share interactive transcript request transcript/captions live captions
download
|
MyPlaylist
KARL PILLEMER: Hi, everyone. Many thanks for coming, and we're glad to see such a great turnout for what is a really, really exciting event. I'm not going to say very much. I just wanted to welcome everyone and introduce them. I'm Karl Pillemer, the Senior Associate Dean for Research in the College of Human Ecology. And one of the unexpected and extremely interesting aspects of this position was to have the MRI facility fall under my purview.
And as a sociologist, of course, a lot of this was foreign to me. I am not an MRI scientist, but I think I could play one briefly on TV now, having learned an awful lot about it. And I do want to say that this is an exciting time. We have spent the last year reviewing the Cornell Magnetic Resonance Imaging Facility or the CMRIF and looking at issues around increasing utilization by researchers improving its systems, et cetera. And along the way, it was suggested that we pursue a partnership with Weill Cornell Medicine and in particular the Department of Radiology.
I am extraordinarily grateful to Dr. Robert Min, who you'll hear from, but Dr. Ajay Gupta and Dr. Sumit Niogi for enthusiastic help with this. And I think we think it can be a model for how this kind of collaboration takes place. I also want to thank Dr. Martin Prince, also Weill Cornell Medicine for speaking here.
And I do want to just acknowledge a couple of other people who have been part of this process. And you could raise your hand. First of all, Peter Farley and Craig Higgins have been extremely helpful in working out the details of this partnership. Emily Qualls. Emily, are you here? And also is Rachel here or not? But our premiere MRI technician. And I also would like to thank-- where's Cindy Monroe? Who helped organize this entire event. So if you could join me--
[APPLAUSE]
And I'm not going to get-- and you will first hear from Rachel Dunifon, our dean. I'm not going to introduce folks, because you have biographies here. And thank you all for coming.
[APPLAUSE]
And since everybody here is more important than I am, they're keeping their own time. I'm not going to stop anybody.
RACHEL DUNIFON: But we'll be on-- we'll respect your time, and we have a reception right afterwards that we'll make sure we can all get to. I'm Rachel Dunifon, Interim Dean of the College of Human Ecology, and I'm really happy to welcome all of you here to this event, which kicks off a really exciting partnership between the College of Human Ecology and the Department of Radiology at Weill Cornell Medicine.
So some people might ask and even some of our alumni ask, what's an MRI facility doing in the College of Human Ecology? And while it may not be obvious at first, the cutting edge research that the MRI facility supports really fits perfectly with our mission in the College of Human Ecology. Here in the college, we're interested in understanding the ways in which social, political, family, environmental, and other factors influence human health and well-being, and we want to use that knowledge to improve human lives.
The MRI facility was the brainchild of human ecology leadership and faculty several years ago, and they understood that this technology would provide an excellent opportunity to truly understand how social and other factors influence our development and specifically the complex workings of the human brain. So in human ecology, research using the MRI facility allows us to ask questions like, how does poverty and material deprivation influence the brain's development? Why do teenagers make the decisions that they do, and how can we influence those decisions? And is there a biomarker in the brain that can help us identify early those who are likely to develop Alzheimer's?
While housed in the College of Human Ecology, the Cornell MRI facility has been a campus wide initiative from the very beginning, with support from the very beginning from the office of the Vice Provost of Research, engineering, the Veterinary College, the College of Arts and Sciences, as well as biomedical engineering, as well as an NIH instrumentation grant. In its very early years, the center was co-directed by Yi Wang from radiology and BME and Valerie Reyna from human development, and I'm really grateful for their enduring contributions to the facility.
Today the facility continues to be a really important resource for faculty across campus. It's actually housed right underneath my office in Martha Van Rensselaer Hall, and I can hear it and sometimes feel it when it's going. So I always enjoy that opportunity. As a testament to the reach of the facility across campus, sometimes I look out my window and I see carts of dogs or bunnies being loaded into the facility for scanning. One of our colleagues here is examining things such as how to understand and diagnose epilepsy in dogs.
So I'm really excited about the new partnership that we're launching today and for the leadership and commitment of the radiology department at Cornell that they're bringing to this facility. This next stage of our evolution will make the Cornell MRI facility a truly university wide resource, allowing us to benefit from the expertise of our Weill Cornell colleagues and fostering cutting edge research across both campuses. It provides both campuses for an opportunity to collaborate together on innovative research projects and grants, and this fits really nicely with one of President Pollack's key priorities, one Cornell, which is all about building connections between what goes on here on the Ithaca campus and the great work taking place in New York City.
So I want to welcome you and thank you very much for joining us today to celebrate this new opportunity and collaboration and for joining me and thinking about ways in which the Cornell MRI facility can lead to new discoveries in our fields. It's my pleasure now to introduce Dr. Niogi, Assistant Professor of Radiology at Weill Cornell Medical and the new Director of the Cornell MRI Facility who's going to talk with about his priorities and vision for the facility.
Dr. Niogi has been actively performing research in MR and cognitive neuroscience for over 15 years, and some of his areas of focus include traumatic brain injury, developmental conditions, neurosurgical applications, and neurodegenerative disease. We were just talking about his interest in athletes and their developing brain as a result of concussions and things like that, which I know is something he'll be working on while he's here.
He started as a PhD and MD student at Weill Cornell back in 2002. He's obtained his doctoral and medical degrees, surgical internship, diagnostic radiology residency, and neuro radiology fellowship at New York Presbyterian Weill Cornell, and we're really excited to have him here now as the director of the facility. Thank you.
[APPLAUSE]
SUMIT NIOGI: Thank you for that overly kind introduction. So I'm going to keep this very brief. I'm really happy, really excited about this opportunity. I see my primary role here is to make faculty across Cornell University, even Weill Cornell, aware that the facility exists and to utilize the facility. And part of that is to know exactly where it is.
So this is a questionably legal drone shot that we took a couple months ago. There's an airport nearby. So it's a little hard to see, but it's over there. And here's another angle. Beautiful view. Another drone shot that we took. There it is over there. And here's a picture, getting a little closer. This is me documenting my first time into the MRI facility. So it does exist. It is there.
And this past weekend, I saw Frozen 2 with my daughter, and the theme for that movie is transformation, and this is, I think, what I would say to be a theme for the facility as well. I want to go from just seeing this as an MRI scanner to a community resource. So this is the 59th Street Bridge, the Queensboro Bridge in Manhattan. It's a couple blocks from my office. And what I would like to see this facility become is a bridge between the two campuses. And there's some barriers there. One is awareness. One is access. Communication. These are all things that we're tackling now.
So this is-- I think you guys should be very aware of this. The bridge is here at Cornell over the gorge. I also want this facility to serve as a bridge between the different colleges. There's so much local expertise here, and I feel that-- when I've come up here and I spoke to faculty, a lot of the different colleges seem siloed. And I feel like this facility can help bridge the gap between different colleges. There's expertise at College of Human Ecology, but we could partner that expertise with the engineering school, the veterinary school, for example.
I'm not doing this alone. This is a huge, huge team effort. There are a lot of players involved. Rachel and Karl mentioned a number of them. And in doing so, I want you to know that I'm not doing this alone. I'm going to be working with a number of people. So there's an existing users group here. And one thing that I want to start doing is having quarterly meetings with the user group.
The users here, I want you to know that I'm always available. I'm going to start doing weekly meetings with MRI facility staff. We had a couple of weekly meetings already. And there's upper administration. There's a lot of people involved, a lot of big players. And we're going to start doing bimonthly meetings, update them, to get feedback, to make sure everyone's on the same page. What I want to do is make this whole facility operations very transparent. So everyone can have feedback. We can implement things to make this better.
So the plan. We'll talk about this later. Really create funding opportunities for current and new users so they can start utilizing the scanner. We want to foster new MRI researchers. This is a challenge. I realize this. And there are a lot of people that want to use the MRI facility, but they might not have the expertise and [INAUDIBLE] to actually get the data or analyze the data. So we're going to deal with that through collaborations, through tutorials.
We're going to help guide people through the IRB process, get them language that they can-- language they can put into the grants, put into the IRB. We want to increase scanner access time. So we're thinking about opening up the facility to potentially off hours, let faculty use the scanners on the weekends, at nights. And we want to update the policies and safety regulations, something that's modern and contemporary. I know a lot of users here talk about the pregnancy testing. That's one thing that we think we can get [INAUDIBLE] very soon.
A lot of this goes around the updated IT infrastructure. We already have a new scheduling program in place that has got very positive reviews. And we're going to change the idea behind data collection, data storage, who's responsible for it. Right now it seems that the individual investigators are responsible for the data. We're going to change it. We're going to make the facility responsible for it. So we're looking into an enterprise solution to store the data, have it fully backed up, and ease data access so investigators can download the data that day, tomorrow, six months from now, as long as the IRB is still active.
This is the founding principle from Cornell University. I would found an institution where any person can find instruction of any study. So if I were to write a core principle for the MRI facility, I would make it twice as long. So to provide MRI facilities so any member of the Cornell University community can utilize the resources within the facility to explore, research, educate, and disseminate knowledge to enhance the lives and livelihoods of the students, people in New York, and others around the world.
I was going through all the different schools that exist here, all the different departments. Not gonna read this out loud. But there are numerous, numerous uses for MRI, spanning just about every discipline you can imagine that's being investigated here. That's just one page. There's another page. And I really think that this is a facility, this is a resource that can really bring not just Cornell University and Weill Cornell together, but really Cornell University, all the different colleges together too, [INAUDIBLE] we can get around.
So thank you very much. And I'm around. I'm available. My contact information is up on the program. I'll be here tomorrow. I'll be here at the reception. So please come talk to me. My goal is to really kind of help everybody here. Thank you very much.
[APPLAUSE]
And it's my great pleasure now to introduce the keynote speaker, Dr. Martin Prince. There are a handful of giants in radiology, and one of the true giants of the field is Martin Prince. Most of us do a lot of great work as faculty. Dr. Prince did his work as a resident fellow. He was transformative for MR technology and MR research. When he did his residency and fellowship at Massachusetts General Hospital, he developed essentially contrast-enhanced MR angiography singlehandedly. And for that, he received the Gold Medal. It's a huge deal. Gold Medal for the International Society for Magnetic Resonance in Medicine. ISMRM Society Gold Medal.
He developed bolus triggering gadolinium MRA technique, which has been commercialized by GE. This is found on every single scanner on every platform now. He is remarkably prolific. So right now he is the professor of radiology at both Cornell and Columbia Universities, and he also serves as the Associate Editor for Radiology, Deputy Editor for JMRI, and he has over 250 peer reviewed publications with over 11,000 citations. Again, a true giant in the field. We're very fortunate to have him here to tell us about MRI and the outlook. Thank you, Dr. Prince.
[APPLAUSE]
MARTIN PRINCE: Thank you, Sumit. It's a great pleasure to be here, and I'm looking around. I see some familiar faces. People who are expert at MR and I see some young faces. Mostly young faces. And I'm very happy to spend a little time talking about some of the things MR can do. I can't go over every little thing, even many of the great things, but I want to try to give you just a taste of the exciting aspects of this technology.
I want to spend a few moments talking about how MR works. Just a few core principles so you can walk away and tell your friends, hey, this cool technology, let me tell you how it works. And finally, I want to share with you a little bit about how I got excited to do MR and to develop technologies and images, imaging techniques that contribute to saving lives every day. And you can tell from my disclosures that these have been extremely rewarding opportunities in the field.
Now, I want to start off, though, with a quiz. This is not the quiz. This is just get used to the quiz. So here we have-- does anyone know what this is? OK, that's a pineapple. I want to show you how we took this image. This is our machine right here.
See, it's got a hole there. That's the cylindrical magnet. You put it on the table. There's a button there. It just sort of slides in. And then you walk out of the room. You hit the button, scan, and boom, you got your image. OK, now here comes the quiz. Why don't we start-- once I get a group here. How about this?
AUDIENCE: Orange.
MARTIN PRINCE: Orange, OK.
AUDIENCE: Apricot.
MARTIN PRINCE: Apricot. OK.
AUDIENCE: Tomato.
MARTIN PRINCE: OK, tomato. That's pretty-- OK.
[AUDIENCE CHATTERING]
OK. Round one.
KARL PILLEMER: Wrong on all counts over here.
[LAUGHS]
AUDIENCE: A banana.
MARTIN PRINCE: OK, which one?
AUDIENCE: The bottom one.
MARTIN PRINCE: OK, bottom one. You guessed banana. OK.
AUDIENCE: Cucumber.
MARTIN PRINCE: OK. What type of cucumber?
[LAUGHTER]
This is like a test if you have kids.
AUDIENCE: Carrot stick.
[LAUGHTER]
MARTIN PRINCE: Cheese stick. And toothpaste. English cucumber. OK. This is the final bonus round.
AUDIENCE: Onion.
MARTIN PRINCE: OK, that's easy.
AUDIENCE: Pepper? Bell pepper.
MARTIN PRINCE: OK. Pretty good.
AUDIENCE: Acorn squash. Acorn squash.
MARTIN PRINCE: Acorn squash. That's good. Oh whoops, I have another round. Let me just jump to the end, because [INAUDIBLE]. More stuff.
Now, MR is fast. We captured the motion of the beating heart and other physiologic motions, peristalsis, and kinematic motions. It's fantastically fast in the hands of an expert. You can inject some gadolinium and track it as it flows down the body, here capturing the arteries of the entire body in less than a minute. I'm happy to report that Cornell University owns a patent covering this invention and has licensed to several companies.
I just spoke to, now I forget who it was. They came in and they said, oh, I've got some iron nanoparticles. This is an image acquired with iron oxide. This is a patient with congenital atrial septal defect. MR capturing the motion of the heart and all the details. MR is safe. We routinely use it to image pregnant women. I can't believe you're testing people to see if they're pregnant.
[LAUGHS]
OK. There is no ionizing radiation. We are doing these cases every day. 20 years ago, we were concerned, gee, is it safe? At this point our experience has developed to the point where we are now using it routinely. But I do have to point out that there are hazards, and because of the hazards, we are very strict about making sure people understand how to be safe using the MR equipment.
One of the greatest and scariest things is that strong magnets like metal stuff. And here is just-- or these are in case anyone missed it. This is a poor lady. She was walking into the scanner with her walker. Boom, it was made of metal. Sucked right in.
Now, we don't like to put patients with pacemakers, cochlear implants, internal implanted electrical devices into the scanner unless we know exactly what that device is and we confirm that it is MR compatible. So this is part of the screening of anyone who goes into the scanner that you have to learn in order to use it safely.
Now, here on the campus I was thinking, gee, what might you run into the most? Tattoos. Actually, most tattoos are safe. But some tattoos have conducting material, copper and other things. And these are first degree burns caused by the conduction of heat through that conductive elements of the tattoo.
MR is able to image flow. We can quantitate flow. Here's an example of a normal carotid artery supplying the brain. This patient has a stenosis. And you can see how it's disturbing the normal flow and starving the brain of precious nutrients. After fixing the stenosis, boom, we're back to normal flow.
MR can capture inflow into a region of interest. Here I'm going to look at the abdomen. What we do is we give a pulse which spoils all of the signal, and then we wait. The blood flows in. Then we take our picture. Here, you can see how the blood in the aorta flows in. Boom. And then it flows out the renal arteries. We get our picture. Everything else is dark. Here's some venous blood has flowed in, but it is flowing less fast. So it fills in less.
Another example. MR capturing the flow. We excite the spins in this axial plane, and then as the heart is beating, we view from the side, and we can watch those excited protons flowing down the aorta. Or you could pick a different view and see branches coming out the side of the [INAUDIBLE]. Now, we can make things bright or we can make things black. I feel like Dr. Seuss.
[LAUGHTER]
Blood can be black. I specialize in the arteries. So a lot of my examples relate to that. But you can take any organ. You could take any material. You could take any plant. Whatever you're interested in, apply the same concepts and come up with ways of utilizing this technology, which is just barely at the beginning of the mountains of possibilities that can be developed.
So for example, here we suppress the blood signal, and we can see the chambers of the heart, and we can see the aorta. We can see the aortic valve. Here another view of the aortic valve. Here the pulmonary valve. Or we can make blood bright. An easy way to make blood bright, inject some gadolinium. When we make the blood bright, we have higher signal to noise. We have higher signal. We have higher signal to noise.
This actually was an undergraduate who had fallen ill during his freshman year. Went to the infirmary. The nurse noticed, gee, the blood pressure in the arm is different than the blood pressure in the leg. You better go back to New York, stop in at Weill Cornell, get checked out. So he came to see us. Look at this. There is a stenosis in the aorta right here. A congenital stenosis known as a coarctation. He's living all these years and getting enough blood to live on, but not the optimum amount.
Now, MR can quantitatively measure fundamental properties of tissues. And this is a slide given to me by Yi Wang, and he also owns the patent that made a lot of money on the imaging of the arteries of the body. Here susceptibility is a fundamental property of tissue. And by looking at the phase data over multiple time points, the evolution of the phase, it's hard to determine the susceptibility, because many different distributions of susceptibility could create that phase field.
But they made the amazing discovery that if you constrain your solution to be similar to what we know the brain should look like, suddenly a solution is possible. And now with this incredible innovation, we can directly calculate the susceptibilities throughout the body. Here in the brain, we see a lot of susceptibility in basal ganglia, where the iron is accumulating. And this can guide surgical interventions, and it can diagnose many conditions, and whether the condition is getting worse or getting better, such as multiple sclerosis and other things.
I want to show you how this susceptibility imaging is applied in other parts of the body. For example, oxygen can make blood dark or make it bright by the susceptibility effect. Here if we look at the left ventricle, you could see the oxygenated blood has low susceptibility. But if we look at the right ventricle, the oxygenated blood has high susceptibility, because deoxygenated hemoglobin has high susceptibility. We also could see the dark aorta from being oxygenated and a bright hepatic vein. So this enables us to quantify the degree of oxygen difference, saturation difference, between the veins and the arteries.
And amazingly, this same concept has been utilized to enable MR to image what you are thinking. Sometimes people call it BOLD. Blood Oxygen Level Dependent. B-O-L-D. Yeah, BOLD. I remember when it was being invented. They asked me to volunteer. And the reason why they asked me to volunteer was because I used to ride my bike a lot over to the MR scanner, and I was in good shape. I could hold my breath for two minutes. So I could get a lot of good oxygen desaturation.
[LAUGHTER]
Look at this. It's lighting up in the visual cortex. What is that guy thinking? Movie. The motor cortex. Finger tapping. Wow, look at that. It's lighting up everywhere.
[LAUGHTER]
All right. MR images molecular diffusion, and we can tell the direction that the protons are diffusing. Protons like to diffuse along the white matter fiber tracks. And collecting this information about direction enables us to map how the brain is connected together and how other things are connected together. This is an incredibly exciting technology that it's just led to an enormous enthusiasm among grant writers, such as in the Connectome Project, which has spent more than 100 million supporting this development.
Well, there's so many things that MR can do, and I guess what we're going to do is we're just going to hang out afterwards, and if you want to learn more, you ask us, and we'll try to respond. Probably what we're really hoping is that things that have not yet been thought of, you guys, because you're not living in our world, will come up with new ideas, new ways of using this technology, new ways of making this scanner have this great impact predicted by Ezra Cornell.
Now, when I got invited to give this talk, Dr. Min said, I'm teaming up with Human Ecology, geez. We're going to come down here and find ways of doing great things. And Dean Dunifon has got it all figured out. Yeah, but she didn't tell me anything.
So I thought, OK, well how are we going to figure out whether this is a good idea or-- well, why don't we just do a Fourier transform? That's what we do in MR. I was looking at the raw data. I just couldn't figure it out. Well, I thought, why don't I take a look at the center of k-space? That's where the core fundamental stuff is. Wow.
[LAUGHTER]
But that's not all there is. There's also this data in the periphery of k-space. Let's take a look there. We're going to have great fun on this project.
OK, I want to move on and just spend a few minutes talking a little bit about how MR works. And forgive me for all you guys who already know how MR works. MR takes advantage of one of the fundamental asymmetries in nature, which is that some atoms, some isotopes, have an odd number of nuclear particles. And that odd number of particles confers a property that we refer to as spin, nuclear spin. And those spinning nuclei can then have a dipole moment, a magnetic moment that we can manipulate in the MRI scanner.
Now, just to give you an idea of how primitive MR is, almost everything I've showed you or I'm going to show you is based on just one of these. And there are all these others, and many others that I haven't listed, waiting to be tapped into. And we use the hydrogen, just one particle. It's actually, it's just a proton. Sometimes we call it proton MRI. And we use that because it's the most abundant nucleus in the body. So it gives us the most signal. It's relatively easy compared to the others. We're picking off the low hanging fruit, which makes good sense for an agricultural school.
And I also want to just mention that the MR signal comes from about one in a million protons at one and a half tesla. Most of the protons are not generating signal, just one in a million. At three tesla, it's two in a million. And luckily, we have a three tesla here at Cornell.
But I do want to just show, and you see here, if you look at the image, where we have protons like fat and muscle and liver, you got signal. And in the air where there are no protons, there's no signal. Although this patient does have a few lesions in the lungs. This patient was an IV drug addict. And we see some abscesses here. These are septic emboli.
Now, I do want to show an image of helium 3, just to give you a feel for how maybe a different nucleus could look differently. Now, there's not a lot of helium in the body, but you can breathe it in. And helium has the nice property that it can be hyper polarized so that you get signal from all of the helium atoms, thus magnifying the signal a million fold compared to proton MRI.
And so here we can get pretty decent images of helium breathing into the lung, even though we know the density is very low. And here you see the trachea, and then we see the bronchi, and then we see the alveoli. And this patient was asthmatic and has a lot of heterogeneity to the distribution of the helium. And of course, every nuclei is going to be unique and is waiting to be explored, and many others can also be polarized in this way. So we're just waiting for people to explore this.
All right, well, why do we need a magnetic field, a static magnetic field, a three tesla magnet? And that's because spins, nuclear spins, are normally randomly oriented. And we can't do anything with that. When we put it in the magnet, they tend to align to the magnet, and this creates a net polarization that we can then work with and we can play with and we can use that make our images. The time it takes for this alignment, we refer to that as T1. Some people call it the spin lattice relaxation time.
Now, the thing that I love about MR, there's only one equation. The whole field, only one equation. And look, it's only got three parameters, and it's just an equal sign. [INAUDIBLE] This is the best. And it's so simple. So this is the resonant frequency, and this is the field strength, and this is the gyromagnetic ratio.
So we give a radio frequency pulse at this resonant frequency, and this net magnetization, which normally would be aligned with the magnetic field, tips out of alignment and starts to precess. Well, it precesses at the resonant frequency. So easy. And it gives off signal. How long does the signal last? Let me see if anybody wants to answer that. How long does the signal last?
AUDIENCE: Depends on the strength of the field.
MARTIN PRINCE: I'm just going to go back, because I want everybody to get this right. Remember, we said this is T1. Oh. T2. It's just two parameters, T1 and T2. OK, signal lasts for T2. That's how long it lasts.
Now, here's a typical machine. Here is the coil. We put it on the patient. And just in case you forget to put the coil on the patient, we have a backup coil built into the wall of the scanner. We call that the RF, Radio Frequency Coil. And we can give pulses, like a 90, a 180 refocusing pulse. And then we can listen to the signal echoing back. We call that the echo. Time to echo. The time between the pulse in the middle of the echo.
So if we look at the echo time and we know the T2 relaxation times of our tissues, we can figure out what the image is going to look like. We can figure out what's going to be bright and dark. So for example, watery stuff like cerebrospinal fluid or bile has a very long T2. The signal lasts a long time. But most other stuff, tumor, muscle, organs, brain, shorter T2.
So if you use an echo time around 500 milliseconds, the only stuff you're going to see is the watery stuff in the body. Here we see the common bile duct. We see the gallbladder. We see the stomach with fluid in it and the duodenum. That's a T2 weighted image. And by playing with the T2, we can determine what's going to pop-- what's going to show up on the image and what's not going to be there.
All right. Now, here I made the echo time shorter. 20 milliseconds. So almost everything is bright. Everything except for flowing blood, because blood is flowing, the motion of the blood spoils the signal, and we get a black blood image where everything else is bright.
Now, let me just show you how a T1 weighted image works. Even my residents sometimes don't even know this. Remember, T1 is the time to align with the magnetic field, the time for these protons to line up with the magnetic field. And we gotta give a lot of pulses to get our image, because it takes a lot of echoes to fill up our Fourier data array and build up enough signal to noise. And the time between the pulses, we call that the time to repeat, TR.
So if the time between pulses is too fast, not enough time for spins to line back up with the magnetic field, tissue is going to be dark. Or if it's long enough, tissue will be bright. So we can see what's going to be bright or dark on a T1 weighted image just by looking at the T1 relaxation time.
See, here's a TR of 400 milliseconds in between pulses. The fat recovers quickly. So the fat is bright. Muscle has a longer recovery time, a longer T1 relaxation time. So muscle is dark. Now look at how this leg is swollen. See the edema and the fat? And she's about 23 weeks pregnant.
I just want to point out that the blood is normally dark. Very long T1 relaxation time. But blood is flowing. And if we get an axial slice, blood will flow into that slice, refreshing the slice with unsaturated, unaffected spins at the rate of one millimeter per 10 milliseconds. So if we use a TR of around 30 milliseconds for a three millisecond slice, what we'll see is everything will be dark except the blood.
And here for example, on the right side, we have bright artery and vein. And on the left, a bright artery but a dark vein. This patient has thrombosis of the left common femoral and iliac vein. Deep venous thrombosis. And that explains why the leg is swollen. So easy to do in pregnant women without having to inject anything. Very easy.
All right, a lot of people have this impression, oh, you know what? I did my thesis on something else. I did my PhD on economic theory. How could I possibly do MR? And we were talking about this at lunch. And the reality is, it's not that complicated. Actually, I did my graduate work on laser physics and then switched to MR and figured it out without any difficulty.
This is me a long time ago working on lasers. We were trying to laser the crud out of arteries. And just about the time I finished my thesis, the stent was invented. And the stent was so good, the laser was completely unneeded, and my entire career had just sort of vanished in front of me.
But also at about that time, someone came by, just like I'm coming by here, and gave a talk about MRI. And being down in the dumps, I was ready to grasp at anything. But it was so exciting, and it had so much potential. I knew that I had to do something with MR. Since I was working on blood vessels, I thought, OK, I got to figure out how to image blood vessels better. And people had been working on these techniques, which did not require injecting anything, but I had the sort of feeling, oh, boy, if we could just inject the gadolinium and image the blood vessel, that would be fabulous.
Well, in those days we had this thing called a certificate of need. I don't know if you ever heard about certificate of need. But it's basically a way of controlling how much medical care is available for the masses so we can contain health care costs. And our hospital had one MR scanner for 1,000 beds, and we weren't allowed to have any more. And it was running basically 24 hours a day, and anyone who wanted to do a project, they were just out of luck.
But then Ronald Reagan came along and said to Gorbachev, tear this wall down in Berlin. I don't know if anyone here remembers that. And sure enough, the Cold War ended, and suddenly there were military bases that we didn't need anymore. And where I was in Boston, there was-- the hospital was over here. And over here in Charlestown, there was a naval yard. And the hospital bought that naval yard and installed an MR scanner. And they said, OK, anyone who wants to try anything, just go over there and do what you want, because we're having trouble getting anyone to use it. Sound familiar?
[LAUGHTER]
So I rode my bike over there, and I got in great shape and volunteered for the BOLD experiment.
[LAUGHTER]
And at that time, it was just impossible to get anything to work. And so I did a literature search. That's usually the way I do things. Do the experiment and then do the literature search. And I noticed that all these great people at the time had written about how, oh, unfortunately, gadolinium distributes in intravascular space, extracellular background signal, blah, blah, blah. And basically, you can't get your project to work. And here's another giant in the field who said MRI angiography requires no contrast. And I had already signed up to work on this. Here's an entry in my lab notebook.
[LAUGHTER]
Luckily, there was a company that was called Kodak. And Kodak worked with us a lot, because they made the film. We took the x-rays on the Kodak film. And they knew that the digital revolution was coming. So all their chemists were desperate to find applications for all those molecules sitting on the shelf that they had been synthesizing.
So a chemist came to me and said, well, can you inject some of these and see what it looks like? Well, it turned out they had a lot of rabbits from the veterinary school, I guess. And she came to visit me and drew pictures of all these molecules. She figured I didn't understand organic chemistry, so she left out the C's and H's. And I was assigned a time when I could go use the scanner. Because during the day, it was used on patients. And then at night, they let people do research for free, actually, because how can you charge at night?
My assigned time was 2:00 AM to 4:00 AM. And I was also given an animal tech, which was critical, because I did not know how to start an IV in a rabbit. But this guy could do it, and we got in there, and I remember the very first rabbit we went in. I thought it was an MR tech, but actually it was a rabbit tech. And so we're both looking at each other. OK, start scanning. And then he looked at me. Start scanning.
[LAUGHTER]
But we sat down and we started pushing those buttons, and pretty soon we got the thing working. And I've discovered now over the years as I've worked with residents and fellows, there's some residents who want to be with you while you're in the reading room interpreting images. And they kind of cling on you. And they get to the end of their four years or the end of their fellowship and I realize they don't have any knowledge of MR. They've just been clinging on me.
And then occasionally, you have a resident or a fellow and you're wondering, hey, well how come they didn't show up this morning? And you realize they were in the night before scanning people or scanning phantoms or working on the scanner. And then as you talk to them, you realize, gee, in a very short period of time, that process of scanning has converted them into expert at MR. Super expert. More expert than me sometimes.
And there I was scanning, and I got some early images. And I was still getting a lot of artery and vein contamination. But I had a trip down to Yale. And on the train, I was just sitting there wondering what to do. I started modeling where the gadolinium goes. And I have this page from my laboratory notebook which shows when I figured out that if you scanned while you were injecting the gadolinium, you could get the arteries without the veins.
And so I was all excited to go back until I realized that I needed to use a higher dose than what was approved by the FDA. But the very next week, a new compound came out, gadolinium compound, approved at the higher dose. So I thought, that's going to be a message. Sure enough, I tried it out. Boom. One of my first MR angiograms.
Now, one of the problems is that in those days, we weren't used to injecting while the patient was in the scanner. We would do some precontrast scans, pull them out, then inject, and then put them back. And that's why nothing was ever showing arteries alone. So I had to kind of figure out how to do that with the stuff that was lying around. And all we had were little syringes. So I would load up a couple syringes with gadolinium, and then I had one of saline to flush it through. That stuff was so expensive. You didn't want to waste any in the tubing.
Well, I put the syringes here, put the first one on, start the scan, start that injection, and then, OK, I'm done with that syringe. Take it off. Oh, shoot. Which one has the gadolinium? And I realized it didn't matter. It didn't matter which syringe. It always came out the same. And eventually, I figured out that what really mattered was when you injected the gadolinium that corresponded to acquiring just the part of the MR which represented the center of k-space, those [INAUDIBLE] spatial frequencies that dominated contrast. That was the key. And I figured that out because of these stupid syringes.
And we patented that and wrote a quick paper on that. And we were getting letters from referring physicians. Your findings by MR angiography were confirmed at surgery and successfully treated. I was very impressed with the detail of this study. You just cannot imagine how exciting it is to be there creating new images and having people write to you and tell you, wow, this made the difference in the patient that I'm operating on and enabled me to fix them and make them better.
And now, of course, MR has accelerated incredibly. This is some work by Pascal Spincemaille, who is a physicist in Yi Wang's group. Now sampling the MR data with a spiral technique and taking advantage of the fact that if you know what the image is supposed to look like, you only need a tiny amount of data, just one spiral, to calculate the entire image. And we can do what used to take five minutes we can do that four times a second now. Here looking at the abdomen.
And here, for example, you can see on this motion it's occurring as the gadolinium is arriving on these three dimensional volumes acquired at four frames a second. We're looking at all kinds of things now. Ureteral peristalsis. You can see how the ureters are pumping the fluid in the ureters.
Here the heart beating. This is done at 50 frames a second. A normal liver, the heart just kind of punches the liver and it absorbs the punch, but a cirrhotic stiff liver, each beat of the heart moves the whole liver like a stiff brick. One of the students was a lover of opera. And decided just to try out his singing voice in the MR scanner.
[MUSIC PLAYING]
All right. I know I'm supposed to talk about the future. Actually, I'm mainly just talking about the past, because people here, you people, you're the future. Your ideas that are different and different from the mainstream. You're off the beaten path. You're in a different environment. You have different challenges. And you could take these tools and discover something amazing.
OK, so we have jumped to the conclusions. So MR is a set of tools. It shows anatomy and flow and motion and chemistry and function and it's good on plants and animals and stuff. MR expertise comes from tinkering and operating the scanner. Of course, safety screening is essential for safe MRI. And there are many tools, many, many tools. We're just at the foothills of the mountains of possibilities that MR has to offer.
I already showed you how most nuclei are waiting to be utilized. And maybe it won't be in humans that they're the most useful. And faster techniques, and maybe what's better are slower techniques but with phenomenal resolution. And everyone's using deep learning now as a way of getting better quality with less data.
So I want to thank you again for this chance to share what's exciting about this technology and hopefully get you interested in thinking about, wow, maybe I can discover a new way to use MR, or I can answer a question with MR more easily than with other methods, or maybe I'll just do it because it's so much fun. Thank you very much.
[APPLAUSE]
KARL PILLEMER: And I think we'll have a little time for questions at the end. But Dr. Min, who is the chair of the department of radiation.
ROBERT MIN: Good job. So being Martin's chair, I hope you didn't get out-- you got what was intended. What I got out of it was that you don't have to learn how to use an MR. You just have to press buttons, which is not true. You don't have to show up to work the next morning, which is also not true. And you don't have to pay for the scanner. For that latter point, we do have some [INAUDIBLE]. And that's what I'm here to talk about.
So Karl said, among other things, I've the chair of radiology down at Weils Cornell for about 15 years. And when I got it, we were on a call, and I was speaking to Rachel and Karl, and you all were somewhat distressed about having this unbelievable state of the art piece of equipment here. By MR terms, a really elite, fairly new scanner here.
And there was some talk about actually closing shop. And as much as I would have loved to pick up a 3T at bargain price, there's no way my wife would allow me to bring it home. I thought better than scrapping it, let's see if there's an opportunity to really have it realize its full potential.
For those of you that don't know, I think you all know how important MR scanning is in the clinical world. We run a huge department down at Weill Cornell. We couldn't do it without MR imaging, and I think most people know that. It's essentially the single most important diagnostic tool that we have in medicine these days. What you probably don't know is it's also the single most important tool that we have in terms of our research.
The vast majority of our research funding is actually tied to MR imaging. And so we fully realize the potential. What hasn't been realized is I don't think we've put the effort in to make people aware of it. And I was joking about the funding part of it, but not having the resources to even acquire some initial experience for pilot data is a big barrier to it.
So one of the things that I felt that we could do to help, besides providing some expertise around just how to operate the equipment, was really to facilitate some of the funding, particularly for the pilot projects. For this facility to work, it can't work just based on what we can do down there, which is clinical imaging. We really need it to be a robust research magnet. And in order to do that, we need grants and particularly federal grants.
Down at Weill Cornell, I can tell you that's our currency when it comes to research. We'd like people to publish impactful papers. But really, it's the grant funding that allows us to continue to survive. And hopefully we can help do that here. But in order to get to that, we realize we need to fund some of the early studies before you have any research funding. So what I told Rachel and Karl that we would be more than happy to do is to fund several pilot grants up to $250,000 cumulative per year for at least three years.
The grants, by and large, I think a $20,000 grant actually would get you a lot of dollars in terms of imaging. You can do a lot of imaging with that. We want, actually, a lot of the dollars for the grants to be used for the imaging rather than purely for travel and things like that. And we're more than happy to do it. I think the benefit of doing that is also hopefully it really will help increase awareness of the great facility that you have here, but it will also increase awareness of some of the potential collaborations that I think we could really foster between Weill Cornell and up here in Ithaca. So I think there are many, many, many potential reasons to do this.
So my team down at Weill Cornell says-- I always say, every time there's a challenge, and my residents say this to me all the time, every time there's a challenge, whatever it is, I need more coverage overnight and on the weekends. And everyone's distressed about how we're going to provide it. I always say one thing. Regards to challenge, what fixes it? Money fixes all. And that sounds ridiculous, but I do realize that when it comes to research, it is helpful.
So I think it's going to be a resounding success. And I really want to thank Karl and Rachel for being so openly willing to collaborate with us to help do that. And you'll be seeing a lot of Sumit here. Hopefully that's a good thing. I think you'll find him to be equally engaging and collaborative. What we ask is you guys come up with the ideas. Let us help you figure out how to use MR to do that.
So I'm really hopeful that next year, if we're having a follow up event, that there'll be a lot of pretty exciting projects that are being done. And there's already some talk of some grants missions going in for federal grants. So it's a pleasure to be here. I want to thank you all for coming. I know you're just a small representative of a lot of the other people who are not here. So get the word out.
And I should say, there's Yi back there, Yi Wang was the PI on the S10 grant, the $2 million grant that helped fund this. And Yi will also be a real-- Yi is one of the most successfully funded MR researchers in the country. And so we're bringing a lot of expertise and talents that Yi has had here also. OK, so thank you guys. I think Karl wants to do the next most fun thing.
KARL PILLEMER: Yes, thanks. And let me just say, do we want to take a moment or two? Because we're not--
ROBERT MIN: Oh, a panel?
KARL PILLEMER: Well, I hate to stand between people and cocktail hour. But I'll do it for one moment. Sumit, do you want to say a word? Has a sign up sheet been circulating? So we'll obviously email the RFA to everyone. And for questions now, Sumit, I think it would be emailing you, right? Do you want to say a word about the deadline, kind of what?
ROBERT MIN: We don't have a deadline yet. It will probably a couple of months from now. And basically, we'll give people an opportunity to submit just some expression of the interest. Ultimately, we'd like to review those. There'll be a monthly disciplinary team that will look at that. We'll send out some materials based on the different criteria on which they'll be judged.
As much as I would like cross collaboration, collaborators from both Weill Cornell Medicine and up here, it doesn't have to be in a formal way, but certainly that would be ideal. Our hope is that we can actually get these selections done sometime this spring. That would be, I think, not overly aggressive. So definitely await-- send us your information. But we're going to send it out more broadly.
The other thing that Sumit's going to do is he's going to go out to all your departments. So we're going to set up meetings. We will have follow ups with each of the departments. We'll even talk about this a lot more. And he's committed or we've committed a lot of his time to this.
[LAUGHTER]
KARL PILLEMER: We have a few more minutes. Do folks have questions? So really for anybody who's spoken, anything?
SUMIT NIOGI: We're going to revamp the website. So all this information is going to be on the Cornell University MRI website as the single place you can go to.
AUDIENCE: Two questions. On the list, you give applications on MRI. I don't see the cancer on the list. Is that one of the one you are going to move into or not? And I saw the healing, also is one subject I'd like to know more about.
ROBERT MIN: Well, from a clinical standpoint, MRI is obviously integral to both diagnosis and follow up of cancer. And so yeah, that was not a all-encompassing-- that was a list that Sumit came up with just to get people exposed to the variety of things that MRI can be used to investigate in every college that exists here, in every department that exists here. It was not in any way to be-- there's no scientific validity in that list, nor is it all encompassing. It was really just to generate some thought and interest and excitement.
AUDIENCE: So you will not exclude anything.
ROBERT MIN: No, no, no, no, no, no, no, no.
SUMIT NIOGI: Just the opposite. Everything is possible.
KARL PILLEMER: Did you want to respond?
MARTIN PRINCE: Yeah, well, obviously yes. I mean, Dr. Min said it. MR is huge in cancer. I mean, it's one of the huge applications. We're hoping to find additional huge applications, but it has room to get bigger.
KARL PILLEMER: Sure.
AUDIENCE: Are you going to continue to work on a smaller bore so that we can get nice resolution on some of our research subjects, which are about the size of a very small [INAUDIBLE]?
ROBERT MIN: I'll tell you my thoughts. They may not be your thoughts.
MARTIN PRINCE: No, I think that's a great question. And I think if Sumit wants to talk about it, I'll just mention that this scanner here at Cornell is an amazing scanner. Three tesla, state of the art gradients, very high performance. It's actually one of the smaller bore sizes that you're able to buy in the kind of human scanner world.
But I think what you're really talking about is are we going to have smaller RF coils so that we can get high SNR of small things, like mice and cheese sticks. And I think the answer to that is that this facility has an amazing coil lab waiting for people.
ROBERT MIN: That's right. Waiting, waiting. No bore size in itself is not what's important. I'm gonna tell you right now, the most exciting stuff in the future MRI is not better imaging with smaller bores. It's actually larger bores at low field. So that that's actually contrary to what you may be saying, but you're really talking about the coils. And sure, that's been one of the things missing here.
But the bore size in itself, you will see stuff that will blow you away in terms of bore sizes that you've never seen before. And from a clinical standpoint, that's particularly important, because what are the downsides of smaller bores? They're very claustrophobic. So for research purposes, that may be important, but most of these are driven by the clinical use. So you're not going to see bore sizes getting smaller. They're probably even bigger.
KARL PILLEMER: Do you want to?
SUMIT NIOGI: Yes. So in terms of coils, one, there's a huge opportunity for research here, for engineers to build coils. Secondly, there's a mouse study, mouse coil. We use a mouse coil to do the studies here. Again, this is cross campus collaboration. We have animal scanners at Weill Cornell. Potentially, we could do scanning at Weill Cornell on the animal scanner [INAUDIBLE].
AUDIENCE: [INAUDIBLE]
SUMIT NIOGI: Exactly.
AUDIENCE: [INAUDIBLE]
ROBERT MIN: [INAUDIBLE] had a great-- I don't know if you went to any of the lecture. Apparently she gave a great-- [INAUDIBLE] was telling us that Simone Winkler, who is one of the investigators that we recently recruited, she has expertise in hardware and in particular ultra high field. I heard she gave a spectacular talk up here in Ithaca. So you're absolutely right. We're bringing on some of that expertise.
KARL PILLEMER: I would add too, and you can tell that I'm not the dean by saying this, but I agree that there are some problems that you can solve by throwing money at.
ROBERT MIN: I was kidding.
KARL PILLEMER: And if there have-- if there are pieces of equipment, if there are barriers to use of the scanner, that can be resolved by purchasing equipment or by additional investment. I think that the [INAUDIBLE] would very much like to hear it. And he's in a position to adjudicate how else we might invest in it. So I'm going to ask that question.
SUMIT NIOGI: In particular, a mouse coil has come up multiple times [INAUDIBLE].
KARL PILLEMER: We're getting a mouse coil. [INAUDIBLE]
AUDIENCE: So I have a question about actually planning to use this and think about writing a grant and questions about the bio statistical kind of expertise in this area. For example, what's the precision of the measurement for various kinds of-- what's an effect size for a region in the brain?
How many samples? How many subjects? How many times would you need to sample the same subject? [INAUDIBLE] Can we access bio statistical expertise, measurement expertise, to address the questions specific to whatever [INAUDIBLE]?
MARTIN PRINCE: That's a great question, because NIH, oh my god, their statisticians fuss on this [INAUDIBLE]. And I think most of the standard techniques, there's a lot of data that you can get from the literature on variability, reproducibility that you can use for your sample size calculations. If you invent something new, it's on you to do that. So that's the double edged sword of inventing new stuff.
ROBERT MIN: So as I mentioned, the majority of our research at Weill Cornell, we have a pretty big NIH funding portfolio. We would love if there is any assistance you need, that's part of this collaboration. It's not just answering a question about the imaging. It's about even some of the grant [INAUDIBLE] that's required. We would love to work with you. So please, anyone, and tell everyone, if there are any questions, just send them our way. That's part of that assistance that we'd be more than happy to provide. 100%.
SUMIT NIOGI: I'm not doing this alone at Weill Cornell. There's a huge team of faculty [INAUDIBLE] what's behind this. So there's access to numerous people. [INAUDIBLE]
AUDIENCE: I'm just curious to know about the basic versus applied emphasis, if that will change at all. It seems like one of the focuses of having a scanner here was that it was sort of intended for non-medical use and obviously medical applications are the [INAUDIBLE] meaning of scientific research. But I'm curious if there will still be a space for the basic research uses.
SUMIT NIOGI: Yes. This scanner's open to all forms of research. It can be fundamental basic design for hardware, for sequences, [INAUDIBLE].
ROBERT MIN: I'd like to say that, in fact, those are the areas that we want to do more of. I think it's been too narrow, and that's of the, I think, challenge that this facility has faced. There is no type of MRI research that I think we would be biased against. So fundamental research, plant imaging, whatever your thing is, you should let us know and we'll work with you.
We inevitably have a very biased view of the type of research. But I think that's what Dr. Prince is trying to say. We should not be the ones talking about thinking about the future of this. It's really everyone out here that doesn't do what we look at every day. Because I mean, that by nature is going to blind us to a lot of the great opportunities and the real advances that will be made. So definitely 100%.
MARTIN PRINCE: I just want to make a comment, because we had a resident who was imaging specimens. And she would put them in the scanner and run them for 12 hours. And that's the kind of thing that's going to be happening at night.
[LAUGHS]
So the stuff in the daytime is going to be probably, I'm just guessing, the shorter slots. If you've got something that is going to occupy, you got to set up some fancy apparatus and you need the whole week, and you're going to be limited in what hours you get access. But the whole point is that we want more of that crazy pioneering, innovative, amazing ideas happening. And I think Sumit wants to find a way to make it happen.
SUMIT NIOGI: We can do that here, probably more easily here than we can Weill Cornell. When I was an undergrad, I did 24 hours of imaging on a mouse and plants to. I imaged plants when I was an undergrad.
KARL PILLEMER: One last question before we-- and if not, everybody is staying around at the reception. So you can certainly meet with people one on. Any last burning? Well, and if not, please join me.
[APPLAUSE]
Leading a new era in Magnetic Resonance Imaging (MRI) research capabilities at Cornell, Dr. Sumit Narayan Niogi, assistant professor of radiology and clinical associate in radiology at Weill Cornell Medicine, has been named Director of the Cornell University Magnetic Resonance Imagining Facility. Under his new leadership, the enhanced Cornell MRI facility capabilities expands to any member of the University community to explore, research, educate, and disseminate knowledge – enhancing the lives and livelihoods of the students, people of New York, and others around the world.
Along with Dr. Niogi, speakers include the College of Human Ecology’s Senior Associate Dean for Research Karl Pillemer, and Interim Dean Rachel Dunifon, as well as Weill Cornell Medicine’s Professor Dr. Martin Prince and Chairman of Radiology Dr. Robert Min.
