New Algorithm Could Substantially Speed Up MRI Scans 115
An anonymous reader writes "In a paper to be published in the journal Magnetic Resonance in Medicine, researchers detail an algorithm they have developed to dramatically speed up the process of producing MRI scans. The algorithm uses information gained from the first contrast scan to help it produce the subsequent images. In this way, the scanner does not have to start from scratch each time it produces a different image from the raw data, but already has a basic outline to work from, considerably shortening the time it takes to acquire each later scan."
Patents etc. (Score:1)
Not to say the patent system isn't unfair/broken, but it's often been said that we shouldn't patent algorithms, and they should be OS etc. But how about if it's someone's livelihood, and years of research went into the algorithm, like I bet this did? Should programmers be given less due just because they're not working with real materials or making physical inventions?
Re: (Score:2, Insightful)
On first read, I thought you meant the patient system. That's also broken. Isn't the significance here the increased utilization of these expensive machines?
Seems like the GPGPU is one of the enabling technologies. If someone looked at this ten years ago, likely they would have pulled up short on the computational barrier.
Re: (Score:3)
Publishing in a Journal was the researchers choice. IIRC that pretty much makes the algorithm unpatentable as it's public disclosure (IINAL).
They have their reasons and motivations. Their sponsoring hardware vendor no doubt has a good jump on the competition.
In any case doubling (or better) the utility of MRI machines should have money flowing.
If the multiplier is good enough we might just sell a few clean used MRIs to countries with socialized medicine. Maybe their wait can come down.
Re: (Score:2)
Publishing in a Journal was the researchers choice. IIRC that pretty much makes the algorithm unpatentable as it's public disclosure (IINAL).
No it does not in the US. But it does start the clock ticking. You have 1 year after that.
In foreign countries however it's first to file, so once you publish it's over.
Re: (Score:2)
Re: (Score:2)
I suspect both the research itself and the education of the researchers involved were paid for by grants and fellowships. The patent will likely go to MIT, with some pay-off to the researchers. (I have no opinion on whether that's good or bad.)
Re: (Score:1)
Re: (Score:2)
If someone patented a way to make a web browser go faster, you would be comfortable with that?
Patents prevent people from being able to build on each others ideas. In a world without software patents, the companies that succeed are the ones that take advantage of everyone else's advancements the best. Patents gum it all up.
No, patents are supposed to be an exchange. In exchange for disclosing the invention to the public where people can build on the idea, you get a short time of exclusivity where the inventor can benefit from investing in making the product come to market either by doing it by her/himself or licensing the product to other more capable companies.
If not for patents, you'd have people hiding/obscuring thier advanced web browser (or other invention) as a trade secrets that need to be reverse engineered only by l
Re: (Score:2)
This sort of thing is exactly why algorithms SHOULDN'T be patentable; you can reduce them to basic lambda calculus. They're just logic and math. The very nature of a program is taking a problem and breaking it down
Re: (Score:2)
I mean, some equations can seem pretty creative, but where would science be if we allowed patents on, say, the Fourier Transform?
Credit for discovering that and all, but you can't lock it up and claim its yours now. If you didn't find it, eventually, unless humanity gives up on academia or dies off, some other mathematician would come to the same equation. They are merely representations of that w
Re: (Score:2)
In a sense though, what you say can apply to physical inventions too, as the same creativeness is required in that case - it's just you're working with physical materials rather than lines of code (I certainly believe all stuff is discovered, it's just that physical inventions probably have many more parts and ideas than most algorithms, and so could be considered an 'implementation'). Anyway, it wouldn't surprise me if there's something like the Fourier transform, but much more complicated and 'inventive'.
Re: (Score:2)
That's the thing yes. I've heard before how algorithms are just too modular (and build off other work) to be worth meriting a patent.
Surely someone here on Slashdot can give a counterexample?
At the least though, it seems prudent to at least protect their implementation by not making it open source (not in all cases of course, as some base libraries and functionality should be open, and that helps humanity, such as the PNG spec), but that's a separate issue I guess.
Re: (Score:1)
Call me when I can get my MRI images in an open format, readable on something else that Windows.
Re: (Score:1)
Re: (Score:1)
Interesting thank you... ...and yes - I've asked and been told that the stuff on the CD is closed format and only available in this 1 way - going to website of the manufacturer confirmed this, so your post is most certainly interesting!
Re: (Score:2)
As commented elsewhere, dicom is an open format. I used Osirix [osirix-viewer.com] on my Mac to view the MRI of my wife's spine.
Re: (Score:2)
I don't think there are many people (at least among those who think patents are valid in some way) who don't believe you should be able to patent algorithms. An algorithm is a process or method, the concept of which has been around since the patent system was created. Algorithms can be factory processes, chemical reactions, implemented in mechanical or electrical hardware, etc.
I think people's main protest is in implementing software that has no inherently novel algorithms - things that just seem like arr
Re: (Score:2)
It's like medical research, the actual manufacturing process might not be that difficult, but it's the research into getting the right "mix" of ingredients is the hard part.
Manufacturing some new "super-drug" might be a straightforward lab bench formula, but all that medical testing, studies, and other bureaucracy takes up all the dough.
One click purchase is more than just clicking the button. You have to bind the current web browser window to a user-id, use the very same user-id for billing, maintain a sec
Re: (Score:2)
One click purchase is more than just clicking the button. You have to bind the current web browser window to a user-id, use the very same user-id for billing, maintain a secure database of personal information, and be able to transfer those orders to a global connected network of warehouses as well as interfacing with the financial networks. There would be ways of guaranteeing fault-tolerance and reliability.
So they need to explain every step of the way and which techniques were used. And if someone comes
Re:Patents etc. (Score:5, Informative)
Not to say the patent system isn't unfair/broken, but it's often been said that we shouldn't patent algorithms, and they should be OS etc. But how about if it's someone's livelihood, and years of research went into the algorithm, like I bet this did? Should programmers be given less due just because they're not working with real materials or making physical inventions?
I've worked as a "programmer" (at least, in part) for 19/22 years in my career, I have over a dozen patents issued - I got a thousand dollar bonus each for maybe three of them, any employment contract I have ever signed immediately transfers ownership of any of "my" inventions to the company, including future royalties, etc. etc. I think my working conditions, with respect to IP ownership, are representative of >90% of programmers out there.
Of those rare programmers who might invent something patentable while not under contract to their employer to transfer ownership of the IP, probably >90% of them can't afford the time and expense of prosecuting a patent application.
Of those especially rare programmers who might successfully get a patent or two of their own, most of them could not afford to do anything about it if a corporation of any size infringed their patent, at best they might hope to sell their IP to a megacorp, but they wouldn't be in much of a position to negotiate its value. An exception to this is when working for ultra-small startups, but 19/22 of my years have been at something like 6 different startups varying from 6-25 people in size, any smaller would be very hard from an income security standpoint - the thing I have never done is sign-on at the moment of inception when company shares are being handed out like toilet paper (hint: you'd usually get better value out of toilet paper than shares in startups that green.)
Of those exceedingly rare programmers who might have the means to negotiate fair value for their own issued patents, most of them probably don't need the money anyway and have better things to do with their time.
Physicians seem to have just enough personal financial juice to get something out of the patent system, but even they are kind of in a commodity market - I've heard it said among investors "these medical device startups, if they have all their IP in order and something worth something to someone somewhere, they pretty much have a standard value of about $3M if you can find an exit buyer." The physicians and their friends tend to sink $500K to $1M into the company "building it up" to a point where it is interesting to the bigger players, getting their FDA clearances, etc. Some get lucky and get bought, many just fizzle out after the early round investors get tired of pumping money in.
Oh, for what it's worth, 80%+ of my patents have nothing to do with software / algorithms, and none of the ones that got me the bonuses did. Sadly, several of them are crap, mostly the ones I got the bonus money for. In the non-bonus environment, if an idea was crap, I'd have little enthusiasm for it, which would tend to turn off everyone else and it would die in development. However, in the bonus driven environment, you'd tend to "read" the rest of the team (including upper management who were rarely present at meetings), and if they liked it, hell yeah, I like it too, let's get this application done and get that bonus!
Re: (Score:2)
I've worked in software for around 10 years, and am a proud patent infringer on at least one patent. I didn't know about my infringement at the time, and by the time I did the software wasn't in use anymore, so it didn't matter. I say proud, because the "invention" is very, very widely used, and is quite obvious. So I'm quite proud to have infringed on this patent, thus proving how useless the patent system has become. No, I won't say what I infringed on, (patent trolls are quite real).
I've no doubt th
Re: (Score:2)
Oh yeah, I was "infringing" on the "use of XOR to display a graphic cursor" patent for years before I knew it existed... I'm pretty sure it's expired now.
Fifteen minutes (Score:3)
Re: (Score:1)
Re: (Score:2)
I wonder if this will ALSO drive prices down. But knowing scientists, and most importantly, *marketers*, this will just be used to develop new fancy scans to add to the current pile...
Reminds me of the early workunit system for SETI@home I participated in back in the days of 300Mhz PC's: as the broadband and turn of the century multimedia PC sale movement started exploding with monster speeds (and little to no modern-day spyware to slow the avg PC down), SETI made big changes in the ~2.0 client and started
Re: (Score:2)
in an electromagnetically charged TUBE is still unfun...
True, but the reduced time makes it less likely that there'll be 'motion blur'. This will result in better diagnoses on the average, and will also make it less likely that you'll have to endure a repeat experience "in an electromagnetically charged TUBE".
Re: (Score:2)
This should actually be a win-win, in that the price of the MRI can stay the same, but now they can process more cattle through the gate per day, so income goes up. Since it's a software change, it will only cost a few million to roll it out through the regulatory system, no real additional cost in hardware or facilities. We can hope that the troll (if any exists, it sounds like this was published in a Journal) would be fed from a portion of the increased income.
So... (Score:1)
Re: (Score:2)
Why wait? (Score:2)
TFA seems to imply that a patient spends more time in the tube because of the slow processing of the images.... Is there a reason they need to be in the tube while the images are developed? Maybe they need to retake images if they didn't come out? Barring that... wtf? Why not just do the scan and process the images offline on another piece of equipment? Storage is cheap nowadays, and caching the received RF data temporarily should be feasible.
Re:Why wait? (Score:5, Informative)
I work at an MRI research lab at UPenn and I can comment to this article (surprised it is on the front page BTW). During a normal clinical scan, the imaging protocol typically requires several different image types to be acquired (T1-weighted, T2-weighted, PD-weighted, T1rho maps, ASL, DTI, etc). Wikipedia can explain what those mean. So with a slew of different image types, the clinician can make a diagnosis based on different contrasts from each image type. The contrasts are based on the inherent T1, T2, T2*, T1rho, diffusion, etc. of the tissue. When creating the image, the scanner plays out a series of RF pulses (between 65-300 Hz depending on MRI field strength - look up Larmor Frequency). The image reconstruction essentially is a 2D or 3D Fourier Transform of a series of data points (which are actually electrical signals picked up by the MRI equipment). The raw data (k-Space in MRI terms or Fourier space everywhere else) consists of high-frequency and low frequency data. The high-frequency data ends up creating edge borders in the images and the low frequency generates the actual gray-scale contrast between tissues. Without reading the article, I assume that the researchers are copying the high-frequency data from one set to another thereby eliminating the need to generate edges. Without the edges, the images would be a big blur. The issue with this is if there is patient movement from one data set to the next it is difficult to accurately guess the borders (can potentially use navigators to correct). The more data you copy the more the contrast does not change from one set to the next so there is a trade-off from speed to accuracy.
As for doing the data reconstruction offline, 2D or 3D Fourier transforms take the computer like 1 second to do so it is not a reconstruction issue.
Sorry for the length.
Re: (Score:1)
Thank you for you insight
Re:Why wait? (Score:5, Interesting)
That's very true with MRI (my research is with MRI). We have our research participants spend 70 minutes in the MRI scanner. This is to capture 6 (7 with a localizer - a quick scan to set up the scanning area) different types of brain images. Our longest individual scan is just over 19 minutes. However, I have a project in the planning stages that will allow me to increase the resolution dramatically (from 1 mm^3 or 2 mm^3 to
This new algorithm seems like it could really help clinicians and some researchers. My type of scanning won't be helped by it for a long time though because I cannot sacrifice quality for speed. I need the sharpest, least distorted images possible. Clinically, that doesn't matter as much because radiologists are not quantifying parts of the brain in the same way that I do.
It's a fascinating technology though. Anything we can do to speed up scan acquisition and post-processing is very welcome.
Re: (Score:2)
It's a fascinating technology though. Anything we can do to speed up scan acquisition and post-processing is very welcome.
I suppose you have your own dedicated MRI... I did a bit of MRI safety testing and was always scrounging (typically for $500 per hour) time in various clinical systems when they weren't being used for patients.
Do you have to restrain your subject's skull in place for 19 minutes while taking your image?
Re: (Score:2)
Compressed Sensing (Score:2)
Hey dude,
Want to reduce scan times? Check out compressed sensing MRI [1]. You should be able to take way fewer scans than thought possible with 20th century math. Regularized reconstructions are the new hotness, but don't take the word of a Slashdot user who says "dude" and "new hotness"; read these fricking things.
[1] M. Lustig, D. Donoho, J. Santos, and J. Pauly, “Compressed sensing MRI,” IEEE Signal Processing Magazine, vol. 25, no. 2, pp. 72–82, 2008.
Re: (Score:2)
I was a little confused about the 15 minute claim, how long is spent "shimming" the magnet for a subject? I'm surprised that alone isn't 10 minutes (I know nothing of MRI, only NRM).
Re: (Score:1)
Re: (Score:2)
Thanks.
Re: (Score:2)
To shorten the AC's reply: MRIs are developed like a long exposure on film. The energy that the images are created from accumulates over time, that is the time the patient must spend in the bore. There are some calculations that can be done after the raw data is acquired (and the patient technically does not have to be in the bore for this), but the newer machines can do those calculations quickly compared to the time required to "ping" the hydrogen atoms throughout the area of interest in your body and l
Re: (Score:2)
It's not the processing of the images that is slow - it is the collection of the raw data. In an MRI scan, the machine performs performs a series of experiments with the following basic format; transmit a high-power RF pulse, distort the magnetic field in a complex manner, and then capture the "echo" of the RF pulse. Each "echo" contributes some information about the image. In the most basic form, each "echo" can contribute an information equivalent to one row of pixels.
So if you wish to acquire a 256x256 i
Technology and medical costs (Score:2)
There are too many people completely misunderstanding that innovation and inventions and new technology drives costs down.
MRI shortens the time it takes to understand what is wrong with a patient, the precision of understanding goes very high, there is very little need (if any) for exploratory surgery with good imaging. Yet people (and on THIS site!) argue that technology is almost supposed to push medical costs higher. This is absolute nonsense.
When this new algorithm makes it into the MRI machines, do you
Re: (Score:2)
MRI shortens the time it takes to understand what is wrong with a patient
That is only true if an MRI is available nearby.
there is very little need (if any) for exploratory surgery with good imaging
An MRI will never replace a biopsy. Never. It is good at seeing structure but it can't tell you what is going on at the cellular level.
Yet people (and on THIS site!) argue that technology is almost supposed to push medical costs higher. This is absolute nonsense.
Everything in this country increases medical costs. No matter what happens - even if nothing at all happens - costs go up because the system is built around the insurance companies. And the insurance companies are constructed to deny care to their customers. We couldn't have des
Re: (Score:2)
That said, it is highly unlikely you would have the power and cooling (just to name a few requirements) in your garage to build a better MRI.
For that matter most people would probably dim the lights in their neighborhood if they brought an MRI to full power in their garage.
MRI is inherently scalable -- that is, you can develop imaging algorithms on an imager for chipmunks then scale up. I have seen essentially desktop-sized NMR imagers used for studies of small archeozoological artifacts, for veterinary imaging of small animals, an for research on NMR/MRI algorithms. So if I were to start doing something like that, I would not be encumbered by anything you mentioned. With groundwork laid, I'd look for VC or buyout to scale it up.
Re: (Score:2)
I usually ignore your free-market trolling, but in this case I have some actual knowledge about how biomedical research works. Also, one of the authors already has the full paper available for download on his website [mit.edu]. Guess who paid for it?
Still think it's great technology? (Yeah, I know - the
Re: (Score:2)
I didn't RTFM at first and I still didn't now. I don't care who and how this came out to be, that was not the center of my point at all, did you even read it?
My point has nothing to do with who invented THIS particular thing or ANY thing, my point is that government money, influence, power, regulations, laws, taxes, subsidies, protections, bail outs, stimulus and GUNS to our heads make it impossible for normal free market to work and drive costs down via competition.
I don't care WHO came out with what, in f
Re: (Score:2)
The trick is that NO true Scotsman (oops, free market) will ever be found, so it can never be wrong.
Re: (Score:2)
If you have these ideas of profiting from providing people with better technology - learn Mandarin.
Okay, now I'm double-posting, but I just couldn't resist: do you really think that China is more of a free market than the United States? Do you really think their government just stays out of the way and never micromanages anything? China may be notorious for letting employers treat workers like cattle and generate massive amounts of industrial pollution, but that's not the same as having a laissez-faire ec
Re: (Score:2)
"Yet people (and on THIS site!) argue that technology is almost supposed to push medical costs higher. This is absolute nonsense."
Nonsense or not, this counterintuitive result it is backed by empirical evidence in the USA.
"When this new algorithm makes it into the MRI machines, do you suppose the costs of using the machines will go or down."
When you get a MRI how much money pays for
(a) the MRI machine
(b) the R&D for the MRI machine
(c) the hospital or facility which owns it
(d) the radiologist
(c) + (d)
Re: (Score:1)
Re: (Score:1)
the pharma has the consumer by the balls because of FDA, because of Medicare, because of special gov't privileges.
Re: (Score:2)
the pharma has the consumer by the balls because of FDA, because of Medicare, because of special gov't privileges.
And once they're all gone, what then?
"Pharma" will still have the consumer by the balls, because it would be terrible if little Timmy died since you refused to shell out a million dollars to prolong his life another three miserable cancer-ridden months, you monster.
Re: (Score:2)
No, if all the costs added by government and all the license fees, patents, copyrights and costs of passing FDA are gone, then the barrier to entry is insignificant. Anybody can come up with new medical instruments and drugs actually, the technology now made it much cheaper - from DNA sequencers to computer tech and tools, to allow anybody basically to come up with new stuff in their garages.
The reason that there is little development in many medical areas today is not that there is no market and no way to
Re: (Score:2)
Tell you what, Roman, can you post a video of your trying out an exciting drug that's been developed in an area of the world that's refreshingly unencumbered by such silly nonsense as safety regulation and clinical trials? Do post it on YouTube, there'd be lots of people who'd love to watch.
Re: (Score:2)
Tell you what, I used a number of drugs that are not FDA approved but are used legally in other parts of the world. Is that good enough for you?
Re: (Score:2)
Tell you what, no. For the absolutely fucking obvious point that the FDA is hardly the world's only government regulator of drug quality.
I meant, and I can't believe you're so fucking dim or deliberately obtuse that I'm having to spell this out, that a decent test of your commitment to your ideals would be if you were to consume a drug that purports to treat a serious condition -- diabetes, dementia, CHF, asthma -- that was made by an unregulated person or entity.
Not popping a tab of fucking MDMA or a medic
Re: (Score:1)
Yes, a large number of people DO consume drugs that are literally created in garages, and it would be much more appropriate to legalize all of this activity, to stop the Drug War in the process, to pardon all of the non-violent drug related prisoners.
The other parts of the world that I am talking about are (in no order) Europe, Russia, China. There are different regulations of-course even in those places, however nothing compares to FDA in cost of doing business with the government.
Re: (Score:2)
You really know absolutely fuck all about pharma development, don't you? The costs of the regulatory burden are effectively equivalent in Europe and the US. Pharmacos are no more pleased with NICE and the MCA et al than they are with the FDA. The costs in Russia and China are lower, but practically no drugs are developed in either country, despite the pharmacos all being giant transnationals with absolutely zero sense of loyalty to a particular country. Care to know why? Because the regulatory bar isn't hig
I'll be more impressed... (Score:3)
Granted, it beats being irradiated, but if I could change anything about it I would make it quieter. Hell I'd tolerate it taking longer if it was quieter; they are kinda cozy and I could take a nap in the scanner if it wasn't so damned loud.
Re: (Score:1)
The sounds come from the Lorentz force shaking the gradients during the RF pulses. There is a lot of research to minimize gradient slew rates to minimize the vibrations (make things quieter). Make sure you get ear-plugs when scanned.
Re: (Score:2)
If I were to attack this, I'd think of installing gradient coils on some sort of electrostrictive material and issue pulses to both to provide opposite loads on the coil structure.
Re: (Score:2)
Re: (Score:1)
Re: (Score:2)
With ear plugs and noise canceling headphones with music on it is quite quiet in there. Some modern scanners actually have noise cancelling speakers in the bore but it gets hard to fit a lot of stuff in there safely without impeding your scans or space. There is a lot of research being done in the commercial space to make them more aesthetic but in the end, you're dealing with quite a strong force in there, it's hard to counteract nature.
Re: (Score:2)
I'm sorry, you're using headphones, with magnets in them, in an MRI?
http://bitsandpieces1.blogspot.com/2005/08/watch-out-for-that-flying-chair.html [blogspot.com]
Re: (Score:1)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
There are off course, those MRI-safe headphones. Ours has a full (non-ferromagnetic) metallic unit (aluminum) inside the scanner room linked with fiber optics. There is a surprising amount of metal that can be in an MRI suite but it has to be specifically designed for it. However everything (cables, solder, chips) have to have specific properties.
What about fMRI? (Score:1)
Re: (Score:2)
You can't get (useful) temporal information that is faster than the phenomenon you are trying to measure. For instance: turn on an oven and measure the temperature. Sure, you can read a temperature sensor at almos
Uh... Caching? (Score:3)
Since when is caching a "new" algorithm? From the basic info in the summary it appears this is just a cache of some initial processing that is then reused for subsequent image generation passes...
Re: (Score:2)
They're using a new algorithm [merriam-webster.com] to capture the MRI data. Why is that unclear? They didn't claim anything like "invented a new basic computer science algorithm".
Re: (Score:2)
I didn't read the article... so I don't know, but the summary doesn't say anything about capturing the MRI data, the data is the data... They are using a new process to generate images from said data.
Now, in the summary, it says they do some up front processing, and then save the result of that, and use it for all subsequent image generation activities... because this processing used to be done on every image generation pass... well guess what, thats the definition of caching, and it has 0 to do with a new
Re: (Score:2)
they are just storing the part that gets repeated over and over, and using it multiple times...
Right, so they can reduce their data acquisition time - no need to re-capture the same data.
Sort of like a spiral CT, I imagine, in general concept.
It doesn't matter (Score:2)
In the end, Dr. House will still barge in, pop a few Vicodin and flip the patient over, revealing the infected tattoo on his ass* (the one his buddies got for him after a night of heavy drinking) that is the cause of all the peculiar symptoms.
* Rabbit tracks down one cheek into the gully and Elmer Fudd with shotgun on the other saying, "Come out of there you Qwazy Wabbit!"
Re: (Score:2)
LOL. This post is priceless.
Re: (Score:2)
original paper (Score:1)
start with modernizing the hardware (Score:2)
Slow data a aquisition (Score:2)
Still, although they are old, they are adequate because the speed at which the data is coming out of the scanner is "pain-in-the-ass" slow.
Scanning is very slow. Emit the radio impulse, wait a little bit, read the signal. Repeat this (with varying magnetic fields) for the number of data point you need. So the current bottle-neck is not the ageing computer, but the data acquisition it self.
The current trend is to find a compromise between the number of data point.that has to be acquired and the definition an
Similarly impressive Technology (Score:1)
While I can not comment on the content or merit of this work since I have not read the paper I can say that there is a lot of research going on in the area of MRI and a number of exciting things are happening. I work in a research group that works on impressive MRI techniques combined with algorithms that allow taking entire MRI movies (not just photographs). This is extremely interesting when looking at the beating human heart, speaking humans or moving joints. We are really just beginning to explore the p
MRI for kids? (Score:2)
Does anyone have experiences with (brain) MRI's for kids (10 and under)? In order to 'keep them quiet' enough for the scan (and scary noise!), is a full-blown anaesthesia necessary? What options do exist?
False Positive Imaging (Score:1)