Error Corrections and clarifications: 1:22 - When we morph the many spins into one M vector, the animation shows M precessing about B. If the spins are in equilibrium then M would be colinear with B and no precession would be occuring, but here I animate as if our spin sample had just experienced excitation and thus M and B are not aligned allowing for precession - defnintely not clear in the video (thanks to @SCramah13!) 20:19 - Δω is not 42.578 kHz, it's 4.2578 kHz (thanks to @johnclarky4375 for catching this!) 36:06 - My labels for 'phase encoding' and 'frequency encoding' are swapped. (thanks to @timangoes for this one!)
I am here because my son is in MRI ! At a university. So, to understand his passion . I'm learning from these videos. The language and understanding of MRI. Thank you . 🎉
I'm an MRI Field Service Engineer and these videos are the best hands down, the visuals you create really connect to what is happening vs. reading a book. BRAVO!
Genuinely amazing how much commitment you have to finishing such an insanely high quality, detailed, easy to understand explanation on a relatively obscure topic. This is, across all sorts of topics, really one of the best RUclips channels I've ever seen for explaining complex topics in an understandable way.
omg dude you are a legend. My journey with your channel started as I was a med student interested in spin dynamics, now I am a noob radiologist and just recently starting to work with MRI, you can not imagine how happy it made to see this video in notifications. Your videos are best explanation to MRI physics anywhere on internet, I just love it!
The greatest video series I've ever seen. Your animations and eloquent explanations are out of this world! If you happen to make a follow-up, I'd love to see discussions of fMRI and BOLD analysis, since these are my areas of application. Cheers!
Your series on MRI is absolutely brilliant, and this episode in particular is especially amazing considering the depth you go into whilst maintaining a clear and coherent train of thought. So happy to see you still uploading :D
You are a legend! I just started my new job as an MRI data analyst, and I absolutely love your videos-they're amazing! I’m currently creating an onboarding document for future employees and students in our group, and your videos will be at the top of the list for everyone to learn MRI easily. Thank you, thank you so much for creating such awesome content!
Arguably, one of the best (if not THE best) Educational Series I have ever seen in my entire + 50 life and BSc Industrial Engineering background. Its painstaking attention to detail, the beauty and virtuosity in all the curated visuals, its deep-dive yet easy-to-understand explanations... a top-level Educational Masterpiece, in a word. My sincerest congratulations and admiration thus to the generous and talented Artist behind this Series. Just can't wait for the next one!
That pile of quantum physics, mathematics and who-knows-how-many-other-fields is so insanely cool, it made me freakin' smile... Thanks a lot for doing such a huge work! It is especially valuable, because primary audience here is very narrow and there kinda no stimulus other than authors own interest and respect for beauty of subject!
I build a company in the ultra low field MRI space and your videos are an invaluable source of educational content for me and my teammates! Thank you so much, can't wait to see what's next. Diffusion please :)
As a graduate student in medical physics, I just have to say this is one of the best lecture series I've come across (on par with 3Blue1Brown). Thank you for enhancing our learning experience. Hope to see more videos in the future!
You are an absolute genius in vulgarization and on top of it you are extremely talented for making animations which help tremendously in the visualization and understanding of such advanced subjects. Thank you so much for continuing this series of video after the years.
This is so far the best MRI lecture I have ever found. I am a first year resident in radiology and have been struggling to understand the complicated concept in MRI for a long time. These fantastic videos really helps me out! Thanks a lot dude!
Does radiology residency really require this much depth in image reconstruction? I thought this much depth particularly applies mostly to medical physicists/engineers while relevant info to rad residents is signal formation and the anatomy etc.
Please, keep this up. This is one of the best channels I've ever seen (I would go as far as saying that you should explain more general stuff: for example, your explanations of the Fourier transform are one of the best I have seen on the internet)
Best videos. Have watched this video 6 times by now. I am starting to get it. The maths part between 30-45 minutes needs some work from me to fully get.
This is the best series on MRI and I'm thankful for you keeping this series going even after years. I have recommended your videos to many radiologists and technicians. Keep going, you have so much potential with educational content.
I studied electrical engineering and am now doing a phd in RF analog electronics. The 3rd and 4th videos of this series give an idea of some of the math I encountered in my studies. Before encountering this video I wasn't aware of the connection between MRI and RF technology. I'll have to check out some research on the transmitter and receiver section of MRIs.
Awesome job! This has helped my studying significantly. I am following a slightly broader course on MRI but this video provides much of the intuitive understanding you need to grasp a lot of related concepts. Wonderful animations as well!
Hey there! I am eagerly awaiting your next installment, particularly in the topic of different imaging modalities, as I've been looking for a comprehensive, intuitive approach to understanding them. This way I can internalize the modalities/sequences by knowing the underlying mechanism, making it much more satisfying and memorable. Thank you for everything you've been doing. You found an excellent method to explain what actually happens when someone undergoes a scan.
Thank you for these videos - much clearer explanation than I was able to find elsewhere! One suggestion: while the background music makes things entertaining for our neurons, there were times when I genuinely found it distracting as I tried hard to focus on the details - sometimes a less shiny video is easier to understand!
@36:12 , it seems that the labeled exponentials for phase encoding and frequency encoding are flipped? based on the preceding sections, frequency encoding should be the one with time dependence and the phase encoding term should be the one where the gradient is only turned on for some time tau
Thank you very much for the insightful lecture! Could you kindly share the solutions to the exercise presented at the end of the video? I'd like to cross-reference my responses with the correct ones. Thanks again!
Oh wow. You're absolutlely right, the bandwidth is 4.258 kHz, not 42.58 MHz. Good catch! A bit embarassing that not once in making the slide, writing the script, and recording it did the abnormally high bandwidth seem odd, ha.Thanks!
Amazing content! Are there any plans to make the code used to generate these graphics and animations available on github? Would love to be able to play with the simulations to get a hands-on feel for MRI signal acquisition.
On the 28:47 you tell about explicit Fourier transform. I tried to understand it and it seems you are not mentioning something. Fourier transform (and the same is true for RFT) of function with two variables should be a function with two variables - F(u,v) = fft(F(x,y)) or f(x,y) = rfft(F(u,v)). In your example, you have S(t) = rfft(M(x,y)). How is it possible? Where the second variable in S(t)?
I found the answer in one book. They say "let's remember that our gradient in y-direction is a function of repetition Gy=Ginit*n. After substitution we get that our signal is function of time and repetition S=f(t,n) and n*tau=pseudo time, where tau is a time of application of Gy"
Yes, exactly, the signal recorded is a function of both time S(t) and k-Space location S(kx,ky). The connection between the two come from the gradients which determine our kx, ky locations and are themselves functions of time. Hope that helps!
Could you please clarify whether or not the net magnetization vector should be visualized as a precessing vector @1:35? My understanding of NMR is that this is NOT true (because all of the x-y components of the magnetic moments of all individual nuclei effectively cancel out)...and if there is no net magnetic moment in the x-y plane, then all of the net magnetism is directed along the z-axis, which means the torque from the perspective of the net magnetic moment vector is 0 (i.e. no precession about the z-axis). However, you seem to suggest that this is the case with your language: "collectively, the quadrillion of spins in space sum to a nuclear magnetization vector M..." and then you provide a picture of this presumed net magnetization vector M (purple arrow) and depict it as precessing. Isn't this incorrect?
You make a good point! I should've made it clearer in the video, but when the M and B vector animation appears it is immediately following excitation, not in equilibrium. You are certainly correct that in equilibrium, M would be stationary pointing in the z direction and no precession would be apparent. An rf pulse (which I didn't show) is necessary to tip M away from B and see precession. Thanks for this observation; comments like this are helpful so I can see where gaps exist in my presentation! Cheers
So does MRI recieves the signal from each say cm2 of the 2nd slice, and then it produces the image, or it just recieves the overall sum and use FFT to differentiate. If it does how does it know the exact location from which the signal is coming? Please answer this. Btw I dont normally comment on videos, but man this is exceptional content. Its very sad that useless channels get 10 or more million subs and views but people like you get less. It shows the reality and useless aims of people's mind. I just dont know how to show gratitude. Hands down one of the BEST youtube channels i have seen. I have a doubt i dont know if you see this but here it goes.
The MRI detects the vector sum signal from all excited spins which the gradients encode, then the FFT transforms to an image. In the gradient echo sequence, the process is done one slice at a time. The scanner does not 'know' where the signal is coming from - but the different frequencies contained in the detected signal are known to have originated from specific locations. Hope that helps? Cheers
@@thepirl903 Ok, but it still did not answer my question (or that im dumb!) But let me clarify my question first. Lets keep it simple and say from different tissues you get signals of different frequencies. Now, suppose we are imaging the kidney. From the renal medulla, a short frequency signal is coming, and from the cortex a long frequency signal is coming. Using FFT you can seperate the frequencies and know that you are scanning two different tissues (lets juts say medulla and cortex is the only parts of kidney). For eg how does it know than renal medulla beneath or below the renal cortex?. How does it correctly 'position' the different tissues or parts in an organ so complex as brain? I understand you will get the information of how many different tissues are present, but how will you correctly postion that different tissues. For another example, im saying to you, suppose you are an artist, that there are 4 different types of tissues in an organ, apart from you use different colours for different tissues, how will you correctly draw the organ with correct position. Hope that is not lengthy!.
Hmm, ok I think I understand better your question, but we'll see. So the location of the separate tissues signals are indeed teased apart by the FFT in order to localize them in the image. The positions of tissues in space are directly mapped to frequencies by the larmor equation Δω=γGΔx. So the individual tissue signals can be precisely mapped in space. The MRI of course doesn't see 'tissue', it only sees protons (hydrogen nuclei). The way tissues are differentiated comes from their different nuclear relaxation properties (T1,T2, etc.), thus the brightness of a voxel reflects these contrast mechanisms, which in turn reflect the different properties of the tissue. Not sure if that helps at all?..
@@thepirl903 Thanks, that helps, so in short, larmor equation maps it. Btw I really appreciate the long detailed videos, exactly what i love. I dont know how to help or say thanks. Im just 15 years old, and you are really an inspiration! May I know what is your proffession?
Hi, I have one question about reciving bandwith, normally when we talk about bandwidth, let's take light for example, the spectrometer may collects 500-600 nm light when it goes to 400-800 nm the bandwidth is higher. Like you excited the slice, you have a band width, because different spin is in different frequency. but when you do reciveing, i suppose the frequency is talking bout K, However no mater you play longer time weak gradient or short time strong gradient the path in k space should be same. so the "bandwith" should be the same right?
The excitation bandwidth is the frequency span which will excite a certain 'distance' of spins in a given magnetic field gradient strength. When encoding during receive the bandwith is chosen so that while applying a gradient that no spins will be precessing faster than nyquist. If a stronger gradient is played then spins will precess faster so the bandwidth will need to be higher. The k will be the same (spatial frequency), but w is not (temporal frequency). Does that help? Great questions!
Would it be correct to say that M(x,y) is, in reality, a function of time due to the ensuing transverse magnetization relaxation process? How does this alter the formulation of the fourier transform application?
Indeed M(x,y) is a complex function of time. Our model for the GRE here assumes that M(x,y) is static. This of course isn't true, but on the time scale of a readout, these effects can generally be safely ignored if the readout is quick and T2* is long (not always the case, but fair for most biologic tissues). Similar to taking a photograph, we find a balance between our subject's motion (analogous to T2* decay) and the camera shutter speed (analogous to readout time) to reproduce the best 'snapshot' of M(x,y). Hope that helps
I use a Siemens Prisma (with 80/200 gradients) and we recently acquired a next gen Siemens Cima which has 200/200 gradients. Does the benefit of the higher gradient strength come from better separation of the phases at each end of FoV, thus yielding greater contrast?
There's a few benefits to higher maximum gradient amplitudes including faster traversal of kSpace (good for collecting signal of short T2* samples, and rapid imaging of the heart) and ability to get to the kSpace edges faster (better resolution), better diffusion weighting (can acheive higher b-values in same time), fewer susceptability artifacts. Though higher gradients strengths can be louder and are more likely to cause peripheral nerve stimulation in some sequences. Cheers!
Excellent presentstion! Thank you fir insughtful dilligence. Pleasexdo axsegemnt in the elecyronics and relsted base software to collect and cinvert the raw dataxand cinvert to image data. Details of ADC sample frequency, how RF frequencies are shifted (mixers), RF PA, LNA and related filters. And maybe post a biography on yourself.
Yes, I will do this at some point. That explanation isn't my favorite of the series, and I think it confuses people more than anything. Probably the best resource for QM treatment of flip angle (imho) is Slichter's Principles of Magnetic Resonance section 2.6 (www.google.com/books/edition/Principles_of_Magnetic_Resonance/pWzrCAAAQBAJ?hl=en&gbpv=0)
ruclips.net/video/yaa13eehgzo/видео.html , ruclips.net/video/rbu7Zu5X1zI/видео.html these are from other creators that do similar if not better animations similar to this style
Hell no. Keep making these MRI videos because the subject is vast and there's much to learn. CT and X-rays had their eras...in the 90s. Now its MR Time!
as someone who already knew the basics of mri that magnetic field strength and phase are varied along the axes to encode location, I can say that this video did nothing but obfuscate the concepts. it seems you are more interested in making fancy looking equations than explaining the concept, clearly exemplified by using an integral to denote a simple linear equation for phase as a function of a linear dimension. i will have to come back to see if this video is useful as a way to solidify knowledge one already has, because it certainly is a terrible way to learn this stuff
Error Corrections and clarifications:
1:22 - When we morph the many spins into one M vector, the animation shows M precessing about B. If the spins are in equilibrium then M would be colinear with B and no precession would be occuring, but here I animate as if our spin sample had just experienced excitation and thus M and B are not aligned allowing for precession - defnintely not clear in the video (thanks to @SCramah13!)
20:19 - Δω is not 42.578 kHz, it's 4.2578 kHz (thanks to @johnclarky4375 for catching this!)
36:06 - My labels for 'phase encoding' and 'frequency encoding' are swapped. (thanks to @timangoes for this one!)
I am here because my son is in MRI ! At a university.
So, to understand his passion .
I'm learning from these videos. The language and understanding of MRI.
Thank you .
🎉
I'm an MRI Field Service Engineer and these videos are the best hands down, the visuals you create really connect to what is happening vs. reading a book. BRAVO!
Genuinely amazing how much commitment you have to finishing such an insanely high quality, detailed, easy to understand explanation on a relatively obscure topic. This is, across all sorts of topics, really one of the best RUclips channels I've ever seen for explaining complex topics in an understandable way.
omg dude you are a legend. My journey with your channel started as I was a med student interested in spin dynamics, now I am a noob radiologist and just recently starting to work with MRI, you can not imagine how happy it made to see this video in notifications. Your videos are best explanation to MRI physics anywhere on internet, I just love it!
Thanks ! It's fantastic you are still completing this series after years...looking forward to more from you
You Sir, are a Scholar and a Gentleman!
The greatest video series I've ever seen. Your animations and eloquent explanations are out of this world! If you happen to make a follow-up, I'd love to see discussions of fMRI and BOLD analysis, since these are my areas of application. Cheers!
Your series on MRI is absolutely brilliant, and this episode in particular is especially amazing considering the depth you go into whilst maintaining a clear and coherent train of thought. So happy to see you still uploading :D
You are a legend! I just started my new job as an MRI data analyst, and I absolutely love your videos-they're amazing! I’m currently creating an onboarding document for future employees and students in our group, and your videos will be at the top of the list for everyone to learn MRI easily. Thank you, thank you so much for creating such awesome content!
Arguably, one of the best (if not THE best) Educational Series I have ever seen in my entire + 50 life and BSc Industrial Engineering background.
Its painstaking attention to detail, the beauty and virtuosity in all the curated visuals, its deep-dive yet easy-to-understand explanations... a top-level Educational Masterpiece, in a word.
My sincerest congratulations and admiration thus to the generous and talented Artist behind this Series. Just can't wait for the next one!
I am really looking forward to MRI part 5. All the possible topics you mentioned at the end sound fascinating!
Stimulated echos and EPGs would be nice
this video is a fantastic for anyone interested in learning about MRI technology. Great work!
That pile of quantum physics, mathematics and who-knows-how-many-other-fields is so insanely cool, it made me freakin' smile... Thanks a lot for doing such a huge work! It is especially valuable, because primary audience here is very narrow and there kinda no stimulus other than authors own interest and respect for beauty of subject!
I build a company in the ultra low field MRI space and your videos are an invaluable source of educational content for me and my teammates! Thank you so much, can't wait to see what's next. Diffusion please :)
dude you created account only so i could pass the NMR classes, real superhero
Incredible video, really impressed by the clarity of explanation.
Top notch lecture, one of the best out there on K-Space and image formation!
Quality and detail are INSANE! Thanks so much for sharing
As a graduate student in medical physics, I just have to say this is one of the best lecture series I've come across (on par with 3Blue1Brown). Thank you for enhancing our learning experience. Hope to see more videos in the future!
You are an absolute genius in vulgarization and on top of it you are extremely talented for making animations which help tremendously in the visualization and understanding of such advanced subjects. Thank you so much for continuing this series of video after the years.
This is so far the best MRI lecture I have ever found. I am a first year resident in radiology and have been struggling to understand the complicated concept in MRI for a long time. These fantastic videos really helps me out! Thanks a lot dude!
Does radiology residency really require this much depth in image reconstruction? I thought this much depth particularly applies mostly to medical physicists/engineers while relevant info to rad residents is signal formation and the anatomy etc.
Thank you for this wonderful series. I hope to see more from you. I loved how you didn't simply neglect the physical coil design.
Your lecture series so far is just outstanding! Please do not stop creating this content! This is a true gift to the MRI community!
Please, keep this up. This is one of the best channels I've ever seen (I would go as far as saying that you should explain more general stuff: for example, your explanations of the Fourier transform are one of the best I have seen on the internet)
Best videos.
Have watched this video 6 times by now. I am starting to get it. The maths part between 30-45 minutes needs some work from me to fully get.
Best mri stuff on yt.
Yooo I immediately thought of string art being like CT scanners
This is the best series on MRI and I'm thankful for you keeping this series going even after years. I have recommended your videos to many radiologists and technicians. Keep going, you have so much potential with educational content.
I studied electrical engineering and am now doing a phd in RF analog electronics. The 3rd and 4th videos of this series give an idea of some of the math I encountered in my studies.
Before encountering this video I wasn't aware of the connection between MRI and RF technology. I'll have to check out some research on the transmitter and receiver section of MRIs.
Awesome job! This has helped my studying significantly. I am following a slightly broader course on MRI but this video provides much of the intuitive understanding you need to grasp a lot of related concepts. Wonderful animations as well!
Amazing! The best video for MRI introduction on RUclips.
Hey there! I am eagerly awaiting your next installment, particularly in the topic of different imaging modalities, as I've been looking for a comprehensive, intuitive approach to understanding them. This way I can internalize the modalities/sequences by knowing the underlying mechanism, making it much more satisfying and memorable. Thank you for everything you've been doing. You found an excellent method to explain what actually happens when someone undergoes a scan.
I have 1 year left to obtain my MRI certification. If you keep explaining MRI the way you have been in this series.... I think I have a shot!
Thank you for these videos - much clearer explanation than I was able to find elsewhere! One suggestion: while the background music makes things entertaining for our neurons, there were times when I genuinely found it distracting as I tried hard to focus on the details - sometimes a less shiny video is easier to understand!
Someone give this man a raise
Best video series!
Please make more videos like this
Thanks for the series, very well made and extremely useful.
Diffusion tensor imaging and fMRI next for sure!
You’re the best for MRI lecture ❤🎉
This is really wonderful stuff! I really appreciate how much work you're putting into these. Thank you!!!!!!
Incredible videos and series! You're the 3B1B of MRI! If you can do it, the video about CT imaging would be great, also.😅
The best!!!! Great content and visual dialogue. Use to study MRSE exam.
@36:12 , it seems that the labeled exponentials for phase encoding and frequency encoding are flipped? based on the preceding sections, frequency encoding should be the one with time dependence and the phase encoding term should be the one where the gradient is only turned on for some time tau
Arg, you are correct. I did swap those. Thanks for catching!
Super fantastic mri series! Keep going ❤❤
Bravo sir! Its like 3blue1brown for MRI. Great work!
Thank you very much for the insightful lecture! Could you kindly share the solutions to the exercise presented at the end of the video? I'd like to cross-reference my responses with the correct ones. Thanks again!
Thanks for your beautiful videos.
Yes, please make videos on CT and reconstruction.
Yes!! He’s back!
nice, i will do my phD thesis on your video....
Minute 21:15, when I calculate delta omega i get 4.2578 kHz. Where am I missing the faktor 10?
Oh wow. You're absolutlely right, the bandwidth is 4.258 kHz, not 42.58 MHz. Good catch! A bit embarassing that not once in making the slide, writing the script, and recording it did the abnormally high bandwidth seem odd, ha.Thanks!
Thank you for this masterpiece!
Great work!
Amazing content! Are there any plans to make the code used to generate these graphics and animations available on github? Would love to be able to play with the simulations to get a hands-on feel for MRI signal acquisition.
see you people in another one year
I will hate who ever will push dislike on these videos, they might be my greatest enemy.
Amazing stuff!🎉
On the 28:47 you tell about explicit Fourier transform. I tried to understand it and it seems you are not mentioning something. Fourier transform (and the same is true for RFT) of function with two variables should be a function with two variables - F(u,v) = fft(F(x,y)) or f(x,y) = rfft(F(u,v)). In your example, you have S(t) = rfft(M(x,y)). How is it possible? Where the second variable in S(t)?
And THANK YOU FOR THE LECTURES!!! With videos and animation mathematics starts becoming less complicated. Really helpful
I found the answer in one book. They say "let's remember that our gradient in y-direction is a function of repetition Gy=Ginit*n. After substitution we get that our signal is function of time and repetition S=f(t,n) and n*tau=pseudo time, where tau is a time of application of Gy"
Yes, exactly, the signal recorded is a function of both time S(t) and k-Space location S(kx,ky). The connection between the two come from the gradients which determine our kx, ky locations and are themselves functions of time. Hope that helps!
Could you please clarify whether or not the net magnetization vector should be visualized as a precessing vector @1:35? My understanding of NMR is that this is NOT true (because all of the x-y components of the magnetic moments of all individual nuclei effectively cancel out)...and if there is no net magnetic moment in the x-y plane, then all of the net magnetism is directed along the z-axis, which means the torque from the perspective of the net magnetic moment vector is 0 (i.e. no precession about the z-axis). However, you seem to suggest that this is the case with your language: "collectively, the quadrillion of spins in space sum to a nuclear magnetization vector M..." and then you provide a picture of this presumed net magnetization vector M (purple arrow) and depict it as precessing. Isn't this incorrect?
You make a good point! I should've made it clearer in the video, but when the M and B vector animation appears it is immediately following excitation, not in equilibrium. You are certainly correct that in equilibrium, M would be stationary pointing in the z direction and no precession would be apparent. An rf pulse (which I didn't show) is necessary to tip M away from B and see precession. Thanks for this observation; comments like this are helpful so I can see where gaps exist in my presentation! Cheers
So does MRI recieves the signal from each say cm2 of the 2nd slice, and then it produces the image, or it just recieves the overall sum and use FFT to differentiate. If it does how does it know the exact location from which the signal is coming?
Please answer this. Btw I dont normally comment on videos, but man this is exceptional content. Its very sad that useless channels get 10 or more million subs and views but people like you get less. It shows the reality and useless aims of people's mind. I just dont know how to show gratitude. Hands down one of the BEST youtube channels i have seen.
I have a doubt i dont know if you see this but here it goes.
The MRI detects the vector sum signal from all excited spins which the gradients encode, then the FFT transforms to an image. In the gradient echo sequence, the process is done one slice at a time. The scanner does not 'know' where the signal is coming from - but the different frequencies contained in the detected signal are known to have originated from specific locations. Hope that helps? Cheers
@@thepirl903 Ok, but it still did not answer my question (or that im dumb!) But let me clarify my question first. Lets keep it simple and say from different tissues you get signals of different frequencies. Now, suppose we are imaging the kidney. From the renal medulla, a short frequency signal is coming, and from the cortex a long frequency signal is coming. Using FFT you can seperate the frequencies and know that you are scanning two different tissues (lets juts say medulla and cortex is the only parts of kidney). For eg how does it know than renal medulla beneath or below the renal cortex?. How does it correctly 'position' the different tissues or parts in an organ so complex as brain? I understand you will get the information of how many different tissues are present, but how will you correctly postion that different tissues. For another example, im saying to you, suppose you are an artist, that there are 4 different types of tissues in an organ, apart from you use different colours for different tissues, how will you correctly draw the organ with correct position. Hope that is not lengthy!.
Hmm, ok I think I understand better your question, but we'll see. So the location of the separate tissues signals are indeed teased apart by the FFT in order to localize them in the image. The positions of tissues in space are directly mapped to frequencies by the larmor equation Δω=γGΔx. So the individual tissue signals can be precisely mapped in space. The MRI of course doesn't see 'tissue', it only sees protons (hydrogen nuclei). The way tissues are differentiated comes from their different nuclear relaxation properties (T1,T2, etc.), thus the brightness of a voxel reflects these contrast mechanisms, which in turn reflect the different properties of the tissue. Not sure if that helps at all?..
@@thepirl903 Thanks, that helps, so in short, larmor equation maps it.
Btw I really appreciate the long detailed videos, exactly what i love. I dont know how to help or say thanks. Im just 15 years old, and you are really an inspiration!
May I know what is your proffession?
oh boy! Christmas in August.
Thank you so much pro
Hi, I have one question about reciving bandwith, normally when we talk about bandwidth, let's take light for example, the spectrometer may collects 500-600 nm light when it goes to 400-800 nm the bandwidth is higher. Like you excited the slice, you have a band width, because different spin is in different frequency. but when you do reciveing, i suppose the frequency is talking bout K, However no mater you play longer time weak gradient or short time strong gradient the path in k space should be same. so the "bandwith" should be the same right?
The excitation bandwidth is the frequency span which will excite a certain 'distance' of spins in a given magnetic field gradient strength. When encoding during receive the bandwith is chosen so that while applying a gradient that no spins will be precessing faster than nyquist. If a stronger gradient is played then spins will precess faster so the bandwidth will need to be higher. The k will be the same (spatial frequency), but w is not (temporal frequency). Does that help? Great questions!
Would it be correct to say that M(x,y) is, in reality, a function of time due to the ensuing transverse magnetization relaxation process? How does this alter the formulation of the fourier transform application?
Indeed M(x,y) is a complex function of time. Our model for the GRE here assumes that M(x,y) is static. This of course isn't true, but on the time scale of a readout, these effects can generally be safely ignored if the readout is quick and T2* is long (not always the case, but fair for most biologic tissues). Similar to taking a photograph, we find a balance between our subject's motion (analogous to T2* decay) and the camera shutter speed (analogous to readout time) to reproduce the best 'snapshot' of M(x,y). Hope that helps
This is beautiful
I use a Siemens Prisma (with 80/200 gradients) and we recently acquired a next gen Siemens Cima which has 200/200 gradients. Does the benefit of the higher gradient strength come from better separation of the phases at each end of FoV, thus yielding greater contrast?
There's a few benefits to higher maximum gradient amplitudes including faster traversal of kSpace (good for collecting signal of short T2* samples, and rapid imaging of the heart) and ability to get to the kSpace edges faster (better resolution), better diffusion weighting (can acheive higher b-values in same time), fewer susceptability artifacts. Though higher gradients strengths can be louder and are more likely to cause peripheral nerve stimulation in some sequences. Cheers!
what software did you use for simulation please ?
Excellent presentstion!
Thank you fir insughtful dilligence.
Pleasexdo axsegemnt in the elecyronics and relsted base software to collect and cinvert the raw dataxand cinvert to image data.
Details of ADC sample frequency, how RF frequencies are shifted (mixers), RF PA, LNA and related filters.
And maybe post a biography on yourself.
PLEASE MOREEEE!
I. Was. Waiting! 🎉🎉🎉
Thank you so much,
How did you animate this? It is absolutely Amazing
I think you made a mistake at 4:49 , because red has lower frequency than blue and so on !
Can we get a physics video fully explaining the flip angle in part 1 of this series around 14min please?
Yes, I will do this at some point. That explanation isn't my favorite of the series, and I think it confuses people more than anything. Probably the best resource for QM treatment of flip angle (imho) is Slichter's Principles of Magnetic Resonance section 2.6 (www.google.com/books/edition/Principles_of_Magnetic_Resonance/pWzrCAAAQBAJ?hl=en&gbpv=0)
@@thepirl903 Thank you so much! I appreciate it (:
Please, somebody know how does he make this animations?
ruclips.net/video/yaa13eehgzo/видео.html , ruclips.net/video/rbu7Zu5X1zI/видео.html these are from other creators that do similar if not better animations similar to this style
Hell no. Keep making these MRI videos because the subject is vast and there's much to learn. CT and X-rays had their eras...in the 90s. Now its MR Time!
a year ago, where is part 555555555555~
dare to delve deep!
as someone who already knew the basics of mri that magnetic field strength and phase are varied along the axes to encode location, I can say that this video did nothing but obfuscate the concepts. it seems you are more interested in making fancy looking equations than explaining the concept, clearly exemplified by using an integral to denote a simple linear equation for phase as a function of a linear dimension. i will have to come back to see if this video is useful as a way to solidify knowledge one already has, because it certainly is a terrible way to learn this stuff