 Hello and welcome to noon conferences hosted by MRI online in response to the changes happening around the world right now and the shutting down of in-person events. We have decided to provide free daily noon conferences to all radiologists worldwide. Today we are joined by Dr. Petra J. Lewis. Dr. Lewis specializes in women's imaging, but her primary interest is in educating students, residents and faculty. She has designed enumerable educational resources such as RAD exam. She is currently the president of the Association of University Radiologist. A reminder that there will be time at the end of this hour for a Q&A session. Please use the Q&A feature to ask all of your questions and we'll get to as many as we can before our time is up. We'll also be using the polling feature for today's conference, so be on the lookout for that. That being said, thank you so much for joining us today, Dr. Lewis. I will let you take it from here. Thank you very much. So welcome everybody. I'm Petra Lewis. I am actually no longer the president of the AUR as of Friday. So I've handed that particular monkey off my back and excuse me if I'm a bit croaky in this, I do not have COVID. I've not had COVID, but this is my third hour of web-based teaching today, so I'm going to look croaky. So our core learning objective for this lecture is for you to be able to do this. So by the end of this activity, I hope that you will be able to interpret breast MRI scans using the fifth edition ACR MR by RADS lexicon. And this talk is really about that lexicon, not so much of interpreting the findings on the MRI. So just realize I am focusing on that for this particular talk. And I'm also, let's go back one. I am really only going to be talking about gadolinium enhanced breast MRI in this section. The ACR by RADS does have a section on implant assessment, but I'm really not focusing on that today. So the ACR by RADS for breast MRI originally came out in 2003 and then had some significant updates in 2013. Now, you will see throughout this talk that I highlight in yellow elements that were changed in 2013 from the 2003 version. And the reason I'm doing that is, A, it's difficult for some of us to lose our old habits of how we interpret studies. And B, you may be looking at reports or research studies that used a prior lexicon, just so for you to know that these are different and why you're not seeing those terms now. Now the by RADS for breast MRI is very similar to mammography in that it develops this standard lexicon for how you interpret and lesions and how you vocalize them in your reports. And it has the same zero to six by RADS categories that we have in mammography and also in ultrasound. So why is by RADS, whether we're talking of ultrasound or mammography or MRI, been developed and it was really to standardize the terminology. And by standardizing this terminology and taking out sort of much more vague terms we facilitate communication between radiologists from one report to another. So we reduce the inter and the intra observer variability. And these factors all lead to some indication of the lesion risk assessment, those by RADS interpretive categories. Now the report structure, like most report structures is going to have certain consistent elements within it and then a few that are more specific. And I'm going to go through each of these in detail. You obviously need to have the indication for it so your insurance company will pay, need to have a delineation of the technique, including the type and the amount of gadolinium contrast that was deployed. There needs to be a indicator of the breast composition, both the density and the enhancement pattern of the background parenchyma, and we'll talk about those. For each finding on the breast MRI, you're going to describe the morphology, you're going to describe the distribution. And then you are going to describe the kinetics of the contrast enhancement within it. And then you may or may not be describing other elements such as a T2 signal. As always, we want to compare every study comparisons. We need to give our assessment by RADS category and then indicate further management. For the indications, obviously we want to indicate why this study is being done. So do they have a personal family history of breast cancer? Is this a screening syndrome? Is this a screening study? Do they have a personal or a family genetic syndrome for the screening for a screening study? Do they have any clinical findings or symptoms in their breasts that's being assessed currently? Is there a history of prior breast surgery, radiotherapy, chemo, or even hormonal therapy, which can be helpful? And then if they are premenstrual where they are in their menstrual cycle because it's going to infect your interpretation. So let's start with looking at the amount of fibro glandular tissue. Now, these four categories, ABC and D, that you may be familiar with from mammography are the same categories that we use in breast MRI. Unlike the new by RADS 5 version for mammography, we do no longer break these down into quartiles. So if you remember, it used to be less than 25% fibro glandular tissue was almost entirely fatty and so on broken down into quartiles. They are not used in MRI at all. So to look at the fatty density, the breast density, you need to look at either the pre-fatsat images or the pre-gadolinium fat-satid images, which are what I'm showing you here. And I often, if I need a little help, I'll also often look at the patient's mammogram because you really, they should be concordant between the two. And you need to, obviously I'm only showing you one slice here, but you need to generally be looking at more than one slice. So we have here a patient with, I would classify as being fatty. Just a little bit in here, scattered, heterogeneously dense. And then this patient, I would say, is extremely dense. Now, obviously there is a not insignificant amount, particularly of inter-observer variability, although we tend to be fairly consistent by ourselves. But we, at the moment, we just need to do the best job we can until any automated methods of breast density assessment for MRI become available. I have to find my, where my mouse is gone. Can you take me off annotate? Please. I didn't seem to want to do it. It should be working. It needs to be taken off draw and onto the mouse, but it won't let me, it's disappeared my cursor on me. Let me see if I can drag it up and make it, there we go. All right, found that trick. Perfect, I'll clear your drawing. Thank you very much. The next assessment we want to do before we start looking for any lesions is an assessment to the background parenchymal enhancement. And this is again divided into four categories, non-minimal, mild, moderate, or severe. And as these categories increase, there is the increasing potential of having a lesion being obscured by the degree of parenchymal assessment, a parenchymal enhancement. And we also have to assess if it is symmetric, or it is asymmetric. Now, again, in the prior version of the MR byrads, we divided these into quartiles, so they were 25% or less was not a minimal 25, 50, 50, 75, and greater than 75. We have taken, they took away these quartiles and they made it more subjective. I'm not sure whether this was a positive thing or not, but it is what it is. Now the most important thing to think about with background parenchymal enhancement is it is the enhancement of the glandular tissue, not the enhancement of the breast. So for example, this first patient here has, I would say at least heterogeneously dense breast, if not extensively dense, but they have no enhancement. This patient has heterogeneously dense breast with just mild punctate enhancement throughout it. So we only look at the amount of fibreglandular tissue. This patient heterogeneously dense with moderate enhancement, so there's a fair amount here, you can see how this might obscure an abnormality. And this patient, they have scattered the heterogeneously dense breast with more extensive, severe background parenchymal enhancement. There was a paper just came out this month in academic radiology that looked at increasing background parenchymal enhancement and it found there were significantly more callbacks in these patients who have more background parenchymal enhancement, but it didn't really seem to affect the overall sensitivity and specificity, which was interesting. Now I also often like to look at the MIP images to get an idea of the enhancement. So, you know, it doesn't matter what their distribution of fibreglandular tissue is, if you've got a MIP that's like this, you know there is virtually no background enhancement going on. In this patient, this would be a mild, moderate and severe, but again, you do have to look at the patient's amount of fibreglandular tissue before you make a final assessment. But no matter what their background enhancement, their background fibreglandular density is, their breast density is, this is going to be severe and this is going to be minimal in this case. In most patients, but not all, BP is going to be symmetrical. However, you're going to have patients like this and you can see that this patient on her left breast has had prior surgery and radiation treatment, which has negated that background parenchymal enhancement compared to her untreated breast on the right. If it is asymmetric, you need to mention that in your report, but the overall assessment pattern is going to be by the denser breast. I'm sorry, the one with the highest background enhancement. If the patient has implants, you want to say if they're present, you want to say the type, saline, silicone, or if they have bilumin mixed device. But I'm not going to talk anymore about implant rupture studies. I'm assuming that the study that we're reporting here is a contrast enhanced MRI looking for breast cancer. Once you've interpreted the background assessment of the breast, the implants, the density, the background parenchymal enhancement to remember that does not include any lesions, you then want to individually identify lesions and analyze each lesion. And so these qualifiers here will be individual for each lesion. And I only recommend if you don't have one already that you build a template within your transcription system such as PowerScribe that has very clearly defined a lesion macros within it that can have dropdown boxes for all the lexicon elements I'm going to talk of. And this is going to make it much easier for you to fill these in and remember to fill them in. So for every lesion you want to say what side it is obviously, how big it is, and depending on the standard air institution, this may just be the longest dimension or it may be bi or even tridimensional. Which quadrant of the breast it's in, or is it retroareola or deep central? What clock face? Depth should really be measured in or reported in a couple of different ways, distance from the nipple as well as the thirds in the breast which I'll explain in just a minute. And if a lesion is close to the skin or the chest wall, you're probably going to want to describe the distance from those structures. So for example here, we're talking on a sagittal image the anterior middle and posterior depth and we're measuring the distance from the nipple for the lesion to the center of the lesion. Here on the axial image again, this would be anterior middle or posterior depth and obviously that's going to, this is why using this as well as centimeters can be helpful because in a very small breast five centimeters might be posterior, but in very large breasts five centimeters might be anterior and it gives us much more help to the surgeons if we use those qualifiers. Here because the lesion is close to the chest wall. I'm also measuring distance in the chest wall. Here because I'm close to the skin. I'm also measuring distance in the skin. We don't routinely measure distance to the skin and the chest wall in lesions that are a long way from them. After we identify the position of the lesion. We're going to say what type of lesion is now with the new version of buyer news, not that new. The latest version of buy rats. There are three types of lesion. There's a focus and note that is different from the focal that we're going to be talking about in a minute. There is a mass and there is non mass enhancement. So non mass enhancement used to previously be called non mass like enhancement that like has been dropped although I still have trouble saying it so I'm sorry if I still keep saying it and they also dropped multiple I think of them like this. So this is a focus. It's a little star. You can't really see discrete borders or anything like that. You can't see the edges, but it's a distinct because it stands out from the rest of the background This is a mass. It has three dimensional contours convex contours. And again, it clearly stands out from the background perencoma and is bigger than a focus and we can see edges on it. This gal at this nebula here is a area of non mass enhancement doesn't really have distinct orders that you can describe it doesn't have a three dimensional contour to it. So it'd be non mass enhancement. So let's go through each of those. So a focus is a tiny dot is usually five millimeters or less, but it's one that clearly can't be clearly defined as being another lesion. And we can only describe its size and its kinetics we can't say anything about the edge characteristics as we can with a mass. So this is a focus of enhancement. The rest of this breast has pretty low levels of enhancement, but this one focus here is standing out and you might wish to describe it depending on other characteristics, like it's kinetic pattern. And we can talk about kinetics towards the end. Masses, however, we can describe more things we can describe their shape. We can describe the margins, their enhancement pattern and their kinetics. Now mass shapes now come down to three round oval, including lobulated and irregular previously lobulated had its own categorization and there was also another one called stellate and that's been rolled into irregular. So let's look at some examples of this. I think everybody knows what around shape is. So this is a round mass of the lymph node in this case. This is an oval and mildly lobulated and how lobulated can something be when it no longer becomes lobulated and it moves into regular. Well, you know, if you've got so two to three, maybe four gentle lobulations like this one here, I would still call it lobulated. Once it starts getting more than that, then I would move into being irregular. And here's two examples of irregular shaped masses. So remember, we're not talking about the edges here, we're just talking about the shape. Once you describe the shape of the mass, we're then going to describe the margins of the mass. So the margins may be circumscribed or not circumscribed. And if they're not circumscribed, they may be irregular, or they may be speculated. So this is an oval mass that has or gently lobulated, which has circumscribed margins. In this case, it was a fibratomoma. This cancer was not circumscribed and it has irregular margins. So small points. And then this one is not circumscribed and has these longer spicules of enhancement that may be longer than the mass body itself, but don't have to be coming out from it. Now, to be honest, I don't know why they discriminated between irregular and speculated. I think what they were trying to do is make it with an increase in the probability malignancy from circumscribed through irregular to speculated. However, there is significant crossover, both benign and malignant pathologies between them. And, you know, there's a certain amount of judgment call when, when is a irregular margin mass a speculated mass. And I'm sure we would all know it may be that some people would fall onto the speculated rather than the irregular with this one. Once you've described the shape and the margins, the mass, you then want to look at the enhancement pattern in the mass, not the degree of enhancement, but the pattern. Is it homogeneous, heterogeneous, rim enhancing, or does it have non-enhancing dark scepter. And the two elements that were taken out of the previous breast tumor byrads were central enhancement and enhancing scepter as opposed to non-enhancing scepter. So this oval mass, which turned out to be just a benign lymph node, has nice homogeneous enhancement throughout. This mass, which turned out to be a cancer and our countness is being lobulated, has heterogeneous enhancement, but it has pretty well circumscribed borders. This irregular mass, which has, again, circumscribed borders, but is rim enhancing. So definitely more enhance than this, and this was also a cancer. And then this is an example, which I showed you before, of these non-enhancing dark scepter within a mass. And this is fairly pathodonomic for fibradanomas, if everything else in terms of the shape and the margins and the intensity of enhancement, the kinetics fitted with a fibradanoma, you could be fairly confident that this is a fibradanoma. Now, the reason that they took out the enhancing scepter is, you know, if I said to you, is a zebra black stripes on a white horse, or is it white stripes on a black horse? You can't really say. And the same here, we would end up with these enhancing scepter in a rim enhancing mass, or are they dark scepter? So really, they took out the enhancing scepter because it was too confusing. Now, non-mass enhancement, you may be amazed to know, is enhancement that we cannot define as a mass. So this is going to be, this can be very variable in size, but it's generally going to be larger than five millimeters, otherwise it's going to be called a focus. It should be an area of enhancement that is discreet from the surrounding breast parenchyma. You remember I showed you like that nebula, and how different that looked from the background galaxy. And it's not, we can't clearly classify it as a mass because it doesn't have three-dimensional convex type borders. So what do we describe about non-mass enhancement? We describe its distribution within the breast, its internal enhancement pattern, and then its kinetics. So we have more options non-mass enhancement distribution. So for distribution, it can be focal. Now realize this is different than focus. So it's non-mass enhancement, not a focus. It's bigger than five millimeters, but it is focal. It is linear. It is segmental, regional, multiple regions, or diffuse. I'm going to share examples of it, of all of them. There were two additional categories previously, ductile and dendritic or reticular, and these have been removed. So this is a focal non-mass enhancement. It doesn't really, we can't really call it a mass. It's not got convex borders. I can tell you it looked sort of more planar in the sagittal plane, but it's bigger than five millimeters. Here we have linear non-mass enhancement. Before they divided linear and ductile, but really anything that looks like this, even if it's got some branches in, we'll call linear enhancement. And you need to confirm it is linear in both planes. Here's an example on a subtraction image of segmental enhancement. So a segment in the breast is going to be triangular shaped or pyramidal shaped. It's going to have the pointy end towards the nipple and the wider end at the base. So this additional enhancement is more extensive than segmental enhancement. It doesn't clearly form that nice triangular shape, but it's not the whole breast. And this is all fairly consolidated into one area. Here's an example on a MIPS scan of multiple regions. There's one region over here. There's another one over here. So I would call this multiple regions. And then finally diffuse, you can't, is pretty much scattered throughout the breast. And this is what you have to be very careful to discriminate from background parenchymo enhancement. One clue is going to be if it's markedly asymmetrical with the other side, the kinetics can sometimes help, but not always. But I know that I have personally followed a patient who had what we thought was just diffuse non mass enhancement for several years before diagnosis of diffuse these low grade DCIS was made. So be careful of this is always a concerning one to me. Now we look at the internal characteristics of that mass non mass enhancement and sometimes this can be a little difficult to assess, but I'll try and give you some idea. These may be homogeneous. They may be heterogeneous. It can be clumped. And then a new one that came in was clustered ring, and they took out the stippled enhancement. The stippled enhancement is now sent to be just the background parenchymo enhancement. So this is a focus. I'm sorry. This is focal. I'm getting myself confused non mass enhancement. So it's focal doesn't really make it into segmental, but it is fairly homogeneous throughout maybe a little bit heterogeneous, but I would probably call that fairly homogeneous. In my experience, the majority of non mass enhancement is heterogeneous. So here we have multiple regions, almost is almost kind of a global generalized enhancement, but it's very kind of spotty. So called this heterogeneous. And then this is clumped and clumped always starts to make us think about DCIS because this is a much more predictive pattern for DCIS. And here we have lots of little foci of enhancement that are clustered together and they often form a linear or a segmental type distribution. And I think of clumped kind of like this. If you look at microscopy of clumping red cells, you have these individual red cells or foci and then you have these spots where they're all clumped together. So just kind of go back to now you can imagine that happening here. Now clustered ring is not something that you see very often but it has an extremely high predictive factor for being DCIS when you see it. So in clustered ring you see multiple ring shapes all grouped in one area and I think of this more, not so much like clustered ring but more like crazy paving. So, you know, think about look at this crazy paving type effect with this, or crazy paving in the lung, such as we see with alveolar protonosis in this particular patient. So think about crazy paving in the breast. Now let's move on and talk about the kinetics. So the vast majority of programs are going to use kinetic analysis by some form of software, several out there, and how they individually display what exact colors that they use are going to depend on that software but most of them are using a blue, yellow and red scheme. All the ones use green instead of yellow until they realize that people who are red, green colorblind couldn't see it very well. But I'm not going to go on so much the meaning of this as I'm going to talk about the descriptors that we apply to it. So for the kinetics and again we're talking about each individual lesion here so all these things I've talked about you need to start with each lesion. So let's talk about the shape of the curve and the shape of the curve includes the initial upslope, which is slow, medium and fast, and the delayed phase which is after the peak, which is persistent plateau or wash out. So how do we categorize the initial upslope. So a slow kinetic curve for the initial upslope is when the intensity increases by less than 50% from its baseline intensity within the first two minutes. And within the first two minutes we're generally seeing our peak enhancement and that's usually the first post enhancement image, unless you do very short interval scans. The medium enhancement upslope is going to be the increases between 50 and 100% from that baseline pre gadolinium value within the first two minutes, and then fast enhancement the intensity increases by more than 100% in the first two minutes. Now, most cats have an option where you can change between these values so you can call up your color image, your kinetic curves and you can flick between seeing 50% not very delicate mouse here. You can flick between being 50% level cut off 75% or 100 and what you're doing there is you're excluding as you increase towards 100 areas that may be enhancing with these slow or medium enhancement curves. Now as you move from to excluding the slow ones to just looking at the fast ones, you're going to increase your specificity, but you will decrease your sensitivity. And so how do I use these in practice? Well, if I have a patient who has virtually no background per ankle enhancement, then I will be looking at anything that is 50% or above. So I will look at the medium and fast, I will set it at 50% and it's only going to highlight with color on my image ones that have medium or fast up slow curves. If, however, I have a very higher background per ankle enhancement. I know there's a lot of fibrocystic or hormonally related stuff going on in the background. Then I'm going to change to having say a 75% or even if it's much higher 100% cut off because I want to get rid of some of that background per ankle enhancement and just see the lesions that are really standing out from that. So here's a graphical demonstration. This is the percentage changed in intensity. So the taking the intensity before we give the contrast media as zero. This is scan one. So it's the pre contrast his scan to which is in our institution probably going to be about a minute and a half. And we have the slow curve. You can see it's enhanced less than 50%. Here the change is nearly 80%. So it's in that medium range. And then here it's over 100%. So this is a fast enhancing curve. Once the initial enhancement has happened. We then want to look at the delayed phase and there are three types of delay phases that can be reported. And as we go similar to with this last curve that we get more and more concerned about lesions as we go up to the fast curves here. So we're going to start from blue type one, yellow type two and red type three. Again, we're more increase. I'm sorry, I'll level of concern about malignancy increases, but just recognizing there is significant overlap. There are many benign processes such as normal lymph nodes which will have these type three wash out kinetics. And there are malignant processes for example low grade DCIS and some invasive lobular carcinomas that are going to have persistent kinetics. And again, we're looking after that first two minute mark and usually that's the peak, or at least the first peak here. So what do these curves do? Well, a curve is described as being persistent. If it continues to increase. So by the end of the study, it's increased more than 10% from this first scan value. It's described as being plateau or type two. If it stays within plus or minus 10% when you look at the end of the scan here. So first we take actually five images, not four images with my study, I was just making it a little simpler here. And here this is a wash out type curve or type three when the contrast washes out of the lesion so that the end value is more than 10% less than the great at the top value. And obviously you don't have to sit and calculate this, the software is going to do all of this for you. So it was part of our lesion descriptors when we want to describe the kinetics we're going to say that it has it has fast inflow kinetics followed by wash out. If the T2 signal is increased or significantly decreased I usually mention it if it's taking part of my decision making process. On the whole, an increased T2 has a benign connotation lymph nodes for example markedly high, fibrocystic change has high T2, and that's likely to make me downrate my level of concern about a lesion, as opposed to the not being increased T2. Cellular fibroidenomas they may have increased T2 example, however there are some tumors that do have increased T2, musinus comes to mind, and we do occasionally see it even with more aggressive tumors. Just got to make sure that what you're not looking at is fluid so you don't want to mistake fluid necrosis within a tumor for a sign of benignity. Like all studies we want to compare to any prior studies. This can be quite challenging in breast MRI. You've got a lot of images to call up. Most software platforms allow you to call up to studies at once you can go between them and also slight differences in the image acquisition parameters. The volume of contrast where there was an infiltrated dose, etc, can really affect those kinetics so comparing kinetics in one study to another can sometimes be quite challenging. The patient scanned a different type of their menstrual cycle again they may have lower or higher background parenchymo enhancement, and that can affect the interpretation and identification of lesions. Another new thing that was added into the newer byrads was the presence of any concerning findings. So we look for nipple retraction, nipple invasion, skin retraction, skin invasion, axillary adenopathy should be described as well as internal mammary adenopathy on all studies, presence or absence, chest wall invasion, and pectoral muscle invasion. So if any of these are present, they should be identified and described clearly related to lesion or lesions. There are some non-enhancing findings which we may want to describe, which also this was a new finding in the latest byrads, and these are generally benign findings, but they can be confusing especially to the uninitiated who's looking at the study. So for example, you may have high signal in pre-contra on the pre-contra study in the ducts and ductilectasia or from bloody nipple discharge. These will not enhance so there'll be nothing on the subtraction image, but it can be quite prominent and look like ductile enhancement on the post-gadolinine image if you don't compare it to the pre-gadolinine image. Cis, either simple or benign, hematomas and seromas and then diffuse edema such as may happen either from post-treatment changes or from lymphadenopathy. More non-enhancing findings here, so non-enhancing masses should be described, not all cancers enhance significantly, the vast majority do, but you need to look at other morphological characteristics and decide if you are concerned about a mass. A speculated mass is still a speculated mass, same as is a mammography, even if it has fairly low enhancement. Describe dilated ducts, describe skin thickening and whether that skin is actually enhancing, as we talked about before, architectural distortion. If you see it, you need to find a reason for it, whether it's benign or you're concerned for more occult malignancy and any signal voids that you see from clips or foreign bodies. Facts with mammography and ultrasound generally speaking, fat-containing lesions are benign. They include lymph nodes, however you need to look at the morphology of those lymph nodes, even if they have fat within the hyalum to make sure there aren't focal lumps and bumps that concern you. Fat necrosis, it's really nice if there is a fat necrosis actually have fat within them, they don't all have fat within them, but you can make a diagnosis of fat necrosis with confidence if you have, for example, a rim-enhancing mass that clearly has a whole lot of fat within it. Hamatomas of the breast and then post-operative seromas that have fat. When you've analyzed each lesion and you've described them as well as all the background enhancement and density patterns, you need to put your breast into a bi-reds category. For me, because an awful lot of our breasts have cancer evaluations, a lot of them are staging exams here. We do use a different bi-reds category for each breast and that is acceptable because we just found that it really helps the surgeons for further management. What I want to point out is, although these categories are the same as you're reading immemography and have exactly the same risk patterns, so benign should be less than 2% likelihood of malignancy and highly suspicious, greater than 95% and so on, you really want to minimize the use of 0 and 3. You want to minimize the use of 3 at the best of times, but you want to minimize the use of 0 and that's because you may be sending the patient to another radiologist to evaluate them. You are likely sending them to a less sensitive modality for evaluation and at some point you have to decide on the basis of the MR if you think this lesion is going to need biopsy or not. So we only use 0 if we're waiting for old examinations that haven't yet come, for example, or if we think that the assessment by follow-up, second-look ultrasound and or mammography is likely to change down regulators. So in other words, if we really think that this is probably going to be a fibratoenoma, and if it looks like a fibratoenoma by ultrasound will be fine, not biopsying that lesion, then we'll call that a 0. If we think it's probably absolutely nothing, it's just background parenchymine enhancement, we just want to make sure with ultrasound there isn't a mass there, we'll call it a category 0. But if we're looking at an area of non-mass-like enhancement with a linear distribution, we're going to call that a 4 regardless of what is seen by ultrasound if one chooses to do a second-look ultrasound in that case, which we may not. So you really want to give the radiologist who is doing any second-look exams guidance, be careful with how to manage it if they don't see anything. Obviously, if they see something completely benign that confirms abnormality, then that's okay. You want to avoid using three, particularly in patients who are, I mean, some people say it's dumping it down the line and hoping that a different radiologist reads it in six months, and we've all done that, I can first. But you particularly want to be careful in cancer patients if you're seeing something either in that breast or the other breast, and you want to call it a CAT3, because this patient may certainly be getting radiation treatment or chemotherapy to one on both breasts, and if you see it disappear in six months, does that mean it wasn't cancer or it's treated cancer? So by red zero, if it's a technical issue or additional imaging may downgrade. Oh, one thing I didn't mention before, with by reds three, so by reds three used to be divided it for MRI into 3a in a 3b, and you used to be able to do this. 3a was an ultra short term interval follow up. And that was the patients who had been imaged in the wrong part of their menstrual cycle. So outside of days seven to 14. There was very high background for income enhancement, and you wanted to repeat it just in a very short interval of time. And that in that case, it was called a 3a, but that's been done away with. But you could in that case call it a by red zero. All right, so we're going to do we're going to bring up some polls here and see if you've learned how to apply the terminology correctly. So I'm not make asking you to make a diagnosis in any of these I'm just asking you to use the terminology so this lesion here I think everybody can see it would be classified as what type of lesion if you could bring up the first poll there so you've got a choice of focus mass or non mass enhancement. And we're going to leave these polls till about 30 seconds or 50% of people have answered. Please answer in the poll not in the question answer box. Okay, great so the vast majority 91% got that this was non mass enhancement and just 4% focus and a small number mass so this is non mass enhancement there's lots of areas of non enhancing perencoma in between it. I would probably call this regional and I would call this heterogeneous could even kind of move towards clumped perhaps. There is more than a few shades of gray when you're reading these things. So another poll, which of the following distribution descriptors is most appropriate to apply to this area of non mass enhancement focal linear segmental regional multiple regions or diffuse. Okay, so about half of you said segmental and that's what I would call it I call it segmental because I think it's pretty much triangular shape to the wider base pointing towards the nipple. And you wouldn't argue greatly with somebody who called it regional which 25% of you did. My reason for calling it segmental in this case is segmental has a higher probability of malignancy than regional and to my mind, this is pretty concerning for a doctoral distribution of enhancement. It's really too big for me to call this focal starting to get shape. And it's more segmental and is linear linear I'm going to think literally of lines. Okay, next poll. So which are the following descript distribution distribution again it's the same question as before with a different case that you would apply to this area of non mass enhancement focal linear segmental regional multiple regions or diffuse. And I'll just tell you that these are blood vessels that's a blood vessel. This is the abnormality that we're looking at. Like, a few people said focal the vast majority you said linear, and I would agree to you I really think that this is linear. If I didn't have really these long bits here just had these smaller bits here then I would call it focal. Okay, now we're talking about shape descriptors we've moved on from non mass enhancement to masses. And you only have three options here makes it a little bit easier. You've got round over lobulated or irregular. Not the margins at the moment. Okay, I called this one over and lobulated. Because it's too lobulated to really call it round. Now, would you say there's too many lobules and you'd call it irregular. That's fine. As I say this is not hard science here. There's great science but hopefully you'll start to have, you know, be pretty consistent within yourself and you can also work with your colleagues in the departments to increase your consistency. Next one. So same mass, but now we're looking at the enhancement pattern. So remember we look at shape. We look at margins, we look at enhancement pattern, and we look at kinetics. So we've got a bit of a split here between heterogeneous rim and dark scepter with most people doing heterogeneous I called this one rim. I can see why you might call it heterogeneous, but to me this really looks like it's rim enhancing, which kind of ups my level of concern for this particular lesion. So, you know, some of this, if the times that we're not worried about rim enhancement is if something's a serum or a hematoma with a nice smooth and preferably progressive type enhancement pattern, or a cyst. If we're not frequently see little inflammatory cysts you're going to see a high T2 area with a nice smooth round, usually again progressive or most plateau enhancement I wouldn't worry about those that they're rim enhancing. But for a mass like this and I can tell you this did not have high T2 signal in the middle. I would be pretty concerned about it. But again calling it heterogeneous is okay. Right now we're talking about borders. So here's our mass. Here's your three border descriptors circumscribe not circumscribed irregular or circums not circumscribed speculated. Yeah, okay majority agrees circumscribe I would definitely call this circumscribed. Right, let's look at some curves. So we're going to, we want to look at both the early here and the late phase of this curve. So let's decide which of these curve parameter which of these kinetic descriptors should be used in this case. And most people voted here 62% for fast then plateau and I would agree that's fast then plateau it comes up to about 170 comes down to maybe just under 160. If it probably fell off a little bit more than we would call it wash out but I would call this fast and plateau. So you don't have to sit there and calculate these amounts in real life. How are these margins best described case so we're not doing shape between margins now circumscribed not circumscribed irregular not circumscribed speculated. And again 83% say speculated a few say irregular and that's fine this has got just too much to my little pointy star bits coming out of it for me so I would end up calling this speculated. It's less than that sort of more like this stuff if it was just going to be a regular. So we're coming back to type of abnormality again here focus mass or non mass enhancement focus. Yeah, I agree it's too small to be called a mass this is probably only like four to five millimeters. And you obviously have to look at this and look at whether there's other ones in the breast to decide if you're particularly concerned about it or not. I always talk about the residence about the you're looking for the black sheep in the family. So if you have somebody with a lot of scattered areas of uptake throughout the breast so they have recently high background perennial enhancement. You're looking for that focus, which doesn't fall in with the rest it looks distinctly different from the rest or it's kinetics are distinctly different kinetics may or may not be that helpful on these little foci to be honest. We're talking about the enhancement so it's a mass we're talking about the internal enhancement pattern. Here are your options homogeneous heterogeneous rim or dark safety homogeneous good I think that's pretty easy one and this although I'd probably say it's over slash it in shape has circumscribed borders and homogeneous enhancement this was a metastatic lymph node internal memory node. Okay, we've got three more questions. Which are the following enhancement descriptors so we're looking at non mass enhancement again. We're looking at the internal enhancement pattern of it I would call this regional. Possibly looks a little segmental but it looks too big to be a segmental so I'd probably call it regional enhancement heterogeneous yeah and the people who said clumped. It's not quite distinct enough to be clumped to me it's definitely heterogeneous maybe this is a little bit more clumped. But you know as I said there's a lot of overlap between these appearances and all these categories sound great when you're doing them until you have to sit and apply. I agree so slow and then progressive was the highest vote. So it's going up it's only gets up here to, you know, like 55%. And then it continues to increase and it's clearly increasing more than 10% from this value here so this slow so this would be a very benign appearing curve that generally speaking this is something particularly worrying about the lesion, I would be ignoring. And then one final one. So again, we're looking at the enhancement characteristics. So this is linear non mass enhancement. So we've got quite a variation here and I sort of went more with clumped because I think there's a clumped here and a clump here and a clump here but you know I wouldn't certainly you know contest it if you wanted to call this homogeneous. I think one thing is a linear heterogeneous is a little difficult to tell from clumped to my mind. All right, so that's the end of what I have to present I hope it's given you an idea. I can I probably got 10 minutes or so that I could answer questions if anyone has some I've just got to work out how to do that. Perfect if you don't mind. It's in the Q&A section. I will let you just hop right in since I know your time is short today. Okay, let me get this. Oh, we have a whole load of them. Okay, I'll try and do as many of these. So please talk about the role of MRI in the detection of calcifications. So it should not be used to detect calcifications mammography is the standard of care for detecting calcifications. Calcifications have been biopsied and you've decided if they're malignant or benign it can then be used for staging and obviously for patients who have high risk screening for DCIS but it's not for detecting calcs. In what phase do you assess background perencoma enhancement phase one or phase two or subtraction. I'm going to use the subtraction. I think it's the easiest one to use. So I'm going to use subtraction but I look at it alongside the the pre gadolinium fat suppressed so I can work out how much breast perencoma someone's got. Okay. How to standardised where the contrast given is sufficient or the rate is enough for the mean. Curve evaluation. So I think what you're talking about here is when you have infiltration of the contrast or, you know, the tuming's gone down it doesn't. I look at the blood vessel I don't have a. I don't have any magic number to give you. But what I look at is I look at the cavity of the heart and I look at the blood vessels. And I make sure that on a subtraction image you're really seeing those vessels stand out clearly and you're seeing the left ventricular cavity stand out very clearly because it should be if you're not seeing that heart there. There's sufficient contrast being given. If I have any question then I will put a little region of interest over a blood vessel and just make sure it has you want it to have a, you know, a night blood vessel should look something like that. If your blood vessels kind of looking like it's a TARDIS parvis curve then that's not going to be adequate contrast. Somebody says how does clock face work for breast. So it's exactly the same as it is with mammography. So, you know, here's the left and the right is if you're looking at them and each of them has a clock face such that, you know, 93, 93. And it's exactly the same as that is just that you have to use a little bit more of your imagination and putting together the, the 3D nature on MRI. Again, as usual, you're going to be looking at the nipple all of the clock faces in mammography. I hope 37 questions not a hope in hell. I'm sorry. I'll do what I can. You're going to be looking at I'm sorry, I just lost track of my thoughts a little bit. You're going to look at it relative to the nipple. So on the sagittal MR you've got your nipple here and you'll be able to tell if it's in the lower or the inferior upper or lower quadrant and you'll check on your nipple here if they haven't been placed very closely. And again, you're going to look at your inner and outer and you take it from that. Most of these software systems will allow you to put a mark the lesion mark the nipple and it will tell you what that clock face is. How do you differentiate? Sorry, I'm just typing nothing. How do you differentiate between the size of the mass on ultrasound, MAMO or MRI and which is most accurate. So you're going to have to take this as a case by case basis. Generally speaking, masses, malignant masses frequently appear larger on MRI than they do by ultrasound. And the MRI size tends to correlate, not perfectly, but it correlates with the size at surgical pathology better. So to my mind, my size on the MRI is going to trump the size on MAMO or ultrasound. But I do, and I usually indicate that if I say, you know, please note this mass looked as if it was two centimeters by ultrasound but is significantly large on MRI because that's going to obviously affect the surgical management. Somebody asks, is a BPE and non mass enhancement the same? No, they're different. Background parenchymoid enhancement is the normal physiological enhancement. It's due to the combination of hormones and some fibrocystic disease and some things that we don't know what the hell it is. And that is going to be there. It's going to be in both breasts usually unless we have those asymmetric patients I spoke to before with ones that had breast treatment. And non mass enhancement is something that's got to stand out over that background enhancement. So the images I showed you of the, probably go back to it, but the stars and the nebula and the planet that those nebula stood out from the background galaxy and that's what you're looking for. I'm not going to go into that one. How to differentiate heterogeneous enhancement. Somebody's asking about differentiating heterogeneous enhancement from, I've lost my mouse again. From, what was that, from non enhancing dark septic. The data septic very clearly look like a stripe. Okay, so think think about the zebras rather than leopards. What is the kinetics of normal breast enhancement is going to depend on the age it may be sub threshold. In other words, we said that we would set that threshold that most softwares it's got to enhance at least 50% above the baseline pre gadolinium signal before it even color codes. So most of it's going to be white if you like so it's going to be non color coded only when it hits that 50% is it going to start to kill it to color code. So most normal breast enhancement is slow and progressive and really most times it's sub threshold in terms of the kinetic analysis. Okay, let me go down some more here. MR by Reds category for abscess is by Reds to the same as it would be on ultrasound. What's the difference between segmental and regional segmental it's really got to form that triangle so we're really expecting to see the that's my thing on. To do that. Sorry. Come back into this. Here so we really expect to see for a. A segmental that it's doing. It doesn't want to draw for me at the moment in mind. So segmental it's going to look at a triangle it's going to have a narrow point to it towards the nipple of regional does not necessarily. What category would you give in cases of post neo adjuvant chemotherapy complete response we still call it by rad six, because we know there's still cancer there but there is that may vary a little bit between different symptoms, but we normally say by red six with a complete response by MRI, because you know we know that when they go to surgery they can have cancer there. Are you doing con no I'm not doing that. If you're an Asian suspicious and ultrasound has to go to biopsy then why do you want to do an MRI totally agree with you. MRI should not be used if you see something suspicious on mammography and ultrasound. You do not use MRI to decide if you're going to biopsy that you should be using all of the ultrasound and mammography parameters that we're already very familiar with to make that decision. MRI they go to MRI in our institution for staging of that breast and the contralateral breast because about 15% of people you're going to pick up another lesion in one or other breast after they've had a biopsy. How to differentiate clump and clustered rings clustered rings really do look like rings. So they remember that crazy paving thing that I showed you bring up again here. So clustered rings they really do look like rings. Okay. And clumped is going to look like lots of little bits stuck together it's that rule of the blood cells there so there's no rings here. Most of the time a cycle for MRI in a premenopausal patient ideally between days seven and day. My shared window. Day seven and day 14. We do not delay, we do not delay our breast MRIs for staging a breast cancer waiting for the optimal time the menstrual cycle because there's just so many delays in this patients care if you start doing that. How to remember these terms in describing lesion someone said okay so this is when you need to make your macros so we have when our we open up a breast MR and macro opens up in we use past crime. Not advertising and the macro opens up that we've built that has dropped down field so it's going to have I should have stupidly. I should have bought a snapshot to show you that but it's going to have it'll say it will have all those different things like background for enhancement and then when you click on that is going to have the drop downs for the mild moderate to extreme. It when you click on the breast density will have the drop downs for that. And then there's a descriptor field for us to put in a verbal description of each lesion but then below that we have a box that says you know lesion one. And then side distant size distance in the nipple quadrant blah blah blah blah blah and then all those other parameters and we have one for masses and we have one for non mass enhancement. So they're all there so all we've got to do is just click the correct fields you don't have to remember them. The ACR also has nice little cards that you can either order from them that have it's a two sided plastic card that's like maybe eight by four inches. And you know we have those on the desk particularly to help residents look at it so it's not going to have all the pictures like the big by Reds Atlas but it's a nice little reminder and you can just print those out from the web as well. When we find multiple foci enhancement how can we be sure which is one is abnormal. Okay well this is, you know this is the, the painful thing about breast MR this is, you know one of those breast MRs that you, you know you open up and then you're kind of look around to see if your colleagues have seen you because you're tempted to, you know, close it off for them to answer, never done that of course never done. And this is the black sheep thing so you want to see if you have multiple photo size of enhancement you want to describe, think about are they, did they all look like themselves, each other are they all pretty much the same size are they all pretty much the same shape are they pretty much all the same kinetics of enhancement or is the one that's sitting there that is, you know, it's bigger, it's got wash out enhancement or it's got perhaps it's got fuzzy borders to it where the others all nice and clean so you're trying to think. What is, you know, what's the black sheep in the family, or do they all look the same because otherwise you end up biopsying an awful lot of fibrocystic disease. Short term follow up, what is short term follow up yes that's six months. Active conditions of the breast really shouldn't be evaluated with MR. Every now and then you're going to have one and it's going to look like if you have a diffuse mastitis like a diffuse mastitis can look like inflammatory breast cancer on mammography is can look like inflammatory breast cancer on an MR but I'm not going to, I don't want to talk about that right now. Fibradenoma by reds two or three. If I'm convinced something's a Fibradenoma I'm going to call it a two. Let me just see what else is here. Software for kinetics and subs we use dinocad. There's not that many options out there. We used to have a logic product which is now gone out of the market so using dinocad which seems to work pretty well. Sorry, I'm trying to read a lot of things very quickly here. We cannot, someone asked about granulometous mastitis by MRI we cannot make that diagnosis by MRI it's going to require biopsy. By reds for positive margins on post-op study but no concerning ML findings so this is, I'm still going to read it by the MR in that situation so even if I know there's positive margins but I don't see anything, then I'd be calling it a by reds two because of the hematoma serum. How we doing for time here? I think we're out of time guys. Feel free to email me if you have any specific questions. And I think they did record this. You are correct. As we bring this to a close I wanted to thank you Dr. Lewis for your time today. I know that you're a little pristine you have to get back to work so we appreciate you being here. And thanks to all of you for participating in our new conference. A reminder that this conference will be made available on MRI online dot com in addition to all previous new conferences. Please join us tomorrow. Dr. Mark Gosselin will be with us for a new conference on pulmonary hypertension. You can register for that and all the others this week on MRI online dot com. And follow us on social media for updates and reminders for upcoming conferences. Thanks again and have a wonderful day.