Vector Map of Middle Earth

When my son was 6 years old, I read all four volumes of J.R.R. Tolkien's The Hobbit and The Lord of the Rings to him as part of our bedtime reading. It took months to finish them all, but we had a wonderful time. One of my favorite memories of this marathon was the night we read the scene where Bilbo finds the Ring in Gollum's cave. As the action mounted, my son became increasingly restless, and by the end of the chapter, was practically sitting on my head, scanning the room like a lemur, with eyes as big as saucers.

"What's the matter?" I asked.

"Your Gollum voice was reeaallly scary!"

I flashed back to that scene tonight while reading Tim Bray's Ongoing blog. He is currently reading LOTR to his 8 year-old son, and printed out the wonderful MearthMap Map of Middle Earth as a companion to the book. As he says,

...it’s astounding. It’s in vector format (a 1.2M PDF), which means effectively infinite scaling, which it turns out you really need if you want to be able to locate Dimrill Dale or Durthang. I extracted just the part of the map where the action occurs and printed it out on a huge glossy sheet of paper and it was a fine companion to the story.

It is, indeed, a lovely map. I've got it stashed away on my hard drive against future need, which will hopefully be sometime before my son graduates from high school. This past December, I got new hope that that might actually come true. New Line Cinema announced that Peter Jackson will indeed be filming not only The Hobbit but also another LOTR prequel. If they stay on their proposed schedule, it looks like we will be needing that map around 2010 - 2011 when the films are released.

Of course, after we watch them, we'll have to keep the map out a bit longer, while we watch the entire LOTR cycle all over again on DVD, possibly even in HD.

Obsolete Radiology Skills

The Obsolete Skills blog has a large compendium of skills that are, well, obsolete. (via Daring Fireball)

A partial list of the items collected there:

• Dialing a rotary phone
  
• Putting a needle on a vinyl record
  
• Changing tracks on an eight-track tape
  
• Shorthand
  
• Using a slide rule

This got me to wondering what a similar list of obsolete skills for a radiologist might look like. Here are some of my random conjectures:

Obsolete radiologic procedures (a very partial list):

• pneumopelvigraphy
  
• pneumoencephalography
  
• silicon foam enema of colon (this one still gives me the creeps)
  
• xeromammography
  
• ascending lumbar venography for lumbar disc disease
  

A pneumoencephalogram, circa 1919

Film skills dying out due to the move to digital imaging (some are not yet completely dead):

• exposing, developing, duplicating, hanging or interpreting actual films
  
• recovering silver washed out of film emulsion during developing
  
• using a "hot light" to view really dark films
  
• using a goniometer to measure scoliosis angles on spine films
  
• photographing films for scientific publications
  
• giving lectures with 35 mm slides and a projector

Measuring the Cobb angle in a spine with scoliosis

Snarky things that might be listed by non-radiologists:

• using a stethoscope
  
• remembering how to do a physical exam
  
• having any clinical skills whatsoever

Anyone working in medicine for very long will have no trouble thinking of plenty more examples in all three categories. Feel free to add others to the comment section for this post.

Grand Rounds - Vol 4, No. 25

The latest issue of Grand Rounds (4.25) just hit the newstands on Canadian Medicine.  This episode is filled with tasty tidbits guaranteed to keep me from my regular chores.

Grand Rounds is a weekly anthology of the best writing from the medical blogscape.

One of my posts, titled Music Made Visible, is listed under the section on "Sexing up the fMRI Dossier".

Striking a Blow Against Breast Cancer

It is every physician's duty to help prevent cancer. For radiologists, this usually involves imaging patients with mammograms, chest X-rays, CT and MR scans. However, we are always delighted to learn from the media of low cost, low tech and zero-radiation alternatives that might reduce the risk of breast or prostate cancer. This news is especially intriguing when the suggested therapeutic tools are, respectively, fellatio and masturbation. (via M.D.O.D)

These are research topics where achieving an adequate sample size just doesn't look like it's going to be a problem. A call for research subjects will attract long lines of not only patient volunteers, but also of medical researchers themselves, all willing to put themselves in the hands, so to speak, of science. Even a (ahem) hardened skeptic would be happy to extend the benefit of a doubt to this kind of research.

There's really no reason why you shouldn't go ahead and launch your own personal in-depth study of these pleasant hypotheses. However, before you apply for federal funding or publish your results, a bit of fact-checking would be a handy next step. As the saying goes, "great claims require great proof". These claims are, indeed, great. Really, really, really great. If they're true.

For example, take the first report: "Fellatio may significantly decrease the risk of breast cancer in women". Gosh, if it's on CNN, there must be something to this. Not only that, the page includes related links to the New England Journal of Medicine, and the American Cancer Society.


Well...... no. If you carefully examine the URL for this report, you find that it's actually hosted on the site of some dude who may not actually work in journalism at all. The researcher's name does not turn up in a PubMed search. Googling the name of the "researcher" turns up a much more believable tale -- that this page was a spoof posted by a college student, who apparently

"responded to legal threats by apologising for the false report to his university and to CNN and to "all men who did not take advantage of this article in time ..."

Harumpf. At least one can still make a convincing case for breast palpation among consenting adults.

Next, let's examine the claim that "Masturbation 'cuts cancer risk'". This time though, the link is to the genuine BBC news site. As in the first example, a researcher's name and institution are given -- this time, however, a PubMed search does actually lead to a paper on this topic from a leading English-language urology journal, titled: "Sexual factors and prostate cancer." The abstract is a bit terse, but delivers the following summary:

Ejaculatory frequency, especially in early adult life, is negatively associated with the risk of prostate cancer.

Gosh, this is sounding better and better all the time! However, a True Skeptical Warrior™ should exercise a bit more due diligence than just reading the abstract.

A perusal of the actual paper itself adds further evidence to the authors' claim. The study population of 1079 cases and 1259 controls would seem to be an exemplary sample size. The authors describe numerous confounding variables they controlled for, supportive evidence from other papers, and several potential (and testable) biological mechanisms whereby whacking off may also whack off some of the cancer risk.

Let's hear it for real science.  It would seem that things are looking promising for a Charge of the Onanist Brigade. Gentlemen, start your engines!

However, replication of results is the cornerstone of science, and is especially important before drawing any causal inferences, even from a reputable study like this one.

A bit more searching on PubMed turns up the following subsequent study: "Ejaculation frequency and subsequent risk of prostate cancer." This much larger study looked at 29,342 men, with 222,426 person-years (or wanker-years, if you will) of followup, and, alas, shows no relation between ejaculation frequency and the risk of prostate CA. I suppose that men will just have to look for other research that will support this kind of use of their therapeutic tools.

But, hope springs eternal that science will find and validate other reasons why we should all have a lot more sex a lot more often. Bettina Arndt of the Sydney Morning Herald has written a delightful article on this topic, and points out several other hypotheses we might as well test while we're at it, such as: "ogling breasts makes men live longer" and "regular sex helps to avoid colds".

Just for the record, I am generally opposed to all forms of quack medicine. However, if one must indulge, this kind is hard to beat.

Imaging Mount St. Helens

Since Mount St. Helens blew its top in 1980, the U.S. Forest Service and the U. S. Geological Survey have kept a pretty close eye on this particular piece of Washington State. As far as I know, they aren't using X-rays on the volcano, but the images they are producing are still quite wonderful.


The U. S. Forest Service mounted a real-time web cam in September, 2004, aimed right down the throat of the volcano. In June 2007, they added a High Definition webcam to the site, with significantly improved images.


In 1992, Johnathan Lees (now a professor of geolocial sciences at the University of North Carolina, Chapel Hill) obtained tomographic images of Mount St. Helens. Instead of X-rays, he used P-wave seismic tomography to produce 3D velocity maps of the magma system beneath the mountain. Another big imaging difference: pixel size. In medical imaging, an image pixel typically represents a piece of tissue less than a millimeter in width. Lees' pixels were much less wimpy -- the smallest of them represented a chunk of the Earth 0.5 kilometer in width! The image below shows a tomographic slice from a 27.5 x 21 x 20 km target volume. For scale, the above-the-ground profile of the volcano has been drawn in green at the top of the image.


Lees, J.M., The magma system of Mount St. Helens: Non-linear high resolution P-wave tomography, J. Volc. Geoth. Res., 53 (1-4), 103-116 1992.

Since the big boom in 1980, St. Helens has continued to erupt. In the past 3 years, the volcano has progressively added 100 million cubic meters to the lava dome in its crater. However, this latest eruption has been so gradual that normal movies of it would be like watching grass growing, only not as exciting. Instead, the USGS created a time-lapse movie of this eruption from data collected between October 2004 and July 2007. (via Ars Technica).


What's in store for Mount St. Helens next? Will someone figure out a way to use MR or PET on the mountain? Perhaps we can harness gravity waves or neutrinos to create tomograms of the whole darned planet. If so, I'd pay cash money to see that.

iPhone SDK -- Not Just for Radiologists

iPhone fans everywhere rejoiced as Apple today released its software developer kit (SDK) for the iPhone. Hopefully, third-party applications will become available with the release of the iPhone 2.0 update in late June 2008. The free beta iPhone SDK is available immediately.

MacWorld editor Jason Snell's liveblogged today's Apple iPhone Event. Apparently, a few weeks before today's release, Apple did a very cool thing: they

... called up a handful of companies, and asked them to send out a couple of engineers each, to see what they could accomplish in two weeks on an SDK they'd never seen before. In fact, most of them had never touched a Mac for development before.

A number of these nifty applications were then shown to the crowd. One of my faves:

10:41 PT: In two weeks, they wrote a game, Touch Fighter. OpenGL game, OpenAL for audio. Using accelerometer to fly the ship by moving the phone, and tap on the screen to fire.

I'm not big on gaming, but can imagine using this same technology to fly through a 3D CT scan like the one below, by banking my iPhone back and forth. I'll bet that the mighty radiologist-programmers at OsiriX are already licking their chops at the thought of this. IMHO, an iPhone-based medical image viewer would immediately launch a lot of iPhones into the hands of not just radiologists, but practicing physicians of all ilk.

10:56 PT: Next: Epocrates. "Every doctor knows about Epocrates." Now Glenn Keighley from Epocrates. Shows a drug lookup UI, so doctors can find a drug, tap to view information about the drug. They used SQLite to store their drug data, and used the iPhone's high-resolution screen to show drug images for the first time on any mobile platform.

This will be the killer iPhone app for my spouse, a clinician who uses Epocrates daily when writing prescriptions.

Just 3 and 1/2 more months -- Woot! Woot!

Driving While Under the Influence of fMRI

What will someone cram into an fMRI machine next? Earlier this week I posted about the use of functional magnetic resonance imaging (fMRI) to study brain activity in keyboard musicians while they were improvising jazz. It would now appear that even drivers with cell phones are not safe from fMRI researchers.

Brain Research just posted an in-press manuscript titled: "A decrease in brain activation associated with driving when listening to someone speak".

Why was this study done?

"Behavioral studies have shown that engaging in a secondary task, such as talking on a cellular telephone, disrupts driving performance. This study used functional magnetic resonance imaging (fMRI) to investigate the impact of concurrent auditory language comprehension on the brain activity associated with a simulated driving task."

Despite the title of the article, they did not study actual driving -- the sheer size of a decent fMRI scanner precludes this. To see the difficulty, imagine driving down the road with a your head stuffed inside a roll of toilet paper that is 7 feet in diameter. By the way, the toilet paper roll also weighs several tons, and has magnetic and radio frequency fields strong enough to send a cell phone flying and frying. Understandably, the investigators chose to simulate driving and cell phone conversations by other means.

"Participants steered a vehicle along a curving virtual road, either undisturbed or while listening to spoken sentences that they judged as true or false. "

How did the simulated drivers do?

"The dual task condition produced a significant deterioration in driving accuracy caused by the processing of the auditory sentences. At the same time, the parietal lobe activation associated with spatial processing in the undisturbed driving task decreased by 37% when participants concurrently listened to sentences."

The bottom line:

"The findings show that language comprehension performed concurrently with driving draws mental resources away from the driving and produces deterioration in driving performance, even when it does not require holding or dialing a phone."

This conclusion is not exactly a bolt from the blue -- they cite a number of other studies that echo the same conclusion: your driving sucks when you use a cell phone, hands-free or not.

For the full article, please see: Marcel Adam Just, Timothy A. Keller, Jacquelyn Cynkar. A decrease in brain activation associated with driving when listening to someone speak. Brain Research (2008), doi: 10.1016/j.brainres.2007.12.075.

(via MedPage Today)