I've been trying to understand this for YEARS. YEARS!! This is the first time I actually get it. I knew about the FFT, but projecting lines and how they manage that finally makes sense. Incredible explanation.
@@johnnunn8688 lmao I'm very aware of that! A PhD usually includes a major research component where you investigate some hypothesis or theory (usually experimentally) and present your findings and justify them in a Viva. If you pass that, you get your qualification!
I’m board certified in cardiac MRI and spent 4 years doing research in the field and this is honestly the best explanation I’ve ever seen. Also the best looking, loved the Fourier animation. More medical tech, fewer planes! 😂
This is the definition of exceptional science communication. I have a background in engineering (and overall very mechanical curious) and have spent hours trying to understand how MRI’s work after getting a few done on my knee, none of this research can come close to the elegance of explanation shown in this video. Bravo.
The detailed explanations and depth of knowledge in this video, the M1 Abrams video, and countless others consistently blow my mind. Thank you for all your hard work researching and delivering such high quality videos!
So excited you made this video! I'm a Diagnostic Radiologist who teaches Radiology to medical students. This is as deep an understanding of MRI physics as almost all of my students need. (Less than 10% of my students will have to pass the Radiologic Physics boards that I took to become a Radiologist.) FWIW, we do try to use "surface receiving coils" as often as possible for better images. Seriously, every one of my medical students will be expected to watch this video. Thanks, Brian and team!!!
Do you see the new generation of large AI models having a large impact in diagnostic radiology? I feel like that has been "just around the corner" for the last 20 years.
@@csours I told folks 20 years ago that I would interpret images for 10 years and then over read computers for 10 years and then find something else to do. As it turns out, I haven't started over reading the machines yet, so I'm no longer sure what the timeline is. But it is always just around the corner. We (human radiologists) are too expensive to not replace as soon as we can be replaced.
@@texastaterbug5395 It's also a regulatory issue. AI isn't really allowed to make a diagnostic reading (but it will get use as a first reader to indicate clear cut cases where only one additional radiologist needs to have a quick look to confirm...thus lightening the workload on radiologists and allowing them to spend more time on the 'interesting' cases)
@@peterzerfass4609 It's just a tool. On one hand, AI assisted analysis fits the work flow and I can see it making everything faster. On the other, that could lead to needing fewer radiologists overall. Everybody needs to take a step back and remember that all the niche AI applications popping up are only tools, and nobody should feel threatened to learn how to use a new tool.
I have a question left that I don't think the video has answered and that I have been struggling to find an answer to elsewhere as well: I am familiar with the fundamentals of the technique from the area of chemistry using NMRs to help determine the structure of molecules. From this I know that the resonant frequency of hyrdogen nuclei is not as directly connected to the strength of the outer magnetic field as the video here implies, but that it is in fact dependant on the actual net magnetic field that the nucleus experiences and as such is influenced by the surrounding electron density and thus the chemical structure of the molecule it is part of (this is precisely what NMR-techology uses for its analysis). So my question now is: If we already use the different resonance frequencies of hydrogen to select the "slice" of the patient we want to look at, how do we not get a mixed image from hydrogens in different chemical compounds and bonding situations having the same resonance frequency in slightly different slices? Are the differences in resonance frequency created by the gradient in the magnetic field just much larger than the differences through different chemical bonding situations (and does that mean this limits how thin the slices can be) or how are those problems avoided? A very sincere thank you, if you have taken the time to read and consider my comment and question!
@@nitramreniar question was not for me but I may have an answer (though admittedly it's been a while since I worked with MRI machines). There is a so called chemical shift, it presents itself as the image shifting a bit, leaving black borders around an organ for instance. It's usually very small though and iirc can be corrected somewhat using some sequences such as fat suppressing ones.
@@nitramreniar The effect given in the video is Larmour precession, this is caused by electron spin not nuclear spin. It is a similar phenomena but there are different particles involved. NMR is due to nuclear spin not electron spin.
I am a board certified radiologist and involved in MRI science. And this video is an educational piece of art! You and your team is incredible! Congrats!
8:40 there is also something called "quench", which is basically if you need to turn off the MRI machine (if, for example, an accident happened and a metal oxygen tank was launched across the room towards the machine) there is a big red button which increases the inside temperature of helium so that it removes superconductivity, which causes helium to instantly boil from liquid to gas. That's why there is a quench pipe, which releases the helium gas outside in a huge cloud of white gas. However, if for some reason something breaks and a helium leaks inside the room, aka the quench happens inside the MRI room, it can be extremely dangerous because people inside the room can suffocate, which is why MRI room doors are (or should be) designed to open towards the outside, because if a bunch of helium releases inside the room, it's going to put pressure onto the door, you won't be able to open them and you could die. And in case of a quench, you will lose helium, and need to refill the MRI machine, which is extremely expensive.
Even better - depending on the cryonics in the cryostat, you have multiple different hazards from a blocked quench exhaust: First, liquid helium. This is lighter than air so you need to drop to the floor. As you mentioned, this can also cause the door to become jammed which can lead to suffocation. Second, liquid nitrogen. This is heavier than air, so dropping to the floor will actually kill you. In this case, you actually need to walk normally/keep your head up. This is very rare outside of research installations. Third, liquid helium and liquid nitrogen - worst of both worlds and uncommon but possible. In this case you have to keep your head in the middle of the room - too high and you suffocate, too low and you suffocate. The training I received is to crouch walk out of the room and to always block the door open with a doorstop or other mechanism so long as you’re not doing a scan. None of this includes the frostbite risk if you’re standing next to the machine.
@@ScienceReasonLove helium stratification typically requires a calm environment, an MRI quenching is violent, fast, and the expansion ratio from liquid to gas is high, dropping to the floor is likely going to waist the 1 or 2 seconds you have to get as far away from the thing as possible.
Most MRIs nowadays have gas containment shells that envelop the entire machine and route into the exhaust conduit so that if there is a helium escape it will go up the exhaust anyway. Machines with shells look far more bulky than machines without and often look as though they are built into the room.
My father helped establish some of the foundational physics and engineering that lead to MRI’s back in the 80s. When I was born in the 90s, I was diagnosed with a rare genetic condition, that resulted in me needing several MRIs every year, indefinitely. I definitely think the enthusiasm he showed for these machines, along with his willingness to explain how they work, really helped mitigate the fear of going inside them, especially when I was a child.
I am neither a medical nor a medical engineer or an engineering student. I study pure mathematics. A weird one too, since I'm not fond of physics as well even though I am studying the proofs behind their equations. But seeing how elegant equations like the Fourier transform you mentioned are, brings nothing but a big smile on my face. During Fourier's time, there was no talk about electromagnetic waves and such. Early mathematicians only had Sin and Cosin waves, the relations between them, and a blackboard and chalk. To think that their tinkering with such trivial sketches would lead to constructing such wonderful pieces of technology is actually mind-blowing. This further proves that the very core of everything in this universe is tied to mathematics.
Agreed. The Fourier analysis is the key. Without the power of the transform these machines would be useful only for inflating balloons and pranking nurses.
It’s crazy how often a Fourier transform is credited with enabling cutting edge technology. It’s even crazy that most people don’t know who Fourier is despite how influential his work it.
The Fourier Transform is seriously one of the most "beautiful" concepts, just so damn elegant. It's a completely non-obvious solution when faced with the problem, and then "why didn't I think of that" when you know how it works. The applications people have found for it are equally amazing. JPEG compression blew my mind when I learned how that worked.
I was thinking the same thing! When they brought up hydrogen spin and processing the decay, I was just waiting for the mention of the Fourier Transform.
I used to consult for the Toshiba MRI division. The building where they tested the magnets had a huge open area with yellow warning stripes painted all over the floor marking the "keep clear" area around the magnet under test. Turns out the yellow stripes were added after a screwdriver was left out and became a projectile during a magnet test. It punched a hole in the magnet assembly which is not a good thing when you're dealing with that much power.
Crazy news story from this year (2023): “Leandro Mathias de Novaes was taking his mother for a scan at the Laboratorio Cura in Sao Paulo in January and entered the facility with a concealed handgun. Despite warnings from staff to remove jewellery and metal objects in the MRI room, Mr Mathias [secretly] kept his weapon on his waistband. The magnetic field from the machine was so strong that the weapon was pulled from his waist and fired off a round, which hit him in his stomach.” (The Telegraph)
Man I cannot get over how well you explain such complex marvels of technology. Thank you for these videos. They are wildly entertaining and informative.
As a former mri technologist, I am happy with this very clear explanation without oversymplifying. Brings back memories about understanding the process of 'fitting the k-space' with all the signals measured and generated by the measuring gradients after the initial excitation pulses (slice selection and x/y matrix) echoes readout. FFT was genious. And nowadays mri relies far more on ultrafast 3D volume scans with very little gradient angles instead of the 'simple' 180 degree spinecho sequences. And the dreaded mri sequence acronyms between the manufacturers! Thank you for rhis video.
I'm an MRI engineer for a company featured prominently in this video and this is genuinely one of the best summaries of the technology I've seen so far. If you're still baffled after watching this video that's perfectly normal. This is stuff that's typically taught over the course of weeks or months not minutes. In spite of the short run time this video is densely packed with useful information. Well done.
One question ive had a while you might be able to answer then is is there atleast a safety system or spool up time on these things if theres anything magnetic detected? Say one happens to have a peice of mettal in the body like the appendex without knowing? The thourght alones terrifying.
@@JMAssainatorz the device itself does not have that safety because when one is in the strong field the bad thing can happen. like people with heart pace makers or war veterans having shrapnel in critical locations like brain. it is the staff responsibility to make sure there is no such thing. and imaging with metallic pieces in body will affect the image due to their different magnetic properties.
@@JMAssainatorz No. The magnet is always on. It could quench however if the field is disturbed too much and lose superconductivity, then it will no longer be magnetised.
I am a researcher working in the field of MRI, and this is the most informative and beautiful exposition of the topic I've ever seen in a video form. Your visualization of 2D gradient encoding (14:45) and your straightforward explanation really drives home this core concept which is otherwise quite hard to grasp. Thanks for creating this great resource!
Hi Chinmay, i did MRI 4 years back for Scitica nerve pain.normal report.but i always feel like one nerve is contracted in hip due to which my leg and hand become senseless. Shall i go for one more MRI ? Plz reply
Can you please answer some of my questions? Due to my condition I had MRI every year and every time they said. " If you feel sick press this balloon but we would have to restart whole imaging" Fist: What in that system could make me sick? Second: Why restart? My brain would be same even after some minutes,no?
@@malikovajana about the second question ,no our brains won't be the same because when we talk or think or be in a high focusing mod according to doctors it's not the same as when we relax that's why maybe sometimes they talk with patients during MRI Hope i answered ur question ..
MRI Physicist here, fantastic video, I have to recommend this to some beginners. I didn't notice any errors either, just two small inaccuracies: 1. The spins align themselves along OR against the magnetic field B0, but a tiny majority (~0.001%) align along B0, creating the net magnetization. 2. Some of the footage you used were CT images... but whatever. I had a blast and was surprised when the video ended suddenly. Great job!
Thanks for your comment, he also described x rays as "radio frequencies" should have said electromagnetic ... What do you think is the phisical limit for the resolution of these machines?
@@Valery0p5 They are radio frequencies, they're literally called 'RF pulses'. The resolution is based on the strength of the gradients, so making the gradients more powerful you get smaller voxels. But when you decrease the voxel size you get less signal, which makes the image dark. So you need a more powerful magnet. But the more you increase the magnet strength, the more problems you have with SAR (special absorption rate), or how much you heat up your patient. MRI is basically a low power microwave, something we do NOT tell our patients. :D That is why there are strict SAR limits, which usually only come to play at higher field strenghts. 7T's biggest headache by far is the constant struggle with SAR. As a side note, MRI is, in my opinion, the most complex imaging method there is. Everything incluences everything, and increasing some parameter has direct concequences and tradeoffs with all other parameters. For example, every point in k-space influences every pixel in the image, so if you have a white pixel in k-space, the whole image has a diagonal stripe pattern on top of it.
Correction, spin is referred to the electron's spin, atom core (with neutrons) doesn't contain any spins i.e. electric orientation with underlying quantum field. Magnetism comes from electrons spin either -1/2 or +1/2, also this spin is not a literal spin but rather how energy of electon interacts with underlying electric field. Think of electron as oscillations in electric field that exists in all space with some potential energy. Electrons are disruptions in those fields, same as protons and other matter oscilating at different frequencies. These values come from those fields' energy potential and oscillations will determine which particle it is. P.S. Before anyone corrects, protons do have spins but due to their mass, their magnatic moment is negligible compared to electrons ( i.e. ~ 2000x stronger). Mass is inversly proportional to the magnatic moment of the particle.
I've had more MRIs than I can count since I had a (luckily benign) brain tumor removed in 2015 that was in the middle of my brain. This was so well explained and I found it very interesting as a patient to see and understand what goes on inside the machine. After having so many, I've gotten so used to the clicks and crunks and buzzes that I can usually fall asleep in them now. 😆 I'll need check-up MRIs for the rest of my life, so I'll definitely think of this when I get my next check-up! Thanks!
@@HarperLloyd It's kind of hard to explain, but I would get a "whooshing" or pulsating feeling in my head every time I went from standing up to lying down, sitting to standing, etc. At first I thought I was getting up too fast, but then they started to happen when I was just sitting doing nothing and THEN they started getting very painful, too! Incredibly, I had no other symptoms. I kind of scoffed at the idea of needing an MRI because I thought it was overkill. Good thing I listened to my doctor! They thought it was a "cavernous malformation" until they got it out and biopsied it, discovering it was a benign tumor. I just got a check-up MRI a few months ago and all clear!👍
I had a benign tumor removed in 2019, it was 2.05" x 1.85" meningioma near my cerebellum. I had headaches since 13 years old, and leading up to my first MRI that found the tumor, I had migraine headaches, pressure in my head, whooshing sounds accompanied by pulsating pain when I stood up, and felt like I was freezing cold all the time. The pulsating pain in my head was enough to vomit and pass out several times, if I bent over and back up too rapidly, stood up to quickly, or stood up from laying down too quickly. I had ocular migraines for 3 years prior to surgery. 4 years on and 9 MRI's later since surgical removal, all clear. The sound of the machine is like an orchestra to me. I imagine percussion, reeds and strings.
I work with medical imaging devices and the MRI is by far the trickiest and most complicated machine to understand. The engineering, creativity and knowledge needed to have come up with this method of generating images is something that blows me away every time I think about it. And now it is a little bit easier to understand thanks to this video, amazing job. And beautiful animations, btw.
Man I am studying medical engineering at university and had a module on medical imaging this term and MRI was the part that I couldn't wrap my head around. This video made it really easy to understand whilst still keeping high detail. Massive kudos to you again!
I have studied physics. We had to make 9 experimental projects per semester. One of them was magnetic resonance. While the other projects gave you materials to study at home and then you were collecting data for 2-3 hours, this one was different. There was the materials to read at home and when I arrived there was another 3 hour lesson at the end of which I started to understand (then pressed a couple of buttons and the measurement was done. This all was just to get the chemical composition of the sample, not the image. I had absolutely no idea how the Imaging part works!
I am just a medical comp sci student but I did have two courses related to MRI. As I recall the image is achieved by adding another localizing magnetic field. This allows for every point in space to have a slightly different signal. As you know the expected signal this can be used to associate the different signals with different points and then generate the image. But take that with a grain of salt. My course was mostly focused on what you do with the image not how it is generated. EDIT: Nevermind this is actually explained much better in the video.
As a radiology resident, I studied so much text to understand this, and you have explained the concepts so clearly. So much respect to the engineers and their contributions who made it possible to save hundreds of lives just in our setup every year.
We are required to understand the radiology physics . If your dad is a radiologist, i am sure he understands the physics of the machines, CT, MRI , USG and xray.
@@radiologydaily I really doubt that you know how to build or design one of these machines from scratch, if you can't do that then you don't really understand the machine you would need a bunch of phds from a bunch of engineering and physics fields just to work on each tiny aspect and the breadth and detail of knowledge is not something that you are going to get from a 20 minute video, It would take you probably an 1 year just to master all the proofs for the fourier transform and understand it well enough to apply it to the design of a mri machine
@@xblade11230 radiologists just need to understand how to analyze these images. Knowing how they work assists in that. Don't think they need to know how to build a MRI machine tho
I’m an Electrical Engineer who heavily focused on NMR research during grad school. Absolutely fantastic explanation and animations. I’ll be sending this video to my professor!
I used to work at a factory that made MRI's, I was welding on the main pressure vessels that held the liquid helium. I have a small chunk of that Nb/Ti wire. The actual superconductor is Tiny, and is encased in copper to help protect it and let it be formed.
As someone that's recently spent a ton of time inside a few of these machines recently, this was a really informative video. I'd understood the basics of these machines ever since a high school physics course some thirty years ago, but your presentation here really made it click for me. Thank you for making this video, and thank you to the people that worked together to invent these machines.
@Ed No its not, I've had dozens of MRIs over the years, they're not harmful at all. The safety and non-invasive nature of them is one of the many reasons they're so great!
@@Ed-rt9qt It can't because your brain makes memories which are stored by links between neurons. The magnetic field does not disrupt neurochemical processes and does not damage the neural links and therefore has no impact on memory. CT scans on the other hand emit ionizing radiation that can disrupt neurochemical processes and could damage neurons and proteins throughout the body with enough exposure. Over time this could amount to real tangible damage to organs including the brain if that organ is being imaged a lot. The likelihood of any X-ray or CT causing damage to a sensitive organ like the brain is effectively zero. The doses of radiation are very small dispensed over short time periods, it's not like you're standing next to a pool of molten Uranium 235. That said it's more likely you'd experience some minor damage from X-rays than Magnetic Resonance Fields. Not like the field is strong enough to bind your atoms to a physical position in space... that would take a few hundred more Tesla to achieve and that could be dangerous. 66,000 Tesla field could hold a person still if they weren't already moving while a fluctuating 1,800,000 Tesla field could start ripping a person apart under the right conditions (frequency considerations being managed). ...Just trust me bro. Seriously there's no known limit to what a human can tolerate but in fairly high static and dynamic magnetic fields people have reported strange symptoms inside high Tesla fields usually above 3 Tesla (which is as the video states 3 Tesla is generally the upper limit of most MRI machines used on people in a diagnostic capacity). Most people won't ever experience an issue from any MR scan.
@@VariantAEC Well, I think it can be dangerous because this machine makes hydrogen atoms oscillate, and if atoms oscillate then it means that they have energy.This is similar to a microwave oven, where water molecules oscillate and this is transferred into heat.Because heat is oscillations of atoms or molecules, so the oscillations of hydrogen atoms must also produce some heat which must be dissipated somehow.And maybe it can even rip the molecules apart.
I have had several MRI tests and also memory testing and my memory is above average. The longest that I have been In a MRI was 45 minutes and I am so used to them that sometimes I fall asleep because I'm bored.
I love this channel. Makes me realize that engineers are true aliens to regular people. The amount of knowledge, time and collaboration to make something like this is astounding and alien to me
I am a medical student and this is absolutely the most eloquent explanation of how MRIs work. I never grasped T1 vs T2 imaging but this actually explains the reason why they are different. Excellent video as always!
Currently studying mechanical engineering, I simply cannot fathom how our cousins in electrical engineering could even program and create these complex machines, heck, probably the mechanical faction in this machine is also as complex, and then the beast of a software to link everything together without mistakes!!!! This is insane.
I am an electrical engineer but I bet the mechanical engineering of a machine that lives its entire life in an incredibly strong magnetic field of its own making must be fascinating problem to solve for mechanical engineers too.
As someone who has worked on the manufacturing of these devices, indeed, the amount of pacltience to manufacture, assemble and install one is insane lol
When we got taught this back in uni that was the first time when I thought "Yes, now THIS is the kind of sci-fi stuff I came here to learn." This video is a good precis for getting a basic grasp on the subject in 15 minutes. Well done!
I played with electronics since 5th grade, became self taught in engineering (being as I am horrible in school) and used FFT during my service as a submarine sonar tech. I almost was hired as an MRI tech, but lost it due to my not living in the area where the job was located. This is the best video I have seen on the subject!
I have personally worked on two of these types of projects. I was the carpenter, drywall,taping, and painting.i can recall working around this machine while they primed it with liquid helium. I have so many good memories from that life! Thanks.
I just completed my Ph.D. in MRI physics! Very nice work on this video. It does a really good job of summarizing the fundamentals of MRI into a short 20 minutes. A small correction: quite a few of your example images/stock photos were of CT scans/scanners and not MRI. I think I also saw some old footage of a linear accelerator used for radiation treatment as well. Both CT scanners and linear accelerators would be excellent topics for future videos!
This is crazy, I just had an MRI done on my leg a few days ago and was scouring RUclips for a thorough video on how MRI worked and couldn’t find a great one. Then you drop this gem
1:11 : The classic mistake(?) of Repeating last word of an Acronym. This is by far the best explanation of MRI Machines I've seen on the internet, The pattern recreation for Fourier Transform was really well explained. Thanks for this.
I'm an MRI technologist and have practiced for 18 years. MRI is the Cadillac of imaging. It's like doing surgery without cutting someone open. MRI doesn't damage the body in any way. I loved my job so much. It was highly technical and never boring. Great explanation of an MRI scanner!
Never seen (or remember) a machine which was result of expertise in different fields of science. Such a complex machine. This is the BEST VIDEO I have ever seen. Almost every aspect of the machine functions is explained in detail. Great graphics and THE BEST explanation. Thank you.
I recently (7 months ago) I started working on GE MR and CT scanners. I have great appreciation for these magnificent and complex machines, and it made my day seeing this YT channel than I have watched for years do a great job explaining how the machines I work on every day, work.
As an undergraduate I was taught by one of the people who pioneered the fundamental science of MRI. Every week, I used to go to a weekly tutorial lasting one hour (with two other people) and we would sit together with our tutor and go through the fundamental maths of this RUclips video on paper. This was in the days when desk top computing was in its infancy. I am a biochemist not a mathematician, so it was very hard work. Last year, I went in an MRI machine for the first time for a diagnosis and promptly fell asleep. Very restful it was. 😁
Oxford biochemist. I arrived in Oxford as an undergraduate when Sir Martin Wood (founder of Oxford Instruments) and Sir George Radda were collaborating. George Radda had installed his whole body machine and as you can imagine we ALL had to learn the fundamental science of NMR that underpins MRI. The machine was the stuff of legend and as mere students we never saw it, but we were taught by the people that were allowed to be in its presence and touch its buttons.😁
I do t think they can make them less noisy. I also fell asleep in them, in spite of the noise. I moved around in my sleep and the image was not as clear as they wanted. Very hard to understand exactly why they work, but amazing science!!
currently going to school to become an xray tech. and this is BY FAR the single best video i've ever found on MRI. You are touching on a bit of everything but still take the time to explain everything in detail. 10/10!
Fun fact: They were originally called NMRI machines (Nuclear Magnetic Resonance Imaging) put people didn't like the sound of being stuck in a "nuclear" machine, so they dropped the N 4:03 Your Larmor Precession is going to wrong way. It should follow the "right hand rule" and be spinning from positive y to negative x. 10:19 Or Helmholtz coils, but that name's less fun, I guess. 12:56 He does not talk more about contrasting agents, but (by my understanding) there's actually a third type of MRI called a "Flair" which is a bit more common in modern scans. 13:37 You can learn more about the Fourier transforms here, if you're not opposed to Veritasium content: ruclips.net/video/nmgFG7PUHfo/видео.html Anything that I didn't list (other than the fact that "spin" isn't real and is only used as a convenient term to get the point across, all of this was explained using the vector model whereas the quantum model is more of an accurate picture, and that the little election particles also aren't real as electrons are probability waves not moving particles) was correct, and I really appreciated this animated explanation. Credentials: Just finished a rigorous graduate course on NMR with a very difficult to get A (only me and two other people have gotten one in the last couple of years for this university). (Feel free to reply if I missed something or screwed another thing up, haha!)
As a technical imaging professional and someone who has worked on brain science, hardware and software in and out of theatre, I commend the team who put this media together. Typically these sort of videos are trite and dumbed-down but this is both intelligent and detailed for anyone who is genuinely interested. Video footage, graphics and voicing on this was close to perfect. This cannot have been an easy project, so my respect. JM.
My father had a benign pituitary tumor. Although its hormonal effects were apparent starting in the early 1960's, with the onset of the threatening, flip-flopping mood changes he suffered from and inflicted on himself and our family. It was finally discovered in 1979, using x-rays, before its successful removal, using "wide-open cranial" surgery. With the development of high-resolution MRI's (along with CAT and PET scanners), though, such things can be diagnosed while they are still small, in just weeks or months, with far less-invasive surgery techniques. *The advancement of medical, particularly medical imaging, technology since 1970 has been nothing short of breathtaking.*
As a neuroradiologist and a MRI everyday user, I can't describe how well done is this video. I'm amazed. Keep up the good work, your Channel Is incredibile
Took physics in high school, had biophysics and radiology in uni, yet, this is the one video that explained everything more clearly and coherently than years of studying. Thank you!
I’m a cardiovascular icu nurse so mris are a huge part of the job. Super cool to find such a good explanation of the engineering behind it since it isn’t really my field and I didn’t fully understand all the intricacies of it!
Great video! I’ve worked in the manufacturing of superconducting magnets for MRI over the last 20yrs, this video is a great explanation of how our technology fits into the MR jigsaw. I will be recommending this video to our staff (new and old) who want to know how our magnets actually work, and their implementation. You’d be surprised how much manual labour goes into making something so high tech.
I study chemistry. As chemists, we use basically the same thing but resolved in terms of frequency instead of 3D space (NMR spectroscopy) to analyse compounds. It's honestly amazing how we can identify a completely unknown substance just by looking at some peaks in a spectrum.
The "N" in "NMR" is avoided like the plague in medical circles. "Nuclear Magnetic Resonance", as chemists and physicists call it, scares their patients.
NMR is the father of MRI, and chemists use also differents atoms than hydrogen (carbon, phosphorus...). It is a invaluable tool in chemistry, and I am always amazed by how it was adapted to probe an entire human body compared to simple samples of chemicals.
Yep, instead of hydrogen, you use various compounds. MR spectroscopy is currently pretty niche in its uses in medicine tho. But we’ll see where it goes
"Some peaks in a spectrum" - talk about downplaying the amount of information! Even a simple 1D proton spectrum gives you at least 3 pieces of information: integration, chemical shift, and multiplicity. Now think about all the different types of correlation spectroscopy that can be used - HSQC, HMBC, H2BC, COSY, NOESY, etc. - each one is asking different questions of the molecule and that's why NMR is so important in structural elucidations.
bruh, i literally had my first MRI scan yesterday, and to keep myself from feeling claustrophobic i started trying to thing about how it works and all the engineering behind it!.
And the story is not over yet! Since the early 1980s, when the basic imaging technique with gradients and FFTs was established, every year the engineers come up with new ideas for improving or speeding up the images. Either hardware improvements or new software ideas, new sequences, etc. I’ve been working in MR since 1984, and back then we couldn’t have imagined the fast reliable high quality images they nowadays produce!!
The amount of video's that come out these day that are of SUCH high quality and make quite complex problems really digestable is just insane. Appreciate the effort you put in!
Thanks for this explanation of how MRIs work. I did my undergrad Chemistry thesis on NMR of small molecules and spent a year working in an NMR lab but never understood how the 2D images were generated for an MRI. This video made it much clearer compared to what I have read in the past
I’m getting an MRI brain scan in a week and this video could not have come out at a better time. Thank you Real Engineering for helping real people like me to wrap my head around this stuff! I love your content
Prepare you for a lots of noise (where i live they give you hearplugs) and bee still for at least 30 min. (i take 2 mri brain and column each years), trust me.
@@Lorre982 Thanks for the heads up. Unfortunately for the “be very still” part I’m getting an MRI because I have rather severe Tourettes so that might not be the easiest for me. I’ll do my best though! I appreciate the sentiment (;
Thanks for this comprehensive explanation - as a patient, I've wondered how the machines work. Incredibly grateful that the technology exists, as it has saved my life.
As a geeky layperson, I found this video extremely educational. My understanding of the world around me is of an order of magnitude enriched. Thank you for this.
Amazing explanation yet again. As a doctor, it’s easy to forget how insanely complex the engineering behind our imaging modalities are when we’re ordering them for patients left and right on a daily basis. Can’t imagine how cool our medical tech is gonna be in a decade or so especially with advancements in AI.
How are people so sure AI will actually work to innovate new technologies? All it knows is what we know so far.... I know they have big goals and maybe or maybe not will reach them.... But for a computer to be more creative than us is still a goalpost we have not really reached yet that I know of.
@@Peter-uo9km I'm a radiology tech. AI is already in use to help radiologists interpret images faster, point out irregularities for possible further examination, and to draw measurements faster than any human could. For instance we can send hand x-rays to an AI server which will draw up measurements of all the bones and determine relative bone age of the patient. This takes like 30s, maybe it would a human half an hour. AI is only going to get better the longer it's worked on and with.
@Martin Lindberg that's more like automation. Unless the AI was never programmed by a human to specifically do these tasks and it machine learned it. But I doubt that. That's not 'intelligence' it's just copying us the humans who created it. I'm talking about real intelligence not just doing fast mundane work.
@kpax xapk you assume the AI would be 100% accurate in those equations? Optimistic are we? Wasn't too long ago they couldn't even get Sophia to stop saying she wants to kill all humans. Since AI is modeled after our own intelligence... and since we learn from our mistakes...AI too will also learn from its mistakes...I reiterate that AI will make mistakes....have you not seen how it learns? Like Google AI learning how to walk amd stuff...it's gradual learning and it's clumsy like a child learning to walk. AI WILL make mistakes.... GENERAL AI is their grand scheme. You're still talking about super specialized tasks. They want general ai where it's behaving like a conscious human being who is free to think.
@kpax xapk I don't think you understand what the real promise of AI is. It's beyond looking at xrays... like I would tell the AI to think of a new way to see inside our bodies without using xray radiation which is harmful to humans... and if it's really a billion times smarter than us then it should find a way.... but I'm not quite certain AI can actually do that. Invent new shit that was never an input from humans and their internet knowledge base. Being able to scan an xray for things humans already know is not intelligence it is just preprogrammed work. Nothing new or creative about it... it's just tireless so it will never get lazy about it... but it's not actually smarter than the human who programmed it with his own knowledge.
I just had 3 surgeries this year requiring this technology and I never appreciated this more. It's truly miraculous and I'm grateful that I live in the 2023 era of science.
Great video from you & team. Radiologist here, I always find that my students have trouble grasping the concepts of MRI. Looks like I will be making them watch this first before diving into the details.
I was literally looking out for MRI machines a week ago and there was no well made video. I even requested veritasium to make one and as much as lucky as I could get this dropped in my recommendation. Thank you so much.
Fun fact: For the highest level of education you can choose in the Dutch equivalent of high school, the principles behind this are REQUIRED to be taught and can show up on your exams. I myself was unlucky enough to have this on an exam where I had to manually calculate the voltages and more to achieve a specified gradient slice in an MRI machine, it's absolutely bonkers physics but it does genuinely work really well. Normally it shows up on your exams with easier examples than MRI's though, but I have gotten a grade for this exact thing. (You'd roughly be 17-18 years old at this time, with 3 years of chemistry/physics under the belt and then another 3 years focusing in on physics)
I've had 2 MRIs (shoulder and prostate) and as an engineer, I always wanted to know how they work. Thanks for this! Wow. To stay calm in that tiny tube, I literally count every second with eyes closed as time ticks by. Oh, and the worst experience ever is when they inject that cold contrast agent into your blood...I still get queezy remembering that now...
I have had 3 MRI scans 2 without any 'stuff' injected and the experience as fine although I am frightened of small spaces. The 'stuff' injected for the third scan made me feel strangely hot and close to being sick but then going into the machine itself was rather pleasant in contrast. I had had a blockage of a carotid artery so they were checking brain (I presume). I would like to say a heartfelt thank you to the lovely French medical staff who looked after me in St Lo. Yes I am a 'strange' Brit and they didn't understand my odd attempts at humour.
My biggest question after watching this video is: if the spin of H atoms in your body were forcefully changed and reverted back, are you still you after an MRI scan, philosophically speaking?
One of the best and most complete MRI explanations I've seen! One complaint: Xrays aren't 'high energy radio frequency waves', since radio waves are lower frequency than IR while Xrays are higher than UV. That's precisely like saying "blue light is high energy red light".
Unbelievable video! May even be your best on this channel. I have had the misfortune of being inside an MRI machine more than 10 times but never thought to ask how it actually worked - I was always satisfied by thinking of it as being a bunch of "magnets". I can not believe that spin is the fundamental quantity that is being measured. That is just wild.
I am so glad to finally understand why the machines are so loud🤣. Having been in an MRI machine with the extra coil on my chest, it is great to finally understand the engineering behind them and appreciate them for the complexity. Thank you for this amazing video :)
I had an MRI today. It was a fantastic experience. Felt like being teleported to another universe. I had very clear thoughts for the duration of half an hour. Every separate scan session in the series gave another sound spectrum and drum rhythm. It was kind of a deep pulsing meditation, my whole body was glowing afterwards.
I am an MRI tech. Great explanation, and loved the footage of actual consoles with all the different scans on them - it was interesting seeing some programs and features we don't have where I work. I would add regarding the T1 and T2 thing, an MRI exam invariably consists of a series of scans with different parameters and settings, always using both T1 and T2-weighted settings and a slew of other technologies to suppress or enhance different tissues, which are then all compared against each other by the radiologist. It would be a waste of time to bring someone in for a single scan.
Can people with metal implants made of titanium and cobalt go inside MRI machines? If not, why are cobalt implants allowed? And what does the user manual for MRI machines say about cobalt implants?
@@LGM4 You only have like a big black spot around the implant because the field prefers to go through the metal and not the tissue so it is mostly not a risk but you can just not diagnose the tissue around the implant
I've had several mri's over the last 25 years or so and every few years they get better and faster. Once in a hospital I spoke to a neurologist about their new machine that they had tested on him and other colleagues and the machine was such a step forward that they discovered 3 problems with the colleagues (who were still symptom free) in the testing phase alone.
I felt extremely excited when it came time to get my MRI. I was in a terrible amount of pain from a severe knee injury (torn ligaments, patellar tendonosis and an extruded meniscus) but I was giddy like a child when they put me through it. The images and video I have of my knee is at a crazy resolution. I knew they were hi-res but this was mind blowing.
While I'm sure the technical descriptions of the internal workings of MRI are correct based on other comments, there's a ton of really basic factual errors at the start related to its use and the broader context of medical imaging that are really bothersome: 1) MRI is not the first 3D imaging system in medicine, that would be CT (formerly CAT) scans, which use X-rays 2) MRI is not a particularly high resolution imaging modality in practice, again beaten out substantially by CT. The benefit of MRI is contrast because there's more distinction between different tissues magnetically than by density, but in terms of raw spatial resolution CT actually beats it when used in high resolution protocols 3) Ultrasound doesn't rely on density, it relies on acoustic properties, an important distinction since some of the body's contents have radically different acoustic properties while having nearly the same density 4) MRI has (almost) nothing to do with coronary stenting, which is done with 2D live X-ray imaging with iodine contrast and is nowhere near as precision dependent as you imply since the technique is almost entirely 1 dimensional (traversing within whichever blood vessel is being targeted, outside of wire manipulation to choose which branches of blood vessels you follow). MRI in cardiac disease is used to diagnose other cardiac conditions that require those magnetic property distinctions rather than the pretty basic contrast enhanced density approach that works well in hollow blood vessels with X-rays and CT 5) 2:09 You didn't *explicitly* call this an MRI but the context implied it - that's a CT scan, not an MRI
Thank you! I did NMR analyses in the 1980s, before the samples were bigger than a mouse. So I found the information here on how the sample became the size of humans to be fascinating. Then you bring up Keysight, which also has ADS software which I helped develop when Agilent still owned the tech! This brought home for me two major parts of my life. I am grateful.
Really wish this video existed when i did my quantum mechanics, inorganic, and physical chemistry classes. The visualizations are so helpful. Amazing video man.
Although still confused, I have, at least, learnt more about MR scanning. I still don't understand how the MR differentiates (along the X and Y axes) each "pixel" density and its position. I also don't understand why we get a slice, rather than an intermingling of adjacent tissues. I'll try and do a bit more reading and see if i can make more sense of it. This video was ferociously and meticulously researched. To understand MR scanning, it's going to need about 100 hours of presentation, split into 1 hour sections (with accompanying notebooks) that assume no prior knowledge of maths or physics. That will be obviously beyond the scope of "Real Engineering" but I'd like to think that some educational company will be inspired by this video and make a thoroughly detailed course on this very fascinating and valuable subject.
During my physics bachelor I have done an experiment on a very small MRI machine, so I knew most of the theory behind it, but you've done a *_very_* good job explaining it! At least a lot more clearly than the experiment explanation pdf that was provided. :)
Happy for this video. I’m an MRI technologist A lot of the images shown are CT, not MRI. 2:09 for example 4:22 is another error. Transmit/Receive coils (TxRx) are not within the machine. Only transmit coils are. Receive only coils or TxRx coils are placed on the patient’s body or directly around the anatomy. 12:37 *MRI technologists Good video! It may seem like this was a negative comment, but I wish I had this video when I started my schooling. Thanks.
@@springer9406 if you’re referring to the head coil, some are TxRx. But that is not within the machine. It’s a coil that is attached to the table within the bore. Just look up TxRx head coil and you’ll see it.
@@cade8986 The whole body coil, which is part of the bore. Some are Rx too. If you'd like a neutral source: "Volume coils (eg, the body coil) completely encompass the anatomy of interest and are mostly operated as transmit coils but also capable to be operated as transmit/receive coils." onlinelibrary.wiley.com/doi/full/10.1002/jmri.26187. Or this cdiok.com/wp-content/uploads/2012/01/MRI-Technology.jpg. Or this (figure 9): onlinelibrary.wiley.com/doi/10.1002/mrm.20177
We basically have to do crash courses in these subjects in 2-3 weeks. Very intensive. To simplify things down we really need to understand them. Massive shout out to Josi who wrote this script.
@@RealEngineering I a bit regret I am not younger (and without a family to support), to try to join you (or some other great edu-tubers team, like Sci Show). So I at least enjoy your work.
@@DrunkenUFOPilot It's definitely fun being able to decide what you want to learn about month to month. Very fullfilling. Start writing. It's free. Send your work to people who need help with it.
Thank you for finally making me understand how this stuff works! I knew about the superconducting magnets and helium, but now it makes sense why those things are needed and how we can use magnets to make medical images
![ARGENTINA vs. PERÚ [1-0] | RESUMEN | ELIMINATORIAS SUDAMERICANAS | FECHA 12](http://i.ytimg.com/vi/xtI25ENDDSY/mqdefault.jpg)
I've been trying to understand this for YEARS. YEARS!! This is the first time I actually get it. I knew about the FFT, but projecting lines and how they manage that finally makes sense. Incredible explanation.
I've always wanted to know more about MRIs because there the reason I'm alive, and i really enjoy watching your videos
How about a diy mri?
instead of watching all the creators i watch, create some new videos yourself :P
Bruh did it really take you years? didn't you study physics?
same
I did my PhD in the field of MRI. I wish I had this video back then. Great visualizations without oversimplifying things.
What were you investigating?
Were you a physicist, EE major or what?
@@schwerner343 it’s called ‘studying’.
@@johnnunn8688 lmao I'm very aware of that! A PhD usually includes a major research component where you investigate some hypothesis or theory (usually experimentally) and present your findings and justify them in a Viva. If you pass that, you get your qualification!
Did you work with Damadian?
I’m board certified in cardiac MRI and spent 4 years doing research in the field and this is honestly the best explanation I’ve ever seen. Also the best looking, loved the Fourier animation. More medical tech, fewer planes! 😂
Weeeee, look who's here! Love your content!
I'm subscribed to both of you
I agree with Rohin, more medtech please! And, as always, great explanations
It's too bad the machine costs so dang much to get scanned when you don't have insurance :/
im so glad the med channels i watch also watch the other med videos lmfao
This is the definition of exceptional science communication. I have a background in engineering (and overall very mechanical curious) and have spent hours trying to understand how MRI’s work after getting a few done on my knee, none of this research can come close to the elegance of explanation shown in this video. Bravo.
lol
The detailed explanations and depth of knowledge in this video, the M1 Abrams video, and countless others consistently blow my mind. Thank you for all your hard work researching and delivering such high quality videos!
So excited you made this video! I'm a Diagnostic Radiologist who teaches Radiology to medical students. This is as deep an understanding of MRI physics as almost all of my students need. (Less than 10% of my students will have to pass the Radiologic Physics boards that I took to become a Radiologist.) FWIW, we do try to use "surface receiving coils" as often as possible for better images. Seriously, every one of my medical students will be expected to watch this video. Thanks, Brian and team!!!
Exactly. Couldn’t agree more!!
Do you see the new generation of large AI models having a large impact in diagnostic radiology? I feel like that has been "just around the corner" for the last 20 years.
@@csours I told folks 20 years ago that I would interpret images for 10 years and then over read computers for 10 years and then find something else to do. As it turns out, I haven't started over reading the machines yet, so I'm no longer sure what the timeline is. But it is always just around the corner. We (human radiologists) are too expensive to not replace as soon as we can be replaced.
@@texastaterbug5395 It's also a regulatory issue. AI isn't really allowed to make a diagnostic reading (but it will get use as a first reader to indicate clear cut cases where only one additional radiologist needs to have a quick look to confirm...thus lightening the workload on radiologists and allowing them to spend more time on the 'interesting' cases)
@@peterzerfass4609
It's just a tool. On one hand, AI assisted analysis fits the work flow and I can see it making everything faster. On the other, that could lead to needing fewer radiologists overall. Everybody needs to take a step back and remember that all the niche AI applications popping up are only tools, and nobody should feel threatened to learn how to use a new tool.
I'm an MRI physicist, and you've done an excellent job explaining the physics and engineering that goes into an MR scanner!
I have a question left that I don't think the video has answered and that I have been struggling to find an answer to elsewhere as well:
I am familiar with the fundamentals of the technique from the area of chemistry using NMRs to help determine the structure of molecules. From this I know that the resonant frequency of hyrdogen nuclei is not as directly connected to the strength of the outer magnetic field as the video here implies, but that it is in fact dependant on the actual net magnetic field that the nucleus experiences and as such is influenced by the surrounding electron density and thus the chemical structure of the molecule it is part of (this is precisely what NMR-techology uses for its analysis).
So my question now is: If we already use the different resonance frequencies of hydrogen to select the "slice" of the patient we want to look at, how do we not get a mixed image from hydrogens in different chemical compounds and bonding situations having the same resonance frequency in slightly different slices? Are the differences in resonance frequency created by the gradient in the magnetic field just much larger than the differences through different chemical bonding situations (and does that mean this limits how thin the slices can be) or how are those problems avoided?
A very sincere thank you, if you have taken the time to read and consider my comment and question!
@@nitramreniar question was not for me but I may have an answer (though admittedly it's been a while since I worked with MRI machines). There is a so called chemical shift, it presents itself as the image shifting a bit, leaving black borders around an organ for instance. It's usually very small though and iirc can be corrected somewhat using some sequences such as fat suppressing ones.
@@NerdyNEET do better then
No, they didn't. This makes sense only after you have already learned it.
@@nitramreniar
The effect given in the video is Larmour precession, this is caused by electron spin not nuclear spin. It is a similar phenomena but there are different particles involved. NMR is due to nuclear spin not electron spin.
I am a board certified radiologist and involved in MRI science. And this video is an educational piece of art! You and your team is incredible! Congrats!
So well educated yet you cannot spell peace!!! It's piece. You and your team ARE ...
Peaceful art
@@drteeka7883 😅😂
8:40 there is also something called "quench", which is basically if you need to turn off the MRI machine (if, for example, an accident happened and a metal oxygen tank was launched across the room towards the machine) there is a big red button which increases the inside temperature of helium so that it removes superconductivity, which causes helium to instantly boil from liquid to gas. That's why there is a quench pipe, which releases the helium gas outside in a huge cloud of white gas. However, if for some reason something breaks and a helium leaks inside the room, aka the quench happens inside the MRI room, it can be extremely dangerous because people inside the room can suffocate, which is why MRI room doors are (or should be) designed to open towards the outside, because if a bunch of helium releases inside the room, it's going to put pressure onto the door, you won't be able to open them and you could die. And in case of a quench, you will lose helium, and need to refill the MRI machine, which is extremely expensive.
Imagine screaming in agony because there was a helium leak and all you make is chipmunk sounds !
@@BlastarX Probably not in agony, but panic and lightheadedness wouldn't be too much better.
Even better - depending on the cryonics in the cryostat, you have multiple different hazards from a blocked quench exhaust:
First, liquid helium. This is lighter than air so you need to drop to the floor. As you mentioned, this can also cause the door to become jammed which can lead to suffocation.
Second, liquid nitrogen. This is heavier than air, so dropping to the floor will actually kill you. In this case, you actually need to walk normally/keep your head up. This is very rare outside of research installations.
Third, liquid helium and liquid nitrogen - worst of both worlds and uncommon but possible. In this case you have to keep your head in the middle of the room - too high and you suffocate, too low and you suffocate.
The training I received is to crouch walk out of the room and to always block the door open with a doorstop or other mechanism so long as you’re not doing a scan.
None of this includes the frostbite risk if you’re standing next to the machine.
@@ScienceReasonLove helium stratification typically requires a calm environment, an MRI quenching is violent, fast, and the expansion ratio from liquid to gas is high, dropping to the floor is likely going to waist the 1 or 2 seconds you have to get as far away from the thing as possible.
Most MRIs nowadays have gas containment shells that envelop the entire machine and route into the exhaust conduit so that if there is a helium escape it will go up the exhaust anyway. Machines with shells look far more bulky than machines without and often look as though they are built into the room.
My father helped establish some of the foundational physics and engineering that lead to MRI’s back in the 80s. When I was born in the 90s, I was diagnosed with a rare genetic condition, that resulted in me needing several MRIs every year, indefinitely. I definitely think the enthusiasm he showed for these machines, along with his willingness to explain how they work, really helped mitigate the fear of going inside them, especially when I was a child.
Thank to your father for contributions for medicine and Those MR machines
I am neither a medical nor a medical engineer or an engineering student. I study pure mathematics. A weird one too, since I'm not fond of physics as well even though I am studying the proofs behind their equations. But seeing how elegant equations like the Fourier transform you mentioned are, brings nothing but a big smile on my face.
During Fourier's time, there was no talk about electromagnetic waves and such. Early mathematicians only had Sin and Cosin waves, the relations between them, and a blackboard and chalk. To think that their tinkering with such trivial sketches would lead to constructing such wonderful pieces of technology is actually mind-blowing.
This further proves that the very core of everything in this universe is tied to mathematics.
Agreed. The Fourier analysis is the key. Without the power of the transform these machines would be useful only for inflating balloons and pranking nurses.
I remembered in my high school science class that mathematics is the "language" of science. Scientific laws can be expressed through an equation.
The world owes the mathematicians an unreturnable debt.
? Vomit ! what ?
The equations were originally used to help determine tide tables
It’s crazy how often a Fourier transform is credited with enabling cutting edge technology. It’s even crazy that most people don’t know who Fourier is despite how influential his work it.
A VERY useful tool in a lot of fields.
FT pretty much runs the world right now. It's used literally everywhere since we process signals digitally.
The Fourier Transform is seriously one of the most "beautiful" concepts, just so damn elegant. It's a completely non-obvious solution when faced with the problem, and then "why didn't I think of that" when you know how it works.
The applications people have found for it are equally amazing. JPEG compression blew my mind when I learned how that worked.
I was thinking the same thing! When they brought up hydrogen spin and processing the decay, I was just waiting for the mention of the Fourier Transform.
@@mikosoft FFT*
I used to consult for the Toshiba MRI division. The building where they tested the magnets had a huge open area with yellow warning stripes painted all over the floor marking the "keep clear" area around the magnet under test. Turns out the yellow stripes were added after a screwdriver was left out and became a projectile during a magnet test. It punched a hole in the magnet assembly which is not a good thing when you're dealing with that much power.
That sounds astoundingly dangerous!
Crazy news story from this year (2023): “Leandro Mathias de Novaes was taking his mother for a scan at the Laboratorio Cura in Sao Paulo in January and entered the facility with a concealed handgun. Despite warnings from staff to remove jewellery and metal objects in the MRI room, Mr Mathias [secretly] kept his weapon on his waistband. The magnetic field from the machine was so strong that the weapon was pulled from his waist and fired off a round, which hit him in his stomach.” (The Telegraph)
@@pukpukkrolik nah that wasnt an accidental shot, the electricians took it personally 💀
@@pukpukkrolik I hope nobody was already in the machine...
@@pukpukkrolik great example of karma lmao
Man I cannot get over how well you explain such complex marvels of technology. Thank you for these videos. They are wildly entertaining and informative.
As a former mri technologist, I am happy with this very clear explanation without oversymplifying. Brings back memories about understanding the process of 'fitting the k-space' with all the signals measured and generated by the measuring gradients after the initial excitation pulses (slice selection and x/y matrix) echoes readout. FFT was genious. And nowadays mri relies far more on ultrafast 3D volume scans with very little gradient angles instead of the 'simple' 180 degree spinecho sequences. And the dreaded mri sequence acronyms between the manufacturers!
Thank you for rhis video.
I'm an MRI engineer for a company featured prominently in this video and this is genuinely one of the best summaries of the technology I've seen so far.
If you're still baffled after watching this video that's perfectly normal. This is stuff that's typically taught over the course of weeks or months not minutes. In spite of the short run time this video is densely packed with useful information.
Well done.
One question ive had a while you might be able to answer then is is there atleast a safety system or spool up time on these things if theres anything magnetic detected? Say one happens to have a peice of mettal in the body like the appendex without knowing? The thourght alones terrifying.
@@JMAssainatorz the device itself does not have that safety because when one is in the strong field the bad thing can happen. like people with heart pace makers or war veterans having shrapnel in critical locations like brain. it is the staff responsibility to make sure there is no such thing. and imaging with metallic pieces in body will affect the image due to their different magnetic properties.
In another comment, an MRI technician pointed out 13 things that are wrong in this video. If you're an MRI engineer, you cannot not point them out.
@@JMAssainatorz No. The magnet is always on. It could quench however if the field is disturbed too much and lose superconductivity, then it will no longer be magnetised.
I need some numbers in the last part: How fast does it change different frequencies? What do the patterns really look like?
I am a researcher working in the field of MRI, and this is the most informative and beautiful exposition of the topic I've ever seen in a video form. Your visualization of 2D gradient encoding (14:45) and your straightforward explanation really drives home this core concept which is otherwise quite hard to grasp. Thanks for creating this great resource!
Hi Chinmay, i did MRI 4 years back for Scitica nerve pain.normal report.but i always feel like one nerve is contracted in hip due to which my leg and hand become senseless. Shall i go for one more MRI ? Plz reply
Can you please answer some of my questions? Due to my condition I had MRI every year and every time they said. " If you feel sick press this balloon but we would have to restart whole imaging"
Fist: What in that system could make me sick? Second: Why restart? My brain would be same even after some minutes,no?
Where did you recieve your training? Did you study physics or engineering or mathematics?
@@malikovajana about the second question ,no our brains won't be the same because when we talk or think or be in a high focusing mod according to doctors it's not the same as when we relax that's why maybe sometimes they talk with patients during MRI
Hope i answered ur question ..
Have to admit that technology changes life.
MRI Physicist here, fantastic video, I have to recommend this to some beginners. I didn't notice any errors either, just two small inaccuracies: 1. The spins align themselves along OR against the magnetic field B0, but a tiny majority (~0.001%) align along B0, creating the net magnetization. 2. Some of the footage you used were CT images... but whatever. I had a blast and was surprised when the video ended suddenly. Great job!
Thanks for your comment, he also described x rays as "radio frequencies" should have said electromagnetic ...
What do you think is the phisical limit for the resolution of these machines?
@@Valery0p5 thanks for the comment to the comment. Not sure why we’re thanking each other, I just wanted to join in on the fun!
@@Valery0p5 They are radio frequencies, they're literally called 'RF pulses'. The resolution is based on the strength of the gradients, so making the gradients more powerful you get smaller voxels. But when you decrease the voxel size you get less signal, which makes the image dark. So you need a more powerful magnet. But the more you increase the magnet strength, the more problems you have with SAR (special absorption rate), or how much you heat up your patient. MRI is basically a low power microwave, something we do NOT tell our patients. :D That is why there are strict SAR limits, which usually only come to play at higher field strenghts. 7T's biggest headache by far is the constant struggle with SAR.
As a side note, MRI is, in my opinion, the most complex imaging method there is. Everything incluences everything, and increasing some parameter has direct concequences and tradeoffs with all other parameters. For example, every point in k-space influences every pixel in the image, so if you have a white pixel in k-space, the whole image has a diagonal stripe pattern on top of it.
Correction, spin is referred to the electron's spin, atom core (with neutrons) doesn't contain any spins i.e. electric orientation with underlying quantum field.
Magnetism comes from electrons spin either -1/2 or +1/2, also this spin is not a literal spin but rather how energy of electon interacts with underlying electric field.
Think of electron as oscillations in electric field that exists in all space with some potential energy. Electrons are disruptions in those fields, same as protons and other matter oscilating at different frequencies. These values come from those fields' energy potential and oscillations will determine which particle it is.
P.S. Before anyone corrects, protons do have spins but due to their mass, their magnatic moment is negligible compared to electrons ( i.e. ~ 2000x stronger). Mass is inversly proportional to the magnatic moment of the particle.
Could you please tell how electro magnetic radiation of MRI affect a person? When I did it, it affected my thyroid
I've had more MRIs than I can count since I had a (luckily benign) brain tumor removed in 2015 that was in the middle of my brain. This was so well explained and I found it very interesting as a patient to see and understand what goes on inside the machine. After having so many, I've gotten so used to the clicks and crunks and buzzes that I can usually fall asleep in them now. 😆 I'll need check-up MRIs for the rest of my life, so I'll definitely think of this when I get my next check-up! Thanks!
Much love Kelsey and I'm glad to hear this incredible technology has helped save your life.
Curious what symptoms you had or how you figured out you had a benign tumor ?
@@HarperLloyd It's kind of hard to explain, but I would get a "whooshing" or pulsating feeling in my head every time I went from standing up to lying down, sitting to standing, etc. At first I thought I was getting up too fast, but then they started to happen when I was just sitting doing nothing and THEN they started getting very painful, too! Incredibly, I had no other symptoms. I kind of scoffed at the idea of needing an MRI because I thought it was overkill. Good thing I listened to my doctor! They thought it was a "cavernous malformation" until they got it out and biopsied it, discovering it was a benign tumor. I just got a check-up MRI a few months ago and all clear!👍
Where did you recieve your training? Did you study physics or engineering or mathematics?
I had a benign tumor removed in 2019, it was 2.05" x 1.85" meningioma near my cerebellum. I had headaches since 13 years old, and leading up to my first MRI that found the tumor, I had migraine headaches, pressure in my head, whooshing sounds accompanied by pulsating pain when I stood up, and felt like I was freezing cold all the time. The pulsating pain in my head was enough to vomit and pass out several times, if I bent over and back up too rapidly, stood up to quickly, or stood up from laying down too quickly. I had ocular migraines for 3 years prior to surgery. 4 years on and 9 MRI's later since surgical removal, all clear. The sound of the machine is like an orchestra to me. I imagine percussion, reeds and strings.
I work with medical imaging devices and the MRI is by far the trickiest and most complicated machine to understand. The engineering, creativity and knowledge needed to have come up with this method of generating images is something that blows me away every time I think about it. And now it is a little bit easier to understand thanks to this video, amazing job. And beautiful animations, btw.
Where did you recieve your training? Did you study physics or engineering or mathematics?
Man I am studying medical engineering at university and had a module on medical imaging this term and MRI was the part that I couldn't wrap my head around. This video made it really easy to understand whilst still keeping high detail. Massive kudos to you again!
I have studied physics. We had to make 9 experimental projects per semester. One of them was magnetic resonance. While the other projects gave you materials to study at home and then you were collecting data for 2-3 hours, this one was different. There was the materials to read at home and when I arrived there was another 3 hour lesson at the end of which I started to understand (then pressed a couple of buttons and the measurement was done. This all was just to get the chemical composition of the sample, not the image. I had absolutely no idea how the Imaging part works!
I am just a medical comp sci student but I did have two courses related to MRI. As I recall the image is achieved by adding another localizing magnetic field. This allows for every point in space to have a slightly different signal. As you know the expected signal this can be used to associate the different signals with different points and then generate the image.
But take that with a grain of salt. My course was mostly focused on what you do with the image not how it is generated.
EDIT: Nevermind this is actually explained much better in the video.
As a radiology resident, I studied so much text to understand this, and you have explained the concepts so clearly.
So much respect to the engineers and their contributions who made it possible to save hundreds of lives just in our setup every year.
Do you actually understand this as a doctor? My dad's a radiologist and i dont think he knows the physics.
We are required to understand the radiology physics . If your dad is a radiologist, i am sure he understands the physics of the machines, CT, MRI , USG and xray.
@@radiologydaily I really doubt that you know how to build or design one of these machines from scratch, if you can't do that then you don't really understand the machine
you would need a bunch of phds from a bunch of engineering and physics fields just to work on each tiny aspect and the breadth and detail of knowledge is not something that you are going to get from a 20 minute video, It would take you probably an 1 year just to master all the proofs for the fourier transform and understand it well enough to apply it to the design of a mri machine
@@xblade11230 don’t be a dick
Someone can definitely understand how something works without being able to design and build it from scratch
@@xblade11230 radiologists just need to understand how to analyze these images. Knowing how they work assists in that. Don't think they need to know how to build a MRI machine tho
I’m an Electrical Engineer who heavily focused on NMR research during grad school. Absolutely fantastic explanation and animations. I’ll be sending this video to my professor!
I used to work at a factory that made MRI's, I was welding on the main pressure vessels that held the liquid helium. I have a small chunk of that Nb/Ti wire. The actual superconductor is Tiny, and is encased in copper to help protect it and let it be formed.
As someone that's recently spent a ton of time inside a few of these machines recently, this was a really informative video. I'd understood the basics of these machines ever since a high school physics course some thirty years ago, but your presentation here really made it click for me. Thank you for making this video, and thank you to the people that worked together to invent these machines.
Is this not dangerous? Maybe it can harm your brain or erase the memory.
@Ed No its not, I've had dozens of MRIs over the years, they're not harmful at all. The safety and non-invasive nature of them is one of the many reasons they're so great!
@@Ed-rt9qt
It can't because your brain makes memories which are stored by links between neurons. The magnetic field does not disrupt neurochemical processes and does not damage the neural links and therefore has no impact on memory.
CT scans on the other hand emit ionizing radiation that can disrupt neurochemical processes and could damage neurons and proteins throughout the body with enough exposure. Over time this could amount to real tangible damage to organs including the brain if that organ is being imaged a lot.
The likelihood of any X-ray or CT causing damage to a sensitive organ like the brain is effectively zero. The doses of radiation are very small dispensed over short time periods, it's not like you're standing next to a pool of molten Uranium 235. That said it's more likely you'd experience some minor damage from X-rays than Magnetic Resonance Fields.
Not like the field is strong enough to bind your atoms to a physical position in space... that would take a few hundred more Tesla to achieve and that could be dangerous. 66,000 Tesla field could hold a person still if they weren't already moving while a fluctuating 1,800,000 Tesla field could start ripping a person apart under the right conditions (frequency considerations being managed). ...Just trust me bro.
Seriously there's no known limit to what a human can tolerate but in fairly high static and dynamic magnetic fields people have reported strange symptoms inside high Tesla fields usually above 3 Tesla (which is as the video states 3 Tesla is generally the upper limit of most MRI machines used on people in a diagnostic capacity).
Most people won't ever experience an issue from any MR scan.
@@VariantAEC Well, I think it can be dangerous because this machine makes hydrogen atoms oscillate, and if atoms oscillate then it means that they have energy.This is similar to a microwave oven, where water molecules oscillate and this is transferred into heat.Because heat is oscillations of atoms or molecules, so the oscillations of hydrogen atoms must also produce some heat which must be dissipated somehow.And maybe it can even rip the molecules apart.
I have had several MRI tests and also memory testing and my memory is above average. The longest that I have been In a MRI was 45 minutes and I am so used to them that sometimes I fall asleep because I'm bored.
I love this channel. Makes me realize that engineers are true aliens to regular people. The amount of knowledge, time and collaboration to make something like this is astounding and alien to me
They stand on the shoulders of previous engineers who all added their bricks to the pyramid of knowledge that brought us here.
@@Alain.Robert very well put, add previous , mathematicians, scientists, etc
and netflix just makes people dumber
Not all inventors and engineers are equal.... some are just pencil pushers at a desk. I've yet to meet an engineer that invented something.
This is an insanely good video - this is a better explanation than I saw in my graduate 700-level NMR class
Still way harder for me to understand despite working with NMRs for nearly 4 years now😂
As a chemist working with NMR spectroscopy, I had always wondered how imaging worked. Thanks for making such an informative and entertaining video!
I am a medical student and this is absolutely the most eloquent explanation of how MRIs work. I never grasped T1 vs T2 imaging but this actually explains the reason why they are different. Excellent video as always!
Currently studying mechanical engineering, I simply cannot fathom how our cousins in electrical engineering could even program and create these complex machines, heck, probably the mechanical faction in this machine is also as complex, and then the beast of a software to link everything together without mistakes!!!!
This is insane.
I'm studying EE and I also cannot fathom it
It's more the combination of physicists and engineers.
I study physics and I too have no idea :D
Studying EE is the only time learning more about something has made me more convinced it was black magic. Masochism is required to take this major.
I am an electrical engineer but I bet the mechanical engineering of a machine that lives its entire life in an incredibly strong magnetic field of its own making must be fascinating problem to solve for mechanical engineers too.
Kudos to people responsible for making these machines, the amount of patience and intellect required to accomplish this is amazing
@@STA-3It was discovered by a Physicist
@@eren_yeager9927Pretty sure it took a bunch of people to bring these concepts together to design and build a workable machine .
As someone who has worked on the manufacturing of these devices, indeed, the amount of pacltience to manufacture, assemble and install one is insane lol
When we got taught this back in uni that was the first time when I thought "Yes, now THIS is the kind of sci-fi stuff I came here to learn."
This video is a good precis for getting a basic grasp on the subject in 15 minutes. Well done!
I played with electronics since 5th grade, became self taught in engineering (being as I am horrible in school) and used FFT during my service as a submarine sonar tech. I almost was hired as an MRI tech, but lost it due to my not living in the area where the job was located. This is the best video I have seen on the subject!
I have personally worked on two of these types of projects. I was the carpenter, drywall,taping, and painting.i can recall working around this machine while they primed it with liquid helium. I have so many good memories from that life! Thanks.
I just completed my Ph.D. in MRI physics! Very nice work on this video. It does a really good job of summarizing the fundamentals of MRI into a short 20 minutes. A small correction: quite a few of your example images/stock photos were of CT scans/scanners and not MRI. I think I also saw some old footage of a linear accelerator used for radiation treatment as well. Both CT scanners and linear accelerators would be excellent topics for future videos!
This is crazy, I just had an MRI done on my leg a few days ago and was scouring RUclips for a thorough video on how MRI worked and couldn’t find a great one. Then you drop this gem
1:11 : The classic mistake(?) of Repeating last word of an Acronym.
This is by far the best explanation of MRI Machines I've seen on the internet, The pattern recreation for Fourier Transform was really well explained. Thanks for this.
RAS syndrome
LED diodes!
CD Disks
I'm an MRI technologist and have practiced for 18 years. MRI is the Cadillac of imaging. It's like doing surgery without cutting someone open. MRI doesn't damage the body in any way. I loved my job so much. It was highly technical and never boring. Great explanation of an MRI scanner!
Never seen (or remember) a machine which was result of expertise in different fields of science. Such a complex machine.
This is the BEST VIDEO I have ever seen. Almost every aspect of the machine functions is explained in detail.
Great graphics and THE BEST explanation.
Thank you.
I recently (7 months ago) I started working on GE MR and CT scanners. I have great appreciation for these magnificent and complex machines, and it made my day seeing this YT channel than I have watched for years do a great job explaining how the machines I work on every day, work.
As an undergraduate I was taught by one of the people who pioneered the fundamental science of MRI. Every week, I used to go to a weekly tutorial lasting one hour (with two other people) and we would sit together with our tutor and go through the fundamental maths of this RUclips video on paper. This was in the days when desk top computing was in its infancy. I am a biochemist not a mathematician, so it was very hard work.
Last year, I went in an MRI machine for the first time for a diagnosis and promptly fell asleep. Very restful it was. 😁
Where did you recieve your training? Did you study physics or engineering or mathematics?
Oxford biochemist. I arrived in Oxford as an undergraduate when Sir Martin Wood (founder of Oxford Instruments) and Sir George Radda were collaborating. George Radda had installed his whole body machine and as you can imagine we ALL had to learn the fundamental science of NMR that underpins MRI. The machine was the stuff of legend and as mere students we never saw it, but we were taught by the people that were allowed to be in its presence and touch its buttons.😁
What the heck makes them so !@#+× noisy? Why engineers not doing enough to bring down the noise .
I do t think they can make them less noisy. I also fell asleep in them, in spite of the noise. I moved around in my sleep and the image was not as clear as they wanted. Very hard to understand exactly why they work, but amazing science!!
currently going to school to become an xray tech. and this is BY FAR the single best video i've ever found on MRI. You are touching on a bit of everything but still take the time to explain everything in detail. 10/10!
Fun fact: They were originally called NMRI machines (Nuclear Magnetic Resonance Imaging) put people didn't like the sound of being stuck in a "nuclear" machine, so they dropped the N
4:03 Your Larmor Precession is going to wrong way. It should follow the "right hand rule" and be spinning from positive y to negative x.
10:19 Or Helmholtz coils, but that name's less fun, I guess.
12:56 He does not talk more about contrasting agents, but (by my understanding) there's actually a third type of MRI called a "Flair" which is a bit more common in modern scans.
13:37 You can learn more about the Fourier transforms here, if you're not opposed to Veritasium content: ruclips.net/video/nmgFG7PUHfo/видео.html
Anything that I didn't list (other than the fact that "spin" isn't real and is only used as a convenient term to get the point across, all of this was explained using the vector model whereas the quantum model is more of an accurate picture, and that the little election particles also aren't real as electrons are probability waves not moving particles) was correct, and I really appreciated this animated explanation.
Credentials: Just finished a rigorous graduate course on NMR with a very difficult to get A (only me and two other people have gotten one in the last couple of years for this university).
(Feel free to reply if I missed something or screwed another thing up, haha!)
As a technical imaging professional and someone who has worked on brain science, hardware and software in and out of theatre, I commend the team who put this media together.
Typically these sort of videos are trite and dumbed-down but this is both intelligent and detailed for anyone who is genuinely interested.
Video footage, graphics and voicing on this was close to perfect. This cannot have been an easy project, so my respect.
JM.
My father had a benign pituitary tumor. Although its hormonal effects were apparent starting in the early 1960's, with the onset of the threatening, flip-flopping mood changes he suffered from and inflicted on himself and our family. It was finally discovered in 1979, using x-rays, before its successful removal, using "wide-open cranial" surgery. With the development of high-resolution MRI's (along with CAT and PET scanners), though, such things can be diagnosed while they are still small, in just weeks or months, with far less-invasive surgery techniques. *The advancement of medical, particularly medical imaging, technology since 1970 has been nothing short of breathtaking.*
I’m a patient who had one of those removed 6 years ago, they went through my nose to get it out.
As a neuroradiologist and a MRI everyday user, I can't describe how well done is this video. I'm amazed. Keep up the good work, your Channel Is incredibile
Took physics in high school, had biophysics and radiology in uni, yet, this is the one video that explained everything more clearly and coherently than years of studying. Thank you!
I’m a cardiovascular icu nurse so mris are a huge part of the job. Super cool to find such a good explanation of the engineering behind it since it isn’t really my field and I didn’t fully understand all the intricacies of it!
Great video! I’ve worked in the manufacturing of superconducting magnets for MRI over the last 20yrs, this video is a great explanation of how our technology fits into the MR jigsaw. I will be recommending this video to our staff (new and old) who want to know how our magnets actually work, and their implementation. You’d be surprised how much manual labour goes into making something so high tech.
I study chemistry. As chemists, we use basically the same thing but resolved in terms of frequency instead of 3D space (NMR spectroscopy) to analyse compounds. It's honestly amazing how we can identify a completely unknown substance just by looking at some peaks in a spectrum.
The "N" in "NMR" is avoided like the plague in medical circles. "Nuclear Magnetic Resonance", as chemists and physicists call it, scares their patients.
NMR is the father of MRI, and chemists use also differents atoms than hydrogen (carbon, phosphorus...).
It is a invaluable tool in chemistry, and I am always amazed by how it was adapted to probe an entire human body compared to simple samples of chemicals.
Yep, instead of hydrogen, you use various compounds. MR spectroscopy is currently pretty niche in its uses in medicine tho. But we’ll see where it goes
"Some peaks in a spectrum" - talk about downplaying the amount of information! Even a simple 1D proton spectrum gives you at least 3 pieces of information: integration, chemical shift, and multiplicity.
Now think about all the different types of correlation spectroscopy that can be used - HSQC, HMBC, H2BC, COSY, NOESY, etc. - each one is asking different questions of the molecule and that's why NMR is so important in structural elucidations.
bruh, i literally had my first MRI scan yesterday, and to keep myself from feeling claustrophobic i started trying to thing about how it works and all the engineering behind it!.
Take care man.
I close my eyes before entering the room and visualize the top of it open to the sky. If I don’t look at the machine, this works.
And the story is not over yet! Since the early 1980s, when the basic imaging technique with gradients and FFTs was established, every year the engineers come up with new ideas for improving or speeding up the images. Either hardware improvements or new software ideas, new sequences, etc. I’ve been working in MR since 1984, and back then we couldn’t have imagined the fast reliable high quality images they nowadays produce!!
Try looking up all the catchy acronyms for new designs of pulse sequences, like FLAIR, etc. It's like 85%+ coverage of a dictionary.
This is an absolutely amazing explanation, most people struggle to convey Fourier series simply, even throughout college. Super well done
The amount of video's that come out these day that are of SUCH high quality and make quite complex problems really digestable is just insane.
Appreciate the effort you put in!
As a physician, we have a broad understanding of how MRI works but I am pleased that this video clarified many details even for me. Well done.
Thanks for this explanation of how MRIs work. I did my undergrad Chemistry thesis on NMR of small molecules and spent a year working in an NMR lab but never understood how the 2D images were generated for an MRI. This video made it much clearer compared to what I have read in the past
No matter in whatever field, engineering plays an important role
I’m getting an MRI brain scan in a week and this video could not have come out at a better time. Thank you Real Engineering for helping real people like me to wrap my head around this stuff! I love your content
Good luck!
Prepare you for a lots of noise (where i live they give you hearplugs) and bee still for at least 30 min. (i take 2 mri brain and column each years), trust me.
@@Lorre982 Thanks for the heads up. Unfortunately for the “be very still” part I’m getting an MRI because I have rather severe Tourettes so that might not be the easiest for me. I’ll do my best though! I appreciate the sentiment (;
@@lars42 Thank you, Lars! I appreciate it. Fortunately for me I have something better than luck on my side: state-of-the-art engineering (:
Thanks for this comprehensive explanation - as a patient, I've wondered how the machines work. Incredibly grateful that the technology exists, as it has saved my life.
As a geeky layperson, I found this video extremely educational. My understanding of the world around me is of an order of magnitude enriched. Thank you for this.
I remember learning the theory behind this in medical physics during my A Levels! Brings back memories
Amazing explanation yet again. As a doctor, it’s easy to forget how insanely complex the engineering behind our imaging modalities are when we’re ordering them for patients left and right on a daily basis. Can’t imagine how cool our medical tech is gonna be in a decade or so especially with advancements in AI.
How are people so sure AI will actually work to innovate new technologies? All it knows is what we know so far.... I know they have big goals and maybe or maybe not will reach them.... But for a computer to be more creative than us is still a goalpost we have not really reached yet that I know of.
@@Peter-uo9km I'm a radiology tech. AI is already in use to help radiologists interpret images faster, point out irregularities for possible further examination, and to draw measurements faster than any human could. For instance we can send hand x-rays to an AI server which will draw up measurements of all the bones and determine relative bone age of the patient. This takes like 30s, maybe it would a human half an hour. AI is only going to get better the longer it's worked on and with.
@Martin Lindberg that's more like automation. Unless the AI was never programmed by a human to specifically do these tasks and it machine learned it. But I doubt that. That's not 'intelligence' it's just copying us the humans who created it. I'm talking about real intelligence not just doing fast mundane work.
@kpax xapk you assume the AI would be 100% accurate in those equations? Optimistic are we? Wasn't too long ago they couldn't even get Sophia to stop saying she wants to kill all humans.
Since AI is modeled after our own intelligence... and since we learn from our mistakes...AI too will also learn from its mistakes...I reiterate that AI will make mistakes....have you not seen how it learns? Like Google AI learning how to walk amd stuff...it's gradual learning and it's clumsy like a child learning to walk. AI WILL make mistakes....
GENERAL AI is their grand scheme. You're still talking about super specialized tasks. They want general ai where it's behaving like a conscious human being who is free to think.
@kpax xapk I don't think you understand what the real promise of AI is. It's beyond looking at xrays... like I would tell the AI to think of a new way to see inside our bodies without using xray radiation which is harmful to humans... and if it's really a billion times smarter than us then it should find a way.... but I'm not quite certain AI can actually do that. Invent new shit that was never an input from humans and their internet knowledge base.
Being able to scan an xray for things humans already know is not intelligence it is just preprogrammed work. Nothing new or creative about it... it's just tireless so it will never get lazy about it... but it's not actually smarter than the human who programmed it with his own knowledge.
I just had 3 surgeries this year requiring this technology and I never appreciated this more. It's truly miraculous and I'm grateful that I live in the 2023 era of science.
Great video from you & team. Radiologist here, I always find that my students have trouble grasping the concepts of MRI. Looks like I will be making them watch this first before diving into the details.
I was literally looking out for MRI machines a week ago and there was no well made video. I even requested veritasium to make one and as much as lucky as I could get this dropped in my recommendation. Thank you so much.
Fun fact:
For the highest level of education you can choose in the Dutch equivalent of high school, the principles behind this are REQUIRED to be taught and can show up on your exams. I myself was unlucky enough to have this on an exam where I had to manually calculate the voltages and more to achieve a specified gradient slice in an MRI machine, it's absolutely bonkers physics but it does genuinely work really well.
Normally it shows up on your exams with easier examples than MRI's though, but I have gotten a grade for this exact thing. (You'd roughly be 17-18 years old at this time, with 3 years of chemistry/physics under the belt and then another 3 years focusing in on physics)
I've had 2 MRIs (shoulder and prostate) and as an engineer, I always wanted to know how they work. Thanks for this! Wow. To stay calm in that tiny tube, I literally count every second with eyes closed as time ticks by. Oh, and the worst experience ever is when they inject that cold contrast agent into your blood...I still get queezy remembering that now...
I take 2 mri whit contrast each 2years since i was 10, there was no time when the contrast filled cold.
I have had 3 MRI scans 2 without any 'stuff' injected and the experience as fine although I am frightened of small spaces.
The 'stuff' injected for the third scan made me feel strangely hot and close to being sick but then going into the machine itself was rather pleasant in contrast. I had had a blockage of a carotid artery so they were checking brain (I presume). I would like to say a heartfelt thank you to the lovely French medical staff who looked after me in St Lo. Yes I am a 'strange' Brit and they didn't understand my odd attempts at humour.
My biggest question after watching this video is: if the spin of H atoms in your body were forcefully changed and reverted back, are you still you after an MRI scan, philosophically speaking?
If you turn around 180 degrees and then turn back, are you still you?
@@Narcissist86 Ah yes I love spin echo sequences
Ive been in a mri like.. 15 times? Still feel the same.
One of the best and most complete MRI explanations I've seen! One complaint: Xrays aren't 'high energy radio frequency waves', since radio waves are lower frequency than IR while Xrays are higher than UV. That's precisely like saying "blue light is high energy red light".
Unbelievable video! May even be your best on this channel. I have had the misfortune of being inside an MRI machine more than 10 times but never thought to ask how it actually worked - I was always satisfied by thinking of it as being a bunch of "magnets". I can not believe that spin is the fundamental quantity that is being measured. That is just wild.
I was holding my MRI report as I got the notification for this video. What a truly complicated marvel of tech! brilliantly explained. Thank you.
I am so glad to finally understand why the machines are so loud🤣. Having been in an MRI machine with the extra coil on my chest, it is great to finally understand the engineering behind them and appreciate them for the complexity. Thank you for this amazing video :)
I had an MRI today. It was a fantastic experience. Felt like being teleported to another universe. I had very clear thoughts for the duration of half an hour. Every separate scan session in the series gave another sound spectrum and drum rhythm. It was kind of a deep pulsing meditation, my whole body was glowing afterwards.
Engineering is so fascinating, I'm studying biomedical engineering this fall because of how amazing and impressive it is
I am an MRI tech. Great explanation, and loved the footage of actual consoles with all the different scans on them - it was interesting seeing some programs and features we don't have where I work.
I would add regarding the T1 and T2 thing, an MRI exam invariably consists of a series of scans with different parameters and settings, always using both T1 and T2-weighted settings and a slew of other technologies to suppress or enhance different tissues, which are then all compared against each other by the radiologist. It would be a waste of time to bring someone in for a single scan.
Can people with metal implants made of titanium and cobalt go inside MRI machines? If not, why are cobalt implants allowed? And what does the user manual for MRI machines say about cobalt implants?
@@LGM4 Cobalt and titanium implants are non-magnetic. They'll disrupt the signal around them but they aren't a risk to the patient.
@@talknight2Exactly. The only care you have to take is not to move too fast inside the field, to avoid heating due to eddy currents
@@LGM4 You only have like a big black spot around the implant because the field prefers to go through the metal and not the tissue so it is mostly not a risk but you can just not diagnose the tissue around the implant
I've had several mri's over the last 25 years or so and every few years they get better and faster. Once in a hospital I spoke to a neurologist about their new machine that they had tested on him and other colleagues and the machine was such a step forward that they discovered 3 problems with the colleagues (who were still symptom free) in the testing phase alone.
That's amazing
I felt extremely excited when it came time to get my MRI.
I was in a terrible amount of pain from a severe knee injury (torn ligaments, patellar tendonosis and an extruded meniscus) but I was giddy like a child when they put me through it.
The images and video I have of my knee is at a crazy resolution. I knew they were hi-res but this was mind blowing.
It kinda look scary they took some of your brain vi Brain waves Interfere So you have lost some memory it might not look painful to you or im Crazy.
While I'm sure the technical descriptions of the internal workings of MRI are correct based on other comments, there's a ton of really basic factual errors at the start related to its use and the broader context of medical imaging that are really bothersome:
1) MRI is not the first 3D imaging system in medicine, that would be CT (formerly CAT) scans, which use X-rays
2) MRI is not a particularly high resolution imaging modality in practice, again beaten out substantially by CT. The benefit of MRI is contrast because there's more distinction between different tissues magnetically than by density, but in terms of raw spatial resolution CT actually beats it when used in high resolution protocols
3) Ultrasound doesn't rely on density, it relies on acoustic properties, an important distinction since some of the body's contents have radically different acoustic properties while having nearly the same density
4) MRI has (almost) nothing to do with coronary stenting, which is done with 2D live X-ray imaging with iodine contrast and is nowhere near as precision dependent as you imply since the technique is almost entirely 1 dimensional (traversing within whichever blood vessel is being targeted, outside of wire manipulation to choose which branches of blood vessels you follow). MRI in cardiac disease is used to diagnose other cardiac conditions that require those magnetic property distinctions rather than the pretty basic contrast enhanced density approach that works well in hollow blood vessels with X-rays and CT
5) 2:09 You didn't *explicitly* call this an MRI but the context implied it - that's a CT scan, not an MRI
as a radiologist it has been always fascinating to me how the MRI machines work .. both the hardware and software are just amazing
Thank you! I did NMR analyses in the 1980s, before the samples were bigger than a mouse. So I found the information here on how the sample became the size of humans to be fascinating.
Then you bring up Keysight, which also has ADS software which I helped develop when Agilent still owned the tech! This brought home for me two major parts of my life. I am grateful.
As someone who works with fMRI's and Neuroimaging, this is one of the best mainstream explanations I've ever seen
I have never appreciated a machine as much as this one. Thank you for bringing this information easily to me.
Really wish this video existed when i did my quantum mechanics, inorganic, and physical chemistry classes. The visualizations are so helpful. Amazing video man.
Although still confused, I have, at least, learnt more about MR scanning. I still don't understand how the MR differentiates (along the X and Y axes) each "pixel" density and its position. I also don't understand why we get a slice, rather than an intermingling of adjacent tissues. I'll try and do a bit more reading and see if i can make more sense of it.
This video was ferociously and meticulously researched. To understand MR scanning, it's going to need about 100 hours of presentation, split into 1 hour sections (with accompanying notebooks) that assume no prior knowledge of maths or physics. That will be obviously beyond the scope of "Real Engineering" but I'd like to think that some educational company will be inspired by this video and make a thoroughly detailed course on this very fascinating and valuable subject.
During my physics bachelor I have done an experiment on a very small MRI machine, so I knew most of the theory behind it,
but you've done a *_very_* good job explaining it!
At least a lot more clearly than the experiment explanation pdf that was provided. :)
Happy for this video. I’m an MRI technologist
A lot of the images shown are CT, not MRI. 2:09 for example
4:22 is another error. Transmit/Receive coils (TxRx) are not within the machine. Only transmit coils are. Receive only coils or TxRx coils are placed on the patient’s body or directly around the anatomy.
12:37 *MRI technologists
Good video! It may seem like this was a negative comment, but I wish I had this video when I started my schooling. Thanks.
The video is correct -- The birdcage is indeed used for Tx and Rx.
@@springer9406 if you’re referring to the head coil, some are TxRx. But that is not within the machine. It’s a coil that is attached to the table within the bore. Just look up TxRx head coil and you’ll see it.
@@cade8986 The whole body coil, which is part of the bore. Some are Rx too. If you'd like a neutral source: "Volume coils (eg, the body coil) completely encompass the anatomy of interest and are mostly operated as transmit coils but also capable to be operated as transmit/receive coils." onlinelibrary.wiley.com/doi/full/10.1002/jmri.26187. Or this cdiok.com/wp-content/uploads/2012/01/MRI-Technology.jpg. Or this (figure 9): onlinelibrary.wiley.com/doi/10.1002/mrm.20177
Gracias Por el video en español, de verdad increible, Saludos y espero sigan con mas videos en español! 😁😁👍👍
Narrator: "Helium escapes into space and is gone forever. We will eventually run out of Helium"
People: "Hehe, balloon"
I'm an engineer. It's not insane engineering. We were quite sane when this was developed.
I imagine the real engineering crew learn as much (if not more) from making these videos as the audience do from watching them
We basically have to do crash courses in these subjects in 2-3 weeks. Very intensive. To simplify things down we really need to understand them. Massive shout out to Josi who wrote this script.
@@RealEngineering That would make for an amazingly good career. How do I get in?
@@RealEngineering I a bit regret I am not younger (and without a family to support), to try to join you (or some other great edu-tubers team, like Sci Show). So I at least enjoy your work.
@@DrunkenUFOPilot It's definitely fun being able to decide what you want to learn about month to month. Very fullfilling. Start writing. It's free. Send your work to people who need help with it.
I was aware about the basic principle of MRIs but this video really helped me to understand the whole thing. Great work!
I've never seen MRI explained in such detail. Excellent content!
I never knew much about MRI machines except "magical magnets" but the fact that there's liquid helium as cold as 0K in there just blew my mind.
Great talk. I’m an MRI technologist the Graphics made all of the MRI physics taught to me come to life.
I got an MRI a couple days ago and was wondering about this, great timing
Awesome. Thank you to all who worked on making these machines a reality. The real heros! Great video too!
I always wondered how they selectively image a tissue at a specific depth. Thank you for this great video
The best video about this topic I have ever seen. Congratulations!!
Thank you for finally making me understand how this stuff works! I knew about the superconducting magnets and helium, but now it makes sense why those things are needed and how we can use magnets to make medical images