Aa
Aa
A
A
A
Close
1453990 tn?1329231426

MRI and What Lesions Can Look Like

OK.  Lots of folks are asking me to look at MRIs that are saved as .JPG files in the photos section.  It is much more complicated than just looking at a picture.  DICOM images that come out of the scanners need to have their "window" and "level" manipulated in a DICOM viewer.  Think of this as the brightness (level) and contrast (window) on your TV set.  For you computer folks, Mamography and digital x-ray come put as 12 Bit images (4096 gray scale levels.)
MRI and CT come out at 16 bit (32768 levels of grey scale) As it turns out, human vision is limited to about 10-12 bits of grey scale and about 7-8 bits of brightness.  The viewer allows the window and level of the image to be loosened or tightened to change what shows up on an image.  Most of the time, the image you get on a disk is "centered" for window and level.  That's why images don't look like the images on the web or in the radiology atlases that are out there.  T2 lesions can be hard to find.  T1 Black holes can be almost impossible to find.  Both require windowing and leveling.

As far as what lesions can look like some examples follow.  These are all axial projections (slices running ear to ear across the nose - axial cross section ). A radiologist and neurologist need the axial plus slice locator, sagittal (vertical slices top of head to chin through the nose) and coronal slices (Top down across the from the front of the face to the back of the head) to really figure out where the lesions are.  Radiologists typically read these things on four 21" high resolution calibrated displays that are synchronized so that a highlight in one view is shown in the other two.  Most Neurologists outside hospitals do not have access to this level of technology.

The arrow in the following image points to a T1 "Black Hole" in the MRI of an MS patient:
http://www.medhelp.org/user_photos/show/169220?personal_page_id=1796003

The following show T2 Lesions
The small bright white areas in image A, the small bright white circles in the larger white-grey area are punctate lesions: (You may have to use Ctlr-+ to see them)
http://www.medhelp.org/user_photos/show/169221  ;

The following is a T1 image show the paramagnetic enhancement of Gadolinium in a T1 sequence.
The window and level have been changed so that this T1 sequence looks much "brighter" than the previous
T1 images.  The same sequence is used, but the window and level of the image can make this much of a difference:
http://www.medhelp.org/user_photos/show/169222

I hope this helps explain some of the nuances in looking at these images.  You really need to trust the Neuro-Radiologists and Neurologists when it comes to reading these things.  I would also keep in mind that most Neurologists are not looking at multiple views or sequences the way a Neuro-Radiologist does.   It is really looking for a needle in a haystack and for many people, they can begin to see pathology in the images that isn't really here.

If this geek stuff is too much, let me know.

Bob
13 Responses
Sort by: Helpful Oldest Newest
Avatar universal
This is the best explanation of MRI images that I have come across. You just made a lot more sense of things in 5 minutes than google has been able to reveal in hours worth of searching and reading.

Thank you for the explanations and examples. This 7T magnet is exciting. Although I think it will be a long time before they're used regularly. The cost is prohibitive. Wasn't the 5T recently deemed safe for use by the FDA?
Helpful - 0
1260255 tn?1288654564
Bob:

Thanks for helping to edjumacate us.

I think that a lot of the questions posted here arise for a variety of different reasons:
- Radiologist reports being too brief, leading people to try to figure things out by looking at their discs (mea culpa).
-Brain areas specified in reports that do not come up easily in web searches, so we neophytes have no idea where this part of the brain is, where to look on images and what symptoms could possibly be related to damage in that part of the brain. The brain areas we see discussed here on the forum are often different from the terms used in the radiologists' reports.
-Trying to understand how the different MRI imaging techniques correlate to one another, thus being able to compare slides from different series.

As LuLu said, we shouldn't even try to read our own MRIs, but when a radiologist's report is too brief, how can one resist the temptation?

Again, thanks for giving us the benefit of your expertise and helping us learn more. I'll have to check out the 7T images.

Audrey
Helpful - 0
1312898 tn?1314568133
Thank you Bob, really great info.  The 7T example is amazing, wish we had that available now.  Helps me understand why different tests are ordered, the MRI or CT etc.

Red
Helpful - 0
1453990 tn?1329231426
OK, one more image.  This is an axial slice off a Phillips 7 Tesla Research MRI.  The resolution on 7 T is so good, that if the imaging is correctly performed and then windowed and leveled correctly, they can see vascular changes (yellow arrows) that they feel are predictive for the formation of MS lesions.  Keep in mind, that this is research data and it has not been proven or disproven.  

http://www.medhelp.org/user_photos/show/169292?personal_page_id=1796008

Bob
Helpful - 0
1453990 tn?1329231426
By the way, I have non human test image off a 7 Tesla MRI.  They are trying to improve the radio receiver on this research MRI, because there is more than enough signal, but the receiver is still introducing noise into the signal data set.  Keep in mind, that these are not real images like and x-ray or a CT scan.  These are images generated inside a computer based on the radio signal strength of water and other molecules vibrating in a magnetic field or retuning to rest coming out of a state of magnetic excitation.  You can match a CT image to an X-ray image, but you can not match an X-ray image or CT image to and MRI image.  MRI is a "derived" image. CT, X-Ray, Mammography are "direct" images.

Bob
Helpful - 0
Avatar universal
All I was saying was, my curiosity got to me.  I do not believe I have been obsessing about my normal MRI.  But when you have so many symptoms, you try to find your own pieces to the puzzle, which in my case, the docs. are having a hard time with.

No one is looking to have ms,, but it would explain and put an end to the mysterious symptoms that all of us that are in limbo might have.
Helpful - 0
1453990 tn?1329231426
I never spoke about enhancement,  All digital DICOM images met be windowed and leveled.  This is not enhancement.  Windowing and leveling changes the histogram of the displayed image.  This is a standard technique needed to differentiate tissue from bone and lesions from normal tissue.  Thats why you have to use a DICOM image viewer to look at the images.

I understand that the field density and receiver signal to noise ratio on the 3.0 T units is much better.  Again, this has nothing to do with the window and leveling of the image histogram in the  viewer.

Enhancement has to do with edge detection, sharpening, colorization, and alteration of the diagnostic data.  The DICOM standard requires that if this is done to a medical image, it must display on the image that it has been altered and not a diagnostic quality image.  

Bob
Helpful - 0
338416 tn?1420045702
I think we're all guilty of trying to read our own MRIs...  What gets me aggravated are the people that are obsessing over their perfectly normal MRIs, even with the radiologist report in front of them.  

Bob, I can't remember, but I think you haven't had a 3T MRI yet, right?  The images from the 3T are much clearer and easier to see, even without enhancement.  I have some photos on my space, if you want to see.
Helpful - 0
Avatar universal
You must have been reading my mind!! For two days now, I have been looking up photos of MRI's of normal brain, and one with lesions, to compare what my photo looks like more... It is definately hard to tell and you have to know what your looking for,

I can see a big white glowing spot in the back part on my brain, and have no clue what to make of it being that my MRI's are considered normal.. My docs. never went over them with me, they viewed them privately and came back to say 'NOT MS".

I love to learn new things so anything you put on here, I try to soak it all in!!

Thanks for the great info,, and keep it coming!!
Pamela
Helpful - 0
1453990 tn?1329231426
Twopack
In those axial projections, it is hard to tell.  The brain is 4-D (X, Y, Z, time.)  CSF actually flow into and out if the verticals.  You need to see all three views and sometimes you need CINE-MRI.  Without having the source image and being able to window and level, it is hard to tell.
Most of the time, there is nothing "floating" there is something that is altering the resonance and the computer is generating a "fantom" to account for that change in resonance.  Sometimes even a very slight movement of the fluid in the ventricle will generate a motion artifact.

Lulu54
The comment about not reading your own (or anybody else's) is kind of why I posted this.  This is not like looking at a photograph.  It is more like building a 3 dimensional model in your head from a bunch of pictures where many of them are out of focus.

Bob
Helpful - 0
1045086 tn?1332126422
Great information Bob!  I love geek stuff even if I can't always follow completely and get easily overwhelmed with learned material as I'm so slow at processing these days.  Thanks for your continued willingness to share your talents.

What are those cocoon-like looking things that seem to float in the ventricles sometimes?

Mary
Helpful - 0
572651 tn?1530999357
Bob, thanks for taking the time to lay out some of this information.  As you now know, when we get wind of an expert we love to pick their brains for every bit of information we can glean.  

As the Co-cls here  will tell everyone, don't even try to read your own MRI's - those radiologists work at this for years to perfect the artform.

Doing so will drive us absolutely crazy with seeing things that are and aren't there.

be well, Lulu



Helpful - 0
1453990 tn?1329231426
By the way,  these are not my images!  Last set MRI had one 3mm T2 lesion with no enhancement.

Bob
Helpful - 0
Have an Answer?

You are reading content posted in the Multiple Sclerosis Community

Top Neurology Answerers
987762 tn?1671273328
Australia
5265383 tn?1669040108
ON
1756321 tn?1547095325
Queensland, Australia
1780921 tn?1499301793
Queen Creek, AZ
Learn About Top Answerers
Didn't find the answer you were looking for?
Ask a question
Popular Resources
Find out how beta-blocker eye drops show promising results for acute migraine relief.
In this special Missouri Medicine report, doctors examine advances in diagnosis and treatment of this devastating and costly neurodegenerative disease.
Here are 12 simple – and fun! – ways to boost your brainpower.
Discover some of the causes of dizziness and how to treat it.
Discover the common causes of headaches and how to treat headache pain.
Two of the largest studies on Alzheimer’s have yielded new clues about the disease