Ignatius J. Reilly, Gene Hunter?!?

Recently Dolores and I received a link to Viruses on Time, a blog post about virology-themed covers of Time magazine, sent to us via a mailing list we joined as part of an ESF Conference on Truth and Trust in Images, which we attended and presented at last fall. While the virus nerds were probably pondering the ethics and aesthetics of the portrayals, I was struck by the March 9, 1981 cover featuring Genetech's Herbert Boyer.  My soft-and-squishy-but-bonily-encased pattern recognition engine leapt immediately into action, and reported (and I paraphrase): "Holy Crap, that's Ignatius J. Reilly".

For those who don't know, Reilly is the protagonist of one of my favourite novels,  A Confederacy of Dunces, introduced to me by my 12th grade English teacher, who could clearly spot a misfit when she met one. Maybe it's a coincidence, but Confederacy won the Pulitzer Prize in 1981 (posthumously, another interesting, but sad, story), the same year as that Time cover.

Maybe the cover artists were one in the same (unlikely).  Or maybe the Time's cover artist was a fan of the book (possible).  Or maybe it was just "in the air".  I'd like to think it was the latter, since it faintly echoes our (and many others', of course) belief that scientific visualizations are, extricably, a product of the broader culture in which they are created.

Youtube channel

In my quest to bring our lab into the 21st century, I have started porting more of our videos onto BSL's YouTube channel.  Please subscribe if you are wont to subscribe to such things. We'll try to keep adding our visualizations, both old and new, as time and energy permit.

A Prof's Life

Stumbled across this blog post of Fil Salustri's, which I think nicely describes the life of a Professor. Fil and I were grad students together at UofT, and had both done our undergrad degrees there.  He'd started two years earlier, so whenever I thought I'd been at UofT for too long -- and I was there for nearly a decade -- I had Fil to look up to (sorry Fil, couldn't resist :-)

Wellcome to the Circulation

Thanks to our friend and self-styled quasi-media activist Roberta Buiani, currently transplanted to Cambridge, for sending this link to "William Harvey and the circulation of blood", a great 1978 film from the Wellcome Collection. (Note early on the comments about Galen being so revered for centuries that no one dared question his presumptions about the circulation. Guess it didn't hurt that Harvey had tenure and the ear of the king :-)

Enjoy it while you can, because the closing of the Wellcome Trust Centre for the History of Medicine earlier this year may be an unwelcome (sorry :-) harbinger of things to come in the Dark Age Ahead...

Too many papers, too little time...

Doing a literature search this morning, I stumbled across "The reliability of high resolution MRI in the measurement of early stage carotid wall thickening" by Boussel et al. What struck me immediately was their Figure 2 (left), which nicely shows that MRI tends to overestimate normal wall thickness compared to ultrasound intima-media thickness (IMT) measurements, something the authors said was "potentially due to the inclusion of the adventitia by MRI".

As it turns out, I had downloaded a copy into my Papers database on January 5, 2008. Why is this relevant? Because, in Febrauary 2008 we submitted a paper entitled "On the overestimation of early wall thickening at the carotid bulb by black blood MRI...". Looking at their figure now (and probably then), I noticed how similar it looked to our predictions. So now, almost two years later, I plotted their data (hand digitized using a great shareware program, getdata) over the data from Figure 8 of our paper. As shown to the right, the trends are indeed remarkably consistent.

There are some discrepancies, of course. Boussel et al.'s data were acquired with 0.6-mm pixels, but would seem to fall between our predictions of 0.3 - 0.5 mm pixels. This may be because their images were segmented manually, whereas our predictions were based on automated edge-detection. (Thanks to Bill Kerwin from the University of Seattle for later explaining to me why the latter probably overestimates the problem relative to the former.)

It would have been nice to have included this comparison in our paper, for it bolsters our conclusion that the spatial resolution of MRI, rather than the adventitia, is the culprit. Oh well, better later than never! But, and per the title of this post, there are just too many papers out there to be able to read and remember them all. My Papers database has 3446 PDF files as of today, most of them probably relevant to my research, but the vast majority probably inhaled and forgotten like a fast food dinner, rather than savoured and remembered like a gourmet meal. A weak metaphor perhaps, but maybe science could do with a "slow food" movement...

All hail Lewis Fry Richardson (1881-1953)

First time I came across his name was indirectly, as an undergrad, learning about something called "Richardson extrapolation", a method for inferring the convergence of a numerical analysis. Then, much (much) later, as I was rethinking turbulence in blood, I came across Richardson as one of the pioneers of modern turbulent flow theory, and as the author of this unsurpassedly clever and concise description of turbulence:

Big whorls have little whorls
that feed on their velocity,
And little whorls have lesser whorls
and so on to viscosity.

Shortly thereafter -- OK, so I didn't pay close attention to my reading of Gleick's "Chaos" many years back -- I learned about Richardson's anticipation of fractals via his musings on the measurement of coastlines, something that arose out of his interest in divining the mathematical rules underlying human conflict; and of his anticipation of chaos theory through his pioneering work in weather prediction. By virtue of what some might call his mania for the latter, he arguably founded modern numerical analysis and finite difference methods.

If that weren't enough, thanks to my reading of Abigail Swillens' fine PhD thesis, I learned that Richardson filed the first patent for underwater echo-ranging (apparently inspired by the sinking of the Titanic), a precursor to sonar and, by extension, medical ultrasound.

In short, virtually everything I do in my research can be traced back, with far fewer than six degrees of separation, to Lewis Fry Richardson. For more about Richardson's life and works, there's the at-your-fingertips Wikipedia of course, but also a nice 1998 review in the Annual Reviews of Fluid Mechanics.

Bibliometric indicators

Many of us have a love/hate relationship with bibliometric indicators, those numbers imposed on, and by, us to quantify the unquantifiable. The latest issue of Radiology provides a nice overview of these, giving some historical perspective and summarizing the pros and cons. (The previous issue of Radiology also provides what appears to be the first in a series of tips and tricks on the use of Powerpoint, something else with which many of us have a love/hate relationship...)

Update: And speaking of love/hate relationships, The New York Times has a funny/frightening article about the US military obsession with Powerpoint.

A personal history of diversity

Just came across the BMES Inaugural Diversity Lecture by Sheldon Weinbaum, in which he gives a broad and personal history of efforts to promote diversity. I've always been impressed by the originality of his scientific ideas, and the passion with which he pursues, presents and defends them. So it should have come as no surprise -- but did anyway -- that he has been on the front lines of the culture wars for nearly half a century(!).

Aneurysms, simulations, and Rolling Stones

In a recent counterpoint to an opinion piece in the American Journal of Neuroradiology on imaging follow-up after coiling of intracranial aneurysms, intenventional neuroradiologist Jean Raymond speaks the words that every simulation jockey in our field dreads (or should dread) to hear:

"Good methodology and hard work may provide some answers to our questions, but contrary to what the Rolling Stones’ song says, time is a false friend in most research enterprises, particularly if it is retrospective (often badly planned and conducted in our young field) or when expediency is favored over accuracy, as exemplified by an expanding literature on computer simulations projecting speculations over the lifetime of individuals, to the detriment of prudent assembling of clinical evidence."

Ouch. It's unfortunate that Dr. Raymond paints with such a broad brush, but I can't say I completely disagree with his sentiments. There's good simulation work in the field of aneurysm research, but a bandwagon's a bandwagon, and it's to our detriment that some, er, less-than-good work gets published, for it only feeds clinical suspicion of all of our efforts. On the other hand, "you can't always get what you want; but if you try sometimes you just might find you get what you need."

Language barriers in translational research

Further to my earlier post about us biomedical engineers tending to focus on the tip of the iceberg when it comes to investigating biomechanically-based cardiovascular risk factors, a recent editorial in Nature Medicine, on the challenges of translating research into practice ("In the Land of the Monolingual"), suggests that this iceberg is also a Tower of Babel (how's that for mixing metaphors!).

On the bright side, the editorial highlights the efforts of Toronto's MaRS Centre to break down the language barriers separating ivory and business towers. It is also an encouraging echo of our own small efforts to break down the verbal and visual language barriers separating bench and bedside.

Somewhere over the rainbow

Among the many things I learned at the Gordon Research Conference (GRC) on Visualization in Science & Education -- an amazing experience, by the way, and highly recommended -- was how poorly regarded we scientists and engineers are when it comes to the use of colour in our visualizations. I can't say I disagree, since in our biofluids community, for example, one can, given an anonymized figure, often guess which software was used to produce it, in large part because users tend to stick to the default colour scheme, which is typically some variation on the execrable rainbow map.

Upon my return from the GRC, I happened upon a Facebook posting by the nice folks at Tecplot (the software we use for most of our visualizations, and also highly recommended), pointing to an enlightening IBM report entitled "Why should engineers and scientists be worried about color" (from where the above image is taken). With great enthusiasm I passed it along to some of the GRC attendees with whom I had discussed my longstanding, albeit naive, interest on this topic. Only later did I realize that the report was actually from the mid-90s, and that those to whom I'd sent it were well aware of it. (I blame it on my jetlag. :-)

Anyway, one of these individuals, Robert Simmon from NASA (check out his work at the Earth Observatory), provided me with the following useful links for helping to design more suitable colour mappings.

http://colorusage.arc.nasa.gov/ColorTool.php
http://colorbrewer2.org/

If you come across others, please feel free to add a comment to this post.

A couple of academic satires

Taking a day off after the ASME Summer Bioengineering Conference, I started, and then promptly finished, "Portuguese Irregular Verbs",  an affectionate satire of academics, tracking the exploits of Professor Dr Moritz-Maria von Igelfeld and his colleagues, Professor Dr Detlev Amadeus Unterholzer and Professor Dr Dr Florianus Prinzel, denizens of the fictional -- in case there was any doubt -- Institute of Romance Philology in Regensburg.  Many thanks to Frau Dr Marilyn Riederer, for sending this book to Frau Dr Dr Dolores Steinman; and to Frau Dr Dr Steinman for encouraging me to read it (and also for encouraging me to take a day off).

While I'm at it, I also recommend "Bellwether", a not-inaccurate satire of life at a research institute, which I read several years ago while working at, well, a research institute.

From biomedical engineering theory to clinical practice

As biomedical engineers, we are often engaged, explicitly or implicitly, in the search for better markers of risk for the diseases or treatments that motivate our research.  A recent scientific statement from the AHA, "Criteria for Evaluation of Novel Markers of Cardiovascular Risk", serves as an informative and important reminder that our efforts are usually just the tip of the iceberg.