Linux Cluster Supercomputer Performs Surgery on Dog

An anonymous reader writes "In April, the Lonestar supercomputer, a Dell Linux Cluster with 5,840 processors at the Texas Advanced Computing Center in Austin, performed laser surgery on a dog in Houston without the intervention of a surgeon. The article describes the process: 'The treatment itself is broken into four stages: 1) Lonestar instructs the laser to heat the domain with a non-damaging calibration pulse; 2) the thermal MRI acquires baseline images of the heating and cooling of the patient's tissue for model calibration; 3) Lonestar inputs this patient-specific information and recomputes the optimal power profile for the rest of the treatments; and 4) surgery begins, with remote visualizations and evolving predictions continuing throughout the procedure.'"

Re:

[mazarin5 (309432)]

I can imagine the dreams I'll have tonight:

"I exist to cut flesh...

PC LOAD LETTER...
PC LOAD LETTER!"

"Nooo!"

New Robot Overlords

[backtick (2376)]

Wow, how about bowing down before a cluster of those? Heheh. Mixing the memes, sorry...

Re:

[moderatorrater (1095745)]

Actually, I believe they were just identifying the type of supercomputer that it was. There are a few types, one of which is to use linux to cluster a bunch of lesser computers until they're to the level of being a supercomputer. The only way that they could have put the same information into the statement while making "linux" less prominent would have been to use a more awkward phrasing, like making the statement passive (ie "surgery performed on dog by linux cluster supercomputer"). Initially, I would hav

The dog died.

[cduffy (652)]

...and they bury that very far down in TFA. The question, of course, is whether that was the planned outcome; I'd like to see it answered a little more explicitly.

If it is the intended outcome... well, so be it. If not, OTOH, that makes me a little less likely to sign up to be an early human test subject. :)

Re:The dog died.

[J'ai Friedpork (1293672)]

Second. I for one would like to know whether the dog died because of the treatment, in spite of it, or because they had to do an autopsy. Probably the latter, but the fact that they didn't specify it is a little worrying.

Re:The dog died.

[crackp1pe (1301669)]

Linux killed a dog? It must have been using ReiserFS, I hear it's a killer file system.

Bad analogy

[jonaskoelker (922170)]

Jokes about killer file systems are like cars with missing passenger seats.

Re:

[DaveV1.0 (203135)]

They state that the dog gave it's life for the test. It does not state that the dog died as a result of the surgery.

A more likely interpretation is that the surgery was completed and that the dog was killed and dissected to determine if the surgery was a success. That is what normally happens during animal trials.

Re:

[DrSkwid (118965)]

I doubt it was a volunteer.

Re:The dog died.

[MiniMike (234881)]

It was in a EULA printed on the back of a doggy treat: 'By eating this buscuit, you agree to be bound and dissected by the terms of this agreement...'

Re:

[morari (1080535)]

Only because we, as a species, don't have the balls to subject ourselves to stupid experiments. We force other living creatures to do it for use.

Re:

[by Anonymous Coward]

Go on mate, be the first. No? Then sit down and have your nice, big, warm glass of the ol' shut the fuck up.

Re:

[Phyrexicaid (1176935)]

And then they installed Linux on it.

Re:The dog died.

[RiotingPacifist (1228016)]

Did they go for yellow dog or puppy?

Re:

[by Anonymous Coward]

The Lonestar may not be able to successfully perform surgery, but I hear it's pretty good at jamming Dark Helmet's radar.

Re:

[192939495969798999 (58312)]

Yeah, it doesn't count to do computer surgery if the thing dies anyways.

Re:

[Tubal-Cain (1289912)]

Look on the bright side: The cancer died too.

Re:

[SUB7IME (604466)]

In medical research using animals, the animal is traditionally sacrificed for the purpose of accessing the tissue, seeing the anatomy, and gaining a more complete understanding of what actually happened during the experiment.

Re:

[cduffy (652)]

Makes good sense. It wouldn't hurt for an article aimed at people who aren't domain experts to be somewhat more explicit, though.

Re:

[SUB7IME (604466)]

Strong point, well taken.

Re:

[SUB7IME (604466)]

There really aren't that many steps in between. It's more like this:

1) Do the basic science
2) Ask your IRB for approval
3) Experiment on dogs
4) Ask your IRB for approval
5) Experiment on people
6) Use evidence to demonstrate that your procedure works
7) Procedure becomes mainstream

In case you were actually wondering, and not just making a roundabout point in opposition to animal research.

Re:

[Slashdot Suxxors (1207082)]

We can't all be as pessimistic as Anonymous Coward.

Re:

[Skrapion (955066)]

I'm not sure if you read further than the headline, but they do say it "went off without a hitch". It's all pretty vague, though.

Re:

[Skrapion (955066)]

The surgery was done with lasers, not exploding batteries.

And they killed the dog....

[by Anonymous Coward]

"Although the dog gave his life to the research, his sacrifice furthers science by allowing researchers to assess the success of the treatment and plan improvements."

Maybe next time the researchers should try it...damn vivisectionists.

Awesome

[timeOday (582209)]

For the big dollars that surgeons pull down, they are after all performing mostly rote procedures for the most part. When you can replace a decade of training a person with a simple file copy to load software on to a robot, think of the savings that represents. Health care costs are a big drag on our standard of living in all other areas and it's only getting worse. Not to mention the millions who die around the world because they simply cannot afford the procedures. I'm by no means saying this technology is ready or that I'd be willing to go under the robo-knife at this point, but I'm sure glad they're working on it.

Re:

[Xtravar (725372)]

We could get the savings now, without machines, if we lowered the bar to doing certain doctor-only actions and stopped artificially limiting the amount of doctors in the market.

In some cases, they already are by allowing doctor's assistants and nurses' assistants the same powers. But I won't really consider it a success until I can go down the street and have eye surgery in Boris's basement, right next to where he makes the bootleg vodka.

Re:Awesome

[Free the Cowards (1280296)]

How very appropriate, to have your sight both destroyed and restored by the same man's products!

Re:

[inKubus (199753)]

In Soviet Russia, sight destroys your bootleg vodka.

Re:Awesome

[HuguesT (84078)]

Of course replacing a surgeon with a reliable fully automated robot would be great.

However your description of surgery is not correct. Surgery is difficult, minutious and different for ever patient. Great surgeons must be able to plan ahead, direct a team and control all the details of a surgery procedure as it happens, as well as improvising with a cool head for hours on end if things go wrong.

It's the exact opposite of rote procedure. Especially now with recent advances in real-time non-invasive imaging and haptic instruments procedures change all the time.

Planning the surgery

[DrYak (748999)]

When you can replace a decade of training a person with a simple file copy to load software on to a robot, think of the savings that represents.

And who is going to plan the surgery ?
A doctor who has gone medical training is still required. The only thing is after a long intellectually preparation part (reflection, selecting the route, specifying the region, everything else that needs to be planned by someone with lots of experience), the doctor can give the instruction to the robot and move to the next case.

The price are going to go down. Not because you'll get rid of the doctors, but because the "planning" doctors we'll be able to handle more cas

Here the *-scopy procedure did it already.

[DrYak (748999)]

I failed to take that into account. Maybe because I live in a country which has already done massive effort to reduce recovery time, thank to less invasive surgery (nothing fancy requiring robots. Just simply using *-scopy instead of *-tomy procedure whenever possible).

Thus, for me, the introduction of robots, both classical surgical and new autonomous one, doesn't translate into shorter hospitalisation times. The procedure that our new overlords are going to replace where already minimally invasive in the

Yellow Dog Linux?

[davidwr (791652)]

If only it was a PowerPC cluster!

Yellow Dog Linux [terrasoftsolutions.com] distro home page.

Autodoc?

[AmiMoJo (196126)]

This sounds like one of the first steps in creating an autodoc from Larry Niven's books. Basically a box (coffin) you put someone in, close the lid and wait for it to fix them. It contains full life support, can perform surgery and produce (cloning?) it's own replacement parts.

Of more immediate use, this sort of thing could be very useful for situations where surgeons are not available. Ships at sea, trips to Mars, NHS hospitals with long waiting lists...

Re:

[awrowe (1110817)]

Do you reckon NASA would get upset at the astronaut who typed 'apt-get dist-update && apt-get dist-upgrade' from mars? "But dude, they finally fixed X in this one!"

Re:

[drinkypoo (153816)]

Actually, the first step was to develop techniques for imaging (parts of) the body (CAT, MRI, PET, X-Ray, whatever.)

It's not a Beowulf Cluster

[russlar (1122455)]

Does this qualify as a Beowoof Cluster?

Always mount a scratch monkey!

[waldo2020 (592242)]

Before testing or reconfiguring, always mount a scratch monkey

"Surgery" is a bit misleading

[ceoyoyo (59147)]

From the article it sounds like the real development is combining thermal MR and thermal ablation therapy with a computer to give you more confidence that you've properly heated the whole target. The computer doesn't do any cutting. It doesn't make the incision, or sew it back up.

If you want to consider what it does surgery then you really should include radiotherapy treatments that have been computer controlled for years.

So, how many

[Reservoir Penguin (611789)]

linux computers doe it take to KILL a dog?

The "New You"...

[Visual Echo (928267)]

The "New You" cosmetic surgery laser gone haywire in "Logan's Run" comes to mind. On the other hand, can I get one of these to melt snow on the road a few centimeters in front of my 70MPH tires? How many development cycles before 5,840 processors will fit in my glove box?

Re:

[MyLongNickName (822545)]

Please tag "linuxkillsdogs". The dog died. If this were a Microsoft product, the dog would have lived. You open source freaks are just evil.

Re:

[thatotherguy007 (1021257)]

Ah yes, blame the fact that they ran a program on an open source operating system. It looks to me like the software that ran the surgical operation wasn't open source at all. It seems it was all closed source software that was doing anything except running the computers. Although I am a bit lost; I don't seem to remember when Microsoft went into the medical field.

Free software is the right tool for the job.

[westbake (1275576)]

Prostate cancer is the target of the research, so your comment is closer to reality than you might like.

In this case, free software was the right tool. HPC with GNU/Linux is both flexible and mature. MD Anderson and everyone has better ways to spend their money than on software licenses for 5,000+ computers required to do this kind of work. Every kind of task will go this way eventually and most are already there. Whenever you start a task, you should look to see if some free program does not already d

Re:

[willyhill (965620)]

That does not mean it's the only tool for the job. FOSS is not perfect, just like commercial software isn't always the best solution. Like in most areas, the cost of software licenses is minuscule compared to salaries, facilities, materials, etc; software is a tool regardless of whether money is charged for it or not.

I'd rather have cancer cured than agendas furthered.

Re:

[sjames (1099)]

The sticker price is minuscule compared to salaries etc. but there are a lot of other costs to licensed software. Like having to keep track of license keys and certificates, having to deal with various license management dane bramage, having licensing prevent quickly getting a service back up quickly and efficiently, etc.

If you consider all the licensing problems and the extra hoops to be jumped through over the life of a server, the up-front cost of the license is really just the beginning.

The BIG adva

Re:

[Mozk (844858)]

Should have read Yellow Dog [wikipedia.org] operates on yellow dog [wikipedia.org].