city centre, possibly the cheapest place that any conference attendee was staying at…
‘So, which hotel are you staying at? The Doubletree? The Hilton?’
‘Uh, I’m actually staying at a (mumble).’
‘A hostel. The Bank Street Hostel.’
In the milliseconds after such a conversation, I can visibly see the person I’m talking to ratchet down their monetary estimation of me; it’s a barely perceptible tilt backwards of the head and a slight raise of the eyebrows. It’s cross-cultural, I tell you. Margaret Mead probably found tribal villagers doing exactly the same in some Pacific jungle.
Philadelphia is actually a pretty interesting city which I’m told that many schoolkids make pilgrimages to, on account of it being the colony where the US declared independance and did any number of things to bug the British. All the places named ‘Independence Hall’, ‘Independence Square’, ‘Independence …’ clued me into this fact which I was not aware of when I landed. The Philadelphia Museum of Art is also world-famous, for a couple of reasons; it has an outstanding collection, and it’s the place where they filmed Stallone running up the steps in the first Rocky movie. Apart from that, it’s not that interesting.
One of the many differences between TED and other conferences is that attendees are told to dress down. That means no suits and no ties; as Richard said, if you’re found with a tie, he’ll personally cut it off with a pair of scissors, unless it’s a really nice tie in which case you’ll be allowed to take it off and keep it. From TED11 it seemed that most people were happy enough wearing shirts/T-shirts and trousers, so I felt it was perfectly fine to turn up wearing shorts and T-shirts on the first day for registration, especially since Philadelphia is quite humid. However, when I walked through the front doors of the Merriam Theatre where the conference was being held, a security guard came up and said:
‘I’m sorry, you’re not allowed in here.’
I was taken aback; this was definitely the right place, there were TEDMED3 posters all over the place. Had I somehow transformed into the very picture of a malcontent?
‘There’s a conference going on.’
‘I’m here for the conference!’ I protested.
‘Oh, okay, sorry,’ he said, backing off (although undoubtedly keeping an eye on me, sneaky tourist-lookalike that I was).
After that, I managed to register fine and positively strutted around with my ID nametag, daring any other security guards to try and throw me out. Alas, it did set the scene for the rest of the conference in that I was perhaps the third or fourth youngest attendee there (and probably the second-youngest looking one, the youngest being about ten years old) and hence was treated a little dismissively by some.
Being at a conference on your own can be quite a lonely affair, moreso when a lot of people aren’t willing to talk to you because you obviously don’t have anything to offer them (namely money or opportunities). I therefore resolved to find myself a group of young people that I could hang around with for the four days of the conference, and I was successful almost immediately when I spotted around eight twenty-somethings from the MIT Media Lab hovering around the theatre lobby. I introduced myself and asked if I could tag along with them to lunch, which I think they were a bit surprised by but were happy enough to acquiese. They turned out to be a very fun group to hang around with, all students or friends of John Maeda, an MIT Media Lab Professor. John had an interesting habit of repeating my name multiple times whenever he saw me, so as to aid his terrible memory for names.
Enough rambling – onto the conference!
Wednesday 11th June – Session 1
The first presentation in the first session of the conference was not from a renowned doctor or scientist, but rather a tap dancing performance by legendary LaVaughn Robinson and students from the nearby University of the Arts. It was a wonderful start to the conference and put everyone in a relaxed mood.
Two things went through my mind during the performance. The first was that we’re so accustomed – you could almost say conditioned – to watch images on screens that there was a tangible pull to watch the blown-up image of the performers projected onto the screen instead of the real performers themselves*. The second was that these guys would be excellent Dance Dance Revolution players.
*It was commented later on that the whole conference had a very television-like feel to it. Everyone ran very slickly, there were cameramen lurking about videoing the speakers who were then projected onto a screen in the middle of the stage and there was a nice mixture of presentations.
William Haseltine is the CEO of Human Genome Sciences Inc. and was a professor at Harvard Medical School; quite an established scientist and entrepreneur in the medical sciences. I understand that Richard chose Haseltine to give the first proper presentation in order to supply a broad outline and context to the conference.
Haseltine spent his WTU* talking about the four areas of medicine that have been most influential in the past 25 years (proteins, antibodies, surgery and imaging) and the four areas he believes will be most influential in the next 25 years. These are gene therapy and RNA interference, the use of cells and stem cells as therapy, the integration of artificial materials into the human body (e.g. prosthetics) and the interface of humans with electricity. That could mean anything, really, but I recall he was referring to reading and interpreting neural signals.
I was not particularly impressed with Haseltine’s presentation; it was far too diffuse for my liking, and I don’t think I learned anything new. It seemed that other attendees agreed with me. I appreciate the need for a more general ‘keynote’ presentation, but this could’ve been better.
*WTU: Wurman Time Unit. Coined by Richard Rockefeller and usually equates to fifteen minutes of time on stage. For favored speakers, this can expand to thirty minutes. You can sense the end of a WTU by increased movement on the part of Richard Saul Wurman.
Professor of Immunology and medical biophysics at the University of Toronto. He started off talking about a ‘genetic map of the world’ and then moved onto to the OPGL and DREAM genes, which are responsible for bone loss and pain perception respectively. Seemed a bit too technical for this sort of conference.
I remember that Penninger expressed a desire to measure the activity of every single neurone in the brain to investigate the circadian rhythm. I’m not sure whether he really believe he could do this, because it is frankly impossible for the next couple of decades at the very least.
He did have an interesting unifying theory of autoimmunity that stated that autoimmune diseases (e.g. multiple sclerosis) only occur as a result of outside infections that initially compromise the immune system. I can’t remember the details and I’m not an immunologist, but it seemed to make sense to me.
A physician who was notably present at Everest during the events covered in the book ‘Into Thin Air’. Ken’s talk was on ‘extreme telemedicine’ – in other words, conducting medical treatment with remote support in a hostile environment such as Everest or indeed outer space. However, he spent most of his time going over the events that occurred when he was at Everest and a number of climbers died. Very compelling stuff – it gave me a real impression of how crazily dangerous climbing Everest is – but it seemed as if Ken had given the talk a number of times before.
Pulitzer prize-winning medical and science writer. I was quite looking forward to this talk because Garrett’s was involved in a nasty spat on MetaFilter not too long ago – and of course she is quite a good journalist.
Garrett started out by saying she didn’t like the hype about the medical and scientific community having beaten microbial diseases via antibiotics. While this probably wasn’t news to the audience, it is to the general populace and it really is a dangerous assumption to think that we have all such diseases beaten, especially with vancomycin-resistant bacteria, etc etc.
I have in my notes that she said the French vetoed the Nobel prize going to the team who beat smallpox in the 70s or 80s. Nasty French, as usual. Garrett believes that the state of readiness for another smallpox vaccination on the scale of that in New York in 1947 (6 million vaccinated in 3 weeks) is far worse now than it was then. Back in the 40s, just after the war, the US medical system was supposedly very well run, well funded and well respected. These days, the implication by Laurie is that this is not the case.
Garrett then asked what would happen if there was a SARS outbreak in a US city? Would the response be better or worse than, say, Hong Kong, China or Singapore? I don’t think she actually answered this question, but she did talk about what she’d learned from her recent trip around China (she’d told Richard that she’d only come to the conference if China got SARS under control, so at least one person thinks things are getting better).
Researchers have found that the SARS coronavirus has been found in a large range of animal species; Garrett believes this is because China has a number of ‘exotic live animal markets’ which consist of overcrowded animal cages jammed right next to each other, overflowing with faeces; perfect places for viruses to jump across the species barrier and mutate all over the place. Indeed, a great number of SARS infections have been tracked back to cooks who deal in exotic animals.
As for the Amoy Gardens outbreak in Hong Kong, Garrett outlined a number of different theories that might explain how one person (a ‘superspreader’) managed to infect hundreds. The first is that there are two strains of SARS with differing virulance. The second is that the external plumbing employed in Amoy Gardens is to blame. The U-tubes in plumbing (you can see them under sinks and toilets) are supposed to always be filled with water. However, due to a combination of them being outside and water conservation efforts in Hong Kong, the water in this particular Amoy Gardens U-tube had dried up. As a result, the faeces produced by the superspreader underwent aerosolization, zipped through the rest of the plumbing system in no time and infected hundreds of people.
A third theory favored by the Amoy Gardens residents is that the bizarre pattern of infection was caused by contact-transmission via the fiendishly strange elevator button system. Who knows.
- Garrett mentioned that while the rest of the world was watching the war on TV, in south Asia they set up 24-hour television networks dedicated to SARS.
- Political infighting was the reason behind the Mayor of Beijing and the Health Minister both being fired; it was a struggle between the old guard and the new Prime Minister.
The May Day shutdown in China was widely publicised and enacted by the Chinese government to prevent transmission of SARS. However, Garrett contends that the real result was the many people thought the shutdown was merely a prelude to martial law being declared, and so thousands of people fled Beijing to the countryside, including many university students – thus further spreading the virus.
She concluded by saying that China’s efforts to contain SARS to date have all involved mass mobilisation of people and resources. This can only work to a point before paralysing the country’s economy. Instead, she believes that China needs a sustainable and permanent response to SARS; in other words, a much improved public healthcare and disease control system.
On a somewhat unrelated note, I first heard of Laurie Garrett when she made something of a splash on MetaFilter, in which she came off looking very bad indeed. While I was very impressed with her talk at the conference, she did seem quite acerbic to me and I don’t think it’d be pleasant to get into an argument with her.
Conversation Break: Seed Magazine
I’d managed to get on the list for a tour around the Kimmel Center with the site’s Project Manager, Sandy McKee. The Kimmel Center is a very impressive building housing two concert halls; the 650-seat Perelman Theatre which can rotate an entire third of its space, and the 2500-seat Verizon Hall.
On entering the Kimmel Center, you have an immediate impression of open space due to the high curved glass ceiling. I remarked to McKee that it seemed very much like the new glass roof at the British Museum. She replied that she’d just visited it a few weeks ago and thought it was very cool – although she claims that the Kimmel Center thought of it first!
The Verizon Hall feels like the insides of a string instrument; the coloring and curves mimic those of a cello or violin. It also has excellent acoustics with surrounding baffle rooms, and sits on a rubber pad to dampen subway vibrations (apparently this is a real problem at the Lincoln Center in New York).
Minsky needs no introduction – he’s one of the most famous pioneers of AI in the world, and to boot, I just found out that he invented the confocal microscope. He is one seriously intelligent guy. Alas, it has been noted by many people (including people I have spoken to at MIT) that he is going a little weird and somewhat crazy in his old age. With that in mind, Minsky’s talk:
Minsky displayed a list of problems that need to be solved. These will be familiar to most of you: infections and epidemics, education, pollution, longevity, energy, wealth distribution, etc. Most of these, he argued, were a result of overpopulation, and so he asked the question: how many people do we need? If, for example, we only had 500 million people in the world, most of these problems would be solved. There wouldn’t be any overcrowding and we could farm everywhere and dump our rubbish without worrying about serious environmental effects.
So, how could we solve this overpopulation problem?
Solution 1: Don’t reduce the number of people, just make them smaller – literally! If you made every six inches tall, why, we’d have plenty of space! To his credit, Minsky did concede that there might be some fundamental limits of miniaturisation with regards to humans in that you can’t really make neurones any smaller than they currently are. It is still weird though.
Solution 2: Currently, when a child is born, it has two parents that each contribute 23 chromosomes. How about a world in which each parent only contributed one chromosome, giving each child 46 parents? This would allow more people to have a genetic contribution in a child and hopefully reduce the birth rate. What’s more, the death of one parent wouldn’t be as traumatic if there were still 45 left.
Solution 3: Time-sharing (cf. ‘Against the Fall of Night’ by Arthur C. Clarke). You could have five billion people in the world, but only 500 million of them are alive and walking around at any one time – the other 4.5 billion are in storage.
Solution 4: Mind-sharing (c.f. Benford’s ‘Galactic Center’ novels). This means ‘more ways to think’ inside each brain’.
At this point there appears to be a discontinuity in my notes, perhaps attributable to my brain exploding. Anyway, Minsky started talking about the problems of increased longevity. If the age of retirement remains at 60 or 70, then we are rapidly approaching an era in which there will only be one active worker per retiree. This is clearly not a good thing when it comes to the economy and quality of life. Minsky presented a rather pretty looking graph and then spoiled the moment by saying, ‘I just made this graph up – it means that I don’t have to do any work!’
Minsky thus concludes that we need robots to do the boring things so ‘people can do more important things.’ Therefore, we need AI! AI could also help us solve those problems that are too hard for human brains (he didn’t elaborate on this).
Things went downhill from here, as Minsky began to vigorously beat the joint strawmen of psychology and philosophy. Psychologists, philosophers and neuroscientists (and I suppose anyone apart from AI researchers) have been asking the wrong questions all along, Minsky claimed. They’ve been asking, ‘What are feelings? What are emotions? What is consciousness?’ instead of more novel and useful stuff like, ‘How do we solve problems? How does common-sense reasoning work?’
I don’t really want to go into how much is wrong with this claim – for one thing, people have been thinking about problem solving for decades, if not centuries – so I will once again move on.
Minsky finished with his new idea for an intelligent machine called a ‘cognitive sandwich’. This machine would be designed akin to Freud’s ego, supergo and id, and for some reason Minsky believes it will work where others have failed. Now, I don’t really want to say that it’s not a new idea because Minsky is a very bright guy, but I can’t see what’s new with this model. Time will tell, I suppose.
Clinical Professor of Surgery at the Yale School of Medicine. Nuland began by drawing an interesting parallel with Narcicuss looking at his reflection in a pool in Greek mythology, and people looking at molecular cell biology today in the hopes of a treatment that will keep us young and beautiful forever. Just as Narcicuss’ heart was broken as he futilely struggled to reach his reflection, so will ours, he claimed.
In today’s atmosphere of biotech hype, I personally think it is important to remain realistic about the potential of new discoveries. Nuland is not just realistic about them, he’s pessimistic. He talked about telomeres. While it’s true that telomerase treatments proved to be a false hope in conferring immortality (to be honest, I don’t think that many people really thought it’d work) that’s hardly the whole story.
‘What’s the point of immortality?’ asked Nuland. ‘What good does it do?’
What indeed? It doesn’t do any good at all, from the traditional perspective, but then neither does watching the Simpsons on TV or playing football. It’s a bit of a non-sequitur. From my point of view, I think it’d be quite fun to live an extra few decades – immortality is stretching it a bit though.
Nuland then moved onto a more fruitful track that proved to set a major theme for the conference; compressed morbidity. He believes that instead of trying to increase longevity, medicine should concentrate on enhancing life and reducing the amount of time that we are unable to live independently.
It turns out that the main reason why we can’t live independently when we are old is strength – we don’t have the strength to move around or to carry out important tasks. To attack this problem, Nuland talked about a study in which they successfully made the ‘oldest old’ (85+) exercise to increase their strength. A careful exercise regimen actually doubled the subject’s strength in only six weeks and it appeared that further gains in strength were attainable. An added benefit was a reduced incidence of osteoperosis. The same could clearly be attempted with minds, by keeping them alert and active.
All in all, an interesting talk. I might not agree with everything Nuland said, but I found his views on compressed morbidity to be quite novel (for me).
Neurosurgeon at the Cedars-Sinai Medical Center in LA. Black has conducted over 3000 operations to remove brain tumours and his work inspired an episode of ER. He’s currently working on a way to non-invasively remove brain tumours. The current practice is to simply cut them out, which clearly is not a desirable thing to do because you invariably end up removing bits of healthy and useful brain at the same time; it might be (relatively) okay to do this with a bit of skin or lung, but you don’t really want to end up not being able to talk or see after having a brain tumour removed.
In other parts of the body it’s possible to use pharmacotherapies to kill cancer tumours, but these don’t work for brain tumours due to the existence of the blood/brain barrier. In short, this barrier serves as a selective filter that only allows a few types of molecules to travel from the blood supply into the brain. If the blood/brain barrier didn’t exist, then you’d probably die shortly after having lunch due to your brain reacting aversely to toxic chemicals in the bloodstream. So it’s obviously there for a good purpose, but it unfortunately also stops medicines from reaching the brain as well.
Black’s main area of research concentrates on selectively opening the blood/brain tumour barrier so that pharmacotherapies will be able to pass into brain tumours, but not healthy regions of the brain. Judging from the results from clinical trials he demonstrated at the conference, it appears to be working. He remarked that it’s a bit underwhelming to simply give a patient a shot in the arm and then have the tumour shrink, as opposed to cutting their brains open.
Other techniques he’s investigating include using microwaves to ablate tumours non-invasively, as well as using vaccines to stimulate the body’s natural immune response to attack the tumour.
All good stuff, not an overly technical talk but it bestowed the conference with an appropriate aura of seriousness.
David was one of the speakers I was most anticipating; if you haven’t seen his Dorling Kindersley books, then go out and buy them, now, and start with The Way Things Work. David is an excellent illustrator and writer, and can explain the most complicated concepts and machines with only a few drawings and judicious use of woolly mammoths. But what’s he doing at a medical conference?
David is writing a book about the brain. Now, this is something that I was never clear about through the entire talk – I think either he wrote a book about the human body about twenty years ago, and now is going to write one about the brain, or he hasn’t done a book about the human body (but is going to do a brain one).
Anyway, his work on the brain book got held up because he’s spent the last two years writing a book about Mosques, following 9/11. During his talk, he showed three different ideas he has for the brain book.
The first idea is based on a romantic artwork nothing, that of building up the brain layer by layer, based on evolutionary time, in other words, starting with the brainstem, then the hindbrain, then midbrain, then limbic system, etc etc. Seemed pretty enough, but I didn’t find it especially inspiring.
His second idea was a bit better; it traced the events that occur when David (of Michalangelo fame) sees an object and then says something. The book would follow the photons of light travelling through David’s eye, hitting the retina, getting converted to impulses to be sent through the optic nerve, then passing to the visual cortex via the lateral geniculate cortex. From there, we zoom into a neurone, then a synapse, and so on. Not bad, but not hugely innovative.
David’s final idea was of course the best, and sees him in top form. He envisages the brain as a theme park the size of several skyscrapers, in which people walk around examining the ears and the eyes and sliding around nerves and things. It was very funny and cool. The Brain Park would of course be quite expensive to build and run, so it would have things like the Caudate Cafe and Cortex Condos. I look forward to seeing it.
My chat with David Macauley
On Thursday (the next day) I spotted David in the lobby in the morning and had a brief chat with him. As I expected, he was a very interesting and approachable guy. I started off with the traditional ‘I loved The Way Things Work’ shtick and then followed on to talk about the difficulty of learning anatomy from a textbook – it’s so dry. I then launched into the now-famous (and probably apocryphal) story of how a biology student at Cambridge managed to get the top first by writing an essay composed entirely of cartoons.
David mentioned something about paper versus electronic media. He sees The Way Things Work CD as being transitional between the two; it had some animation and some sounds. Apparently the sounds helped people tremendously, which baffled him a bit.
He confessed that he still prefers paper media to anything else; it’s so much more permanent and not subject to the feeling of transcience that comes from electronic products, i.e. that there’s always something better coming out soon. We then started talking about when he had to write the cartoon of Jack and the Beanstalk in only four pages; he found it to be a very difficult task. David claimed that it makes you have to choose your words so much more carefully when you have so little space, and then you really to have work on boiling down to the essence of the story; which is always a good thing to do.
So the idea is that even if you don’t intend to use a cartoon to convey your final message, they are always good to use for intermediate stages for designing and editing stages of (say) an essay; they stop you from fudging things. You can’t fudge cartoons – it’s all out there in black and white.
Vice president for drug discovery at the R. W. Johnson Pharmaceutical Research Institute. I have to confess, I really didn’t like this guy at all. He gave an awful introduction and had a voice that grated. Plus, he had a weird accent – not quite English, not quite Scottish and not quite Californian, but 100% bizarre and annoying.
He did say some interesting things about the difficulty of data analysis these days. When you conduct gene expression experiments on 2000 different cell types, you have several million data points. He thinks he can automate acquiring the data and (here’s the clever bit) cross-referencing it with web-accessible Medline absracts using character and word analysis. So say you wanted to find out what happens to gene expression when you remove or alter a particular gene – you get all the data, and then you press ‘go’ (presumably) and the program will tell you, ‘Hey! It seems that cluster ABC genes were all consistently affected, and then there are a few papers on the function of these genes as well, with reference to the one you altered.’
Could be vapourware though.
There was some really cheesy early 90s CGI, and some stuff about a computational model of diabetes, which I said in my notes might be ‘sorta helpful for clinical trials’.
IBM Life Sciences Reception Reception
And that’s the end of the first day of sessions. A busy day, followed up by a reception by IBM Life Sciences and Bombay Sapphire at Solmssen Court, a few hundred yards down the road. The reception was interesting enough with lots of nice food and sushi, but the place was slightly too small. Still, we had free drinks including probably the most potent cocktail I have ever tasted. It looked blue, so I thought it’d be okay, but it must have been 99% gin and 1% food coloring or something – I stayed well away from it for the rest of the night.
But forget about the food and drink – let’s talk about the swag, for this was the night when we received our goodie bags. I’ll admit that I was a bit disappointed given that at TED11 I received (as a speaker) a very nice Palm VII wireless PDA. This year the best thing I got was a Hasbro board game of ‘Operation’ (which is still admittedly very cool). Other things included the Merck Manual for Medicine, which should be very useful. Some of the MIT folks played ‘Merck lottery’ where you’d guess a three digit number, turn to that page, and whatever was there is what you’d die from. Fun stuff! There was more in the bag but it can’t have been interesting because I’ve forgotten it.