Hello, everyone. This is the Bio-Ed 6th edition. All right. So we're going to start with mammography. And I wanted to start with where we are with the current status of the 6th edition. Every section has been reviewed. We've looked at the evidence, expert consensus, and expert opinion. The review process included approximately a year with each subcommittee. Each subcommittee has five members. And then approximately a year after that with the subchairs all together that were up here in our committee chair. And then about several months with the full committee, all the members. And now it's gone through several rounds of additional groups of outside experts in the U.S. and also international. We just received the last set of comments. So the format has been built to optimize search for a particular subject. And we've updated the atlas overall. And to highlight new content, all of this, though, is preliminary until it's published. There's been a real movement to move towards modality neutral and structure clinical indications. So you will see that on the left you see the major indications, such as asymptomatic screening, diagnostic workup, and then diagnostic current breast cancer. In every major indication there's optional subcategory indications for asymptomatic screenings, elevated risk, dense breast, completed treatment for breast cancer, and then any relevant history to report, such as the gene mutation or risk assessment model, if you've used one. With diagnostic workup we included clinical findings, imaging findings, follow-up or probably benign BI-RADS 3 findings, follow-up after biopsy, implant assessment, and then again, any clinical finding or imaging finding. And then for current breast cancer diagnostics, the extent of disease before definitive surgery or response during or after neoadjuvant therapy. And here, location, the relevant history to report is important, such as location and size of cancer on any prior imaging evaluation. So why are modality neutral structure clinical indications important? Well, they will allow us to have vendor templates and structure reporting that's streamlined. Also, it will help us get coding and billing correctly, which is important for us in clinical practice. And it will allow us to collect data on method of detection, so we can retrieve what type of modality the cancer was detected and whether it was screening or diagnostic or some other format. Now for mammography, the updates from the fifth edition, all the images and the lexicon and illustrations have been replaced and expanded. The examples no longer include any screen film, but we have now fulfilled digital tomosynthesis and synthetic mammography. And you will see throughout the sixth edition that we have captions, including capital letters, perhaps a descriptor that we're interested in, such as in this example, a mass oval. But we also have the margins, what the density is, and whether there's any pathology. This was a simple CIS, but if there's pathology available or stability, we'll give you that information. We've added quite a bit on the guidance chapter on tomosynthesis. We wanted to help the radiologists interpreting with the management of several of the clinical scenarios that we may face in clinical practice daily. And we want to have consistency across breast imaging practices. So you can see that we've defined screening and diagnostic and added it to include tomosynthesis, and also the benefit of lower recall rate and higher cancer detection. We've made comments throughout about better margin visualization and also how it improves localization of findings. We have lesion management in the guidance chapter, because previously when we published the tomosynthesis supplement for the ACR, we're unable to say that if you see a lesion in only one projection, it's a mass. So we couldn't call it a mass previously. But now we say that if you see a lesion on a tomosynthesis projection, it can be a real finding, even if it's only on one view, because of the increased conspicuity, the localization improvement with lesions on tomo. So this may eliminate any additional diagnostic views for you. Now breast density in the fifth edition, which was 2013, from the fourth edition, we classify breast density. We went from percentage of dense tissue to base it on the likelihood that there's an island of dense tissue that can obscure a cancer. So this principle is followed in the sixth edition. So if there's any region on the mammogram that's dense, potentially you could miss a cancer in. You have to upgrade that to the heterogeneous dense pattern, even though the rest of the pattern may not be like it. So if you have a breast that's more dense than the other, you have to assign the greater density of the two. We also added things on breast density about risk. So we have breast density is relevant for two reasons, mammographic accuracy, and also the risk. So we have the masking effect, but we also have an increased inherent risk for developing breast cancer. So we did add that in, and we've also added some relative odds ratios of breast cancer risk based on breast density. Now we had changed A, B, C, and D assessment by risk categories for a brief period of time, and we're very happy with the changes, but the FDA came out with strict language that we have to use in September of 2024. So MQSA final rule says we have to say A, B, C, and D exactly as they are there. We can't change it. Even if you're only talking about one breast, you have to say it that way. So we changed it right back for all 50 states in the United States. So the sixth edition mass descriptors, so we added lobular for a while. It's now become lobulated. So from the fifth to the sixth edition. In the fifth edition, lobular was removed from the fourth. So I want to say that this is an addition again. This is not something new. We just brought it back. Because it can be confusing with margin, with microlobulated, we took microlobulated out. And so we call a lobulated mass something that displays one or more indentations resulting in undulating contour, usually benign, but it may be a little bit more on the suspicious indeterminate side than an oval mass. And like I said, the margin descriptor microlobulated was discontinued. These are the changes of the sixth edition calcifications for typically benign. We took out popcorn-like, and now it's coarse only. Dystrophic is included in coarse. And the milk of calcium has been renamed layering. And punctate has been removed, and it's now under round. And these changes, so we took out popcorn-like, it's under coarse, like I said. Also dystrophic is under coarse. The definitions haven't changed. You can turn the lights back down, because there's some images. Whoever turned them up, thank you. So the definitions of layering calcifications have not been changed versus milk of calcium, but we've changed the names. Punctate was a subset of round calcifications, and now we just took that out. It's not reasonable to try to measure less than 0.5 millimeters. So we describe it in the round. Asymmetries, you see here, developing asymmetry has been removed. And this was a thoughtful change. We've discussed it quite a bit. But it's consistent with the lack of temporal comparison in the other aspects of the atlas. We don't say developing mass or developing calcifications. And we clarify, though, just because the term was taken out, if you see something that is showing a change, a temporal change over time, that's a suspicious feature, and you have to work this out. Dilated ducts, we've added multiple, and obviously solitary. And so we did review the solitary dilated duct scenario in the fifth edition. If you could not define a benign etiology, a biopsy was recommended. We really looked at the literature on this. So if you're a screening patient, it is a rare finding. And if it's not associated with any suspicious imaging findings or clinical findings, it's benign. However, if you have a baseline patient and you feel uncomfortable letting go of that solitary dilated duct, bring her back. Do an ultrasound. Do spot views, whatever you do to make yourself feel better. I don't think that's unreasonable at all. If the patient has a bloody nipple discharge or they have other symptoms that are clinical or imaging, obviously you have to work this up. This is for screening asymptomatic. We've added a new section. We removed lymph nodes from associated features, and we've added lymph nodes that are intramammary, normal or abnormal, and axillary. You'll see that throughout each modality. And under normal and abnormal lymph nodes now, we included some management clarity, what to do for symmetry with the other axilla, review the record of the patient for any autoimmune disorder, and then obviously ultrasound if you have unexplained indeterminate abnormal nodes. I'm showing you some examples of metastatic, reactive nodes, and infection or CLL. So in the sixth edition, we removed axillary adenopathy, architectural distortion, calcifications now on their own. We've left skin changes, nipple retraction, skin thickening. We've also added a new section on implants and gynecomastia. We're defining where the location of the implant and the type of implant, and historically they were called retroglandular. We're now naming them prefectural. Gynecomastia has also been added. We do talk about management causes and appearance of mammography and guidance on reporting. We've also added under special cases mastectomies and described scenarios where mammography may be reasonable to be performed on the mastectomy side. FDA requires an assignment of breast density. We have additional discussion on what drives the BI-RADS, the primary finding, but also you have secondary findings, and we define secondary findings versus associated features. So if there's a suspicious mass, you're going to do an ultrasound biopsy. If there are associated calcifications, that's a secondary finding. But you may also have some skin changes, and those, you know, are associated features that you will define in your reporting. We've talked about non-surgical management of atypical findings. Practices are different, and this is evolving in the management of high-risk lesions. If you have a non-resected high-risk lesion that hasn't changed, you know, then you give it a BI-RADS 2 over a defined period of time. However, if it's changing and increasing, you're going to give it a BI-RADS 4, recommend repeat biopsy. So I'm going to highlight some of these changes for you. So this is a solitary dilated duct that you can see, and I do believe I can play this, hopefully. Maybe. Is it playing? Yeah, there it is. There we go. So you can see on tomosynthesis, the MLO view, you have a solitary dilated duct on this patient. So how to deal with that? Well, we're lucky to have prior films. So we have the prior year, a couple years before, so it's stable. This is a BI-RADS 2. You leave this alone. As opposed to this case, let's say it was exactly the same imaging on the mammogram, but in this case it was a baseline. So it's reasonable to bring her back to an ultrasound. We found the small mass in the duct. We biopsied it. It was benign, dilated duct stromal fibrosis. This is a patient with enlarged nodes. She was called back. She came in for screening. So you may do spot views, or you may go straight to ultrasound. Whatever it is your practice is, it's bilateral. On ultrasound, obviously, this is an abnormal node. There's no recent vaccinations in her record or illnesses. So we do a biopsy, and this is a lymphoma. Of course, you know, we're going to look carefully. If something in imaging is new or increasing, we're going to have to investigate that. So because bilateral adenopathy can be benign, systemic conditions, and non-breast malignancies, many different things, vaccinations, and then you will report it appropriately. You know, you may say abnormal nodes. If you don't have any reason for those nodes to be there, then you report location, their number, and then you give the appropriate bi-rads and you bring the patient back for a biopsy. This is a non-resected high-risk lesion. You see microcalcifications here in this patient. We biopsied this. This was flat epithelial tibial stereo biopsy. The patient and her surgeon chose imaging surveillance and not surgical excision. So now she comes back. Fourteen months later, this is stable. Whether you choose 14 months or 18 months or two years, patients are not always compliant coming back. At some point, you can let go of this if it's not changing to a bi-rads 2. If it increases, you call it a bi-rads 4, recommend surgical excision or re-biopsy. If you think that post-biopsy changes are relevant in that case and it's still benign, give it a bi-rads 2, let it go. And I'm running out of time. I want to show you this case really quickly if I can. Multiple findings in this patient. And let's see if I can make this play. Maybe? Okay. So this patient, as you can see, has quite dense tissue. And there's masses, architectural distortion. We have skin thickening. We have trabecular thickening. We have nipple changes. So there are multiple imaging and clinical findings in this patient along with a lump. So this patient has a lot going on. So let's see if I can move this forward. Okay. So multiple findings. Primary finding are the suspicious masses, architectural distortion, associated features, the skin thickening, inverted nipple. So you will report that appropriately. By adding primary and secondary findings that are imaging as well as associated features, you're giving a more thorough, more nuanced report to the referring physician and also to the patient as per the abnormalities. And this is a breast density evaluation. Really quick. You see you have this asymmetric islands of more dense tissue in the upper outer quadrants. You upgrade that to heterogeneously dense. It doesn't have to be that the whole breast has to be more dense. We're going to upgrade it based on any island of dense tissue. Our assessment will be based on that, not the overall tissue. And then this is a male patient with gynecomastia. And so we have to report breast density on male patients on mammography similar to the female patients. And thank you for your time. Good afternoon. So updates and some interesting cases. Disclaimer, you heard that we're still working on it even though it's getting to the final stage. I want to focus here your attention on the right side of the screen. That's how I kind of think of the overall approach to the BI-RADS Atlas. That is that we have changes in the lexicon. We have new lexicon. We have new constructs. And we have expansion of prior topics. And I'm going to show a little bit of highlights of each just to get a flavor. Certainly in the 12 minutes, you know, we don't have time for the entirety. So first, I'm going to just start with a new construct. This new construct was quite controversial whether we should include it or not. And while it is still, you know, in the work in progress, I do believe that it will be included. And this is the concept of glandular tissue component or GTC. In the current edition, which is the fifth edition for BI-RADS or really the second edition for ultrasound, because remember that the edition first, second, and third did not include ultrasound or MR. It only included mammography. So we're really dealing with the second edition only of ultrasound at this time. We talk about tissue composition at ultrasound. And that is homogeneous or heterogeneous, fat or fibroglandular tissue. Now, this will be an adjunctive component. And we call it glandular tissue component. We try to base as much as possible on the literature. And this is the sort of the classic article that our concepts are based on from Korea. So the idea is that in a normal breast, you have the mammary zone, fibroglandular tissue. And in it contains your lobules, terminal ducts, your TDLUs, in other words, along with the tissue between the stroma, interlobular stroma. So you can kind of think of this as the active part of the breast, as opposed to the more superficial skin, cupule ligaments, subcutaneous fat, and the retromammary fascia and fat and the muscle behind it, posterior to it. So focus here on what the action is, if you will. So when you look at ultrasound, what happens is that the terminal ductal lobular units appears relatively hypoechoic, darker, compared to the interlobular stroma, which is relatively hyperechoic. So you can see in this schematic depiction here of the four ultrasounds, we're proposing that we're calling the percentage based on quartiles less than 25, 25 to 49, 50 to 74, and then 75 or greater. So this is a little bit of a visual examination, just like when you do breast density on a mammogram. So this is considered to be about greater than 75% or greater of hypoechoic TDLUs compared to the more echogenic interlobular stroma. And of course, the minimal is the other end, much more echogenic stroma. And why is this important? Because literature has been showing that with marked glandular tissue component, that you have higher risk. And while this is not expected that it goes into every ultrasound report, we thought it would be a good idea to include it to start sort of collecting data, to see how it works in the clinic, to see whether this is an important part of breast ultrasound. You know, try to get more out of it and not just use ultrasound for, you know, cyst versus solid, to really get biological information. And I kind of think of this as a parenchymal background enhancement on an MRI or on contrast enhancement mammography. Okay, so on that note, that's a new construct. Now let's go on to talk a little bit about changes in lexicon and new lexicon. So morphology, we know, is important in evaluating breast mass. That's true in mammography, true in ultrasound, true in MR. And as Sheila already said, where we, including lobulated as a shape. Ultrasound, like mammogram, this is going to go somewhere between oval and irregular. And let me show you an example of that. What we have here are MLO and CC mammograms. Patient presented with a palpable lump, which is demarcated by this triangle here, you see. And this mass, what oval circumscribed. And it turns out that this mass had been stable at ultrasound mammography for a number of years. But patient feels now, you know, maybe there's some change in her physical exam and she's coming to attention. There are some coarse calcifications, which have been stable. But interestingly, there's actually another mass, right? You see that, yes? Anterior posterior, kind of posterior part of the breast, which the patient did not palpate. And that's kind of the magic of mammography, right here by the red arrows behind there in the upper outer posterior breast. So you can see that the margins are suggested that indistinct, maybe speculated. So let's look at that. This is the yellow mass, which is the one anteriorly with the yellow arrow. This is oval versus the one in the back was this. The shape, some of us may say irregular, look kind of roundish. Certainly the margins are very different, not circumscribed, right? And of course, the yellow here is your fibroannoma. And this was biopsy in the red square. That was an invasive ductal carcinoma. So mass and mass, different, similar, but different and very different in terms of what they stand for and what they represent. But additionally, in this patient at ultrasound, she has two additional incidental masses. And this is my example of the lobulated mass, okay? And this also got biopsied and this was another fibroannoma. And I really show you this as introduction of what we mean by lobulated in this case. And she had yet another mass, which we would call oval, which was another fibroannoma. So to show you how morphology is important for assessing breast masses and lobulated, okay? Another kind of change in lexicon is that the new term will be mixed solid and cystic. And the differential, of course, includes a cancer, formerly known as the complex cystic and solid mass, okay? Now it's going to be called mixed solid and cystic mass. So this is an example of a mixed solid and cystic mass. There is cystic parts and there are hypoechoic components. When you are questioning whether this is debris or actually solid materials, color doppler can be helpful. And here you have vascularity demonstrating that the hypoechoic components are actually tissue and not just debris. And this at biopsy turned out to be a triple negative breast cancer, mixed solid and cystic mass. To show you the entire complement, what we have here is mammography, MRI, and then here's actually CT showing you this. The hypoattenuated components here are the cystic parts. Ultrasound, again, term will be mixed solid and cystic as opposed to complex cystic and solid. And another term will be the introduction of the term echogenic rind, okay? The term rind has been a little bit controversial. That's what we're standing with at this moment to kind of give a, you know, a little bit of a shocking effect because this is meant to show you something that is suspicious for malignancy. So echogenic rind, what is it? It is a thick band of echogenic tissue surrounding all or part of a breast mass. This is the opposite of the echogenic pseudocapsule. The echogenic rind is less sharply demarcated, it is thicker, and it is more variable in thickness as opposed to echogenic pseudocapsule. I'll show you an example of that. It's the opposite. This is very sharply delineated and thinner. It is important that you include the rind part in your measurements. So we're going to recommend that you have a choice. You can say that without the echogenic component, it measures a certain size. With it, it measures a bigger size, or you can just give one size. And if you do give that one size, give the one that includes the echogenic component. Why? Because it's been shown to correlate better with pathology. So this is the echogenic pseudocapsule, which is opposite from the echogenic rind. So the pseudocapsule is thin, and usually it's very well demarcated, and usually the mass is oval, okay? And of course, your classic fibroid anoma is one that the echogenic pseudocapsule goes with. As opposed to this, this is the rind. This is the invasive ductal carcinoma. You can see that this is fuzzier, less well demarcated, less sharp, thicker, variable. Some parts are thicker, some parts are thinner, and the mass can be any shape, including a suspicious shape such as this one. Okay. How about new constructs? So I think this is probably the biggest construct, and I would say that this is sort of the biggest change in the fifth edition. Excuse me, sixth edition. What am I saying, fifth edition? Sixth edition, okay, of biorads for ultrasound, at least. We're talking about ultrasound only now. This is a non-mass lesion. What is it? Well, given the current advances in ultrasound equipment capabilities and knowledge of breast ultrasound, there is increasing awareness and identification of abnormalities at ultrasound that do not, do not rise to the criteria of a mass. I don't want to date myself by telling you when I first learned ultrasound, but back in the day, it was kind of like, is this a cyst or is this a solid? Okay. And of course, both solid or cystic, mass or masses. Okay. So those are all masses. So we're kind of looking at the more nuanced, you know, borrowing Tula's words, of nuances in imaging. So more nuanced finding. And we think that especially with the advent of additional modalities like MRI and like contrast enhanced mammography, where then you go and do quote unquote second look or, you know, image directed ultrasound, we're finding all kinds of things, you know, that may not be a mass, but that has significance because they could be a cancer. So that's what we're talking about. Now, a mass is space occupying, seen in at least two different imaging planes versus a non-mass lesion is identified in at least two imaging planes, but it lacks the three dimensionality or conspicuity of a mass. It doesn't have that oomph, you know, that is the mass. And it may not have a characterizable marginal shape, which are classic features associated with a classic mass. Non-mass lesion, our proposal is that going to be four categories or four considerations for it, which is echogenicity, number one. The words are going to be hypoechoic, isoechoic, hyperechoic, and mixed echogenicity, just like a mass. Distribution, this is very akin to the parenchymal breast, background parenchymal enhancement of MRI or of contrast enhanced mammography. And where else have you seen these words? Distribution, right, of breast calcifications. Okay, so those we try to be very aligned and because it makes sense anatomically, you know, in predicting malignancy. So we have the words regional, focal, and these regional especially tends to suggest benign. Linear, you're getting to the suspicious category. And lastly, segmental, which is a suspicious term. Anywhere you use it, be it mammographic calcifications or contrast enhanced mammography, or in my case, we're talking about non-mass lesion at ultrasound. Now shape and margin, which is so key for the description of a mass, not so applicable when it comes to non-mass lesions. And lastly, orientation, and the final words we've chosen are parallel and not parallel. We've debated with words like anti-parallel versus others, and it's a long story to why those were not selected, but parallel and not parallel, okay? It is often subtle and detected only because of the disruption of background tissue, while it's identifiable in two planes only, it may be visualized primarily in one plane. Size can be difficult to determine because it lacks definitive shape or margin. And again, segmental distribution, highly predictive of malignancy. That makes sense to us common sense wise, but it's also shown in literature, okay? And interestingly, common things are common, where you hear hoofs coming in, you think horses are not zebras. And when you see a non-mass lesion that is a cancer, we're thinking of invasive ductal carcinoma, just common things being common. But interestingly is that invasive lobular carcinoma and DCIS have been shown to be disproportionately high in such non-mass lesions. Again, evidence-based shown in the literature. So let's look at a case, case number one. And like masses, if it has a clinical symptom, like a palpable lump, the chance of having a cancer is higher. And this patient had a palpable lump and her calcifications morphologically were pleomorphic and distribution wise was segmental. And the yellow is supposed to point to the posterior part, the red the anterior part, okay? So suspicious finding. We then go on to do ultrasound of the calcifications of the region where the calcifications reside and we have this. And we believe that this is a pretty good demonstration of a non-mass lesion. You have something, something, it correlates with the calcifications, but it's not a mass, okay? Here's the nipple I'm pointing to here. And to show you additional images of that, you can have associated findings, just like a mass. This is the echogenic rind. This is post-acoustic shadowing. And additional features suggest that it is malignant. And you have additional findings like these foci, which are calcifications. You have, these are calcifications here. You have hypervascularity. Those additional associated features increase the probability of malignancy. And indeed, of course, this was a cancer. Palpable, okay? And also the number one rule, if you will, to a non-mass lesion being malignant is the fact that you see a correlate in other imaging modalities. And in this particular case, we saw it at mammography because those, the calcifications were the mammographic correlate. So this was a case of high-grade DCIS. Case number two, and this will be the last case, patient had a palpable lump, which is here in the upper inner posterior breast. Again, palpable lumps increases the probability of malignancy. And this is meant to show you this finding, which is a subtle focal asymmetry that was a bit different from prior images, okay? Formerly known as the developing asymmetry, if you will, okay? So there's that. And we're going to go to ultrasound of that. And here's the ultrasound. Show you the longitudinal view. You know, in the transverse plane was quite difficult to detect, so it didn't have that oomph of a mass. But certainly has something, something, and we had a palpable lump. We have something at mammography. And it shows hypervascular along that distribution, you know, along that line there. So we thought this was the correlate. We then do ultrasound guided biopsy, which is the biopsy clip at the outside images that you see here. And you can see the biopsy clip correlates with the initial mammographic finding and with the clinical finding. So we do believe that the ultrasound and the mammal were one and the same, represent the same lesion. And here's the MRI finding. Sagittal on top, axial on the bottom, okay? I will call this non-mass enhancement in MRI. And our pathology showed, anyone guess? Because we're in a classroom, of course, we're going to think lobular, we're going to think DCIS. In this particular case, it was three centimeters of DCIS. There was a bit of microinvasion, like several foci, and there were a bit of invasive ductal carcinoma, but they were microscopic. So what I believe we're seeing with our eyes at imaging, at ultrasound, at mammography, and now at MRI, were the DCIS, the ductal carcinoma cytocomponent, okay? So lastly is the expansion of prior topics, and that, I mean lymph nodes. Lymph nodes were talked about before, but a lot more now, especially in ultrasound and MRI. Why do we do that? Because it's important. It's part of the N of TNM, and therefore part of the clinical stage, and part of the staging means prognosis, treatment, et cetera. So the big thing is that the different levels of the axilla will be introduced. Pectoralis minor muscle defines the level one axillary nodes, which are shown here in green, level two in red, level three in yellow. For the purpose of MRI and ultrasound, we're also going to talk about the internal mammary nodes here. And for the purpose of ultrasound, we're going to also talk about the supraclavicular node, which is here, internal jugular vein, the clavicle, and the omohiori muscle, the triangle, definition of the supraclavicular node. So I like to say that with ultrasound, these are all the nodal regions of TNM when it's still regional disease and not M1 metastatic disease, okay? And those will be all expanded in the ultrasound section. So here we go. This is meant to be an internal mammary node, adenopathy, supraclavicular adenopathy, and here is an axillary node with a compressed hilum. Three millimeters will be the cutoff, as demonstrated in the ultrasound literature. I know that MRI will have something else, and we're looking forward to expanding the use of ultrasound in nodal staging of our patients. So on that note, thank you very much. Okay, so I'm talking about breast MRI. Let's get going. So these are the updates I'm going to cover today. Most of the time, we're going to spend on the lexicon and reporting system, but really quickly, and this was already discussed, that there are structured clinical indications now across modalities. And I just wanted to mention that for breast MRI specifically, most of the time we're going to be interpreting the exams that are either in the asymptomatic screening category or in the current breast cancer category, either for extent of disease prior to definitive surgery or response during or after new adjuvant therapy. The acquisition parameters section of the breast MRI subsection has been revised and expanded. Expanded in the contrast-enhanced MRI section to further describe a standard full protocol contrast-enhanced MRI, which typically includes a pre-contrast localizer, T2-weighted image, and some kind of pre-contrast image without FATSAT, often a T1, and then a pre-contrast and at least two points contrast, T1-weighted series. This would be your DCE. Abbreviated contrast-enhanced MRI is introduced without a standard definition, but typically less than 10 minutes, and after the localizer, it includes at least a pre- and one early-phase post-contrast T1-weighted series. And other techniques, such as ultra-fast, are also introduced and discussed. And I want to specifically mention this all-important early-phase post-contrast image, which is in the 60- to 120-second range. This was previously called initial, and the name has now changed to early-phase because it might not be the initial image, particularly in many current protocols. DWI is also discussed further as a non-contrast technique that currently augments contrast-enhanced imaging, but acknowledging that research is ongoing. No reporting guidelines were introduced into BI-RADS at this time, but the readers were referred to the International Breast DWI Working Group Statement. Okay, now we're really going to get into the lexicon and reporting system and the big news that the focus has been removed. And this is because most small dots of enhancement, which are foci, are benign, either by distribution, multiplicity, or features like this, where you have this tiny focus, which has T2 hyperintensity. And these can be dismissed. There's also inconsistent and heterogeneous use, both in research and clinical practice. There may be unique, suspicious findings that are less than 5 millimeters, but with modern MR techniques, these can either be described as a mass with appropriate mass features or focal non-mass enhancement, and that's what's recommended moving forward. Moving into mass, where there's some updates to the lexicon. Just like the other modalities, lobulated has been added as a shape, defined as one or more indentations resulting in an undulating contour. This is a nice example of a lobulated-shaped mass with circumscribed margins and homogeneous and also probably dark internal septations. And this was a fibroadenoma. Margin was also updated, so it still falls under circumscribed and non-circumscribed, but under non-circumscribed, instead of irregular and speculated, getting rid of irregular and going to indistinct and speculated. Indistinct is defined as uneven or jagged, as can be seen in this mass that I have the arrow pointing to. And this avoids the irregular-shaped mass with the irregular margin. It would be indistinct moving forward, and again, is consistent with the other modalities. This is a new feature that will be introduced for masses of T2 signal intensity. So benign and malignant masses can both be T2 hyperintense, but when there is T2 hyperintensity in combination with other specific features, it can be a sign of benignity. It's not required to be reported, but it should be reported when it influences assessments and recommendations. And then when it is reported, it will be reported as hyperintense or not hyperintense. And this should be defined as compared with a normal lymph node as the internal standard. So if the mass is uniformly as hyperintense as a lymph node, then it's T2 hyperintense. This is the same mass we were just looking at on the left-hand side of the screen on the post-contrast image. On the right-hand side, you have the T2 with fat suppression. And in the left breast, those are cis. So you can see how bright this mass is. This mass is T2 hyperintense. And again, all these features together go along with a fibroadenoma. Lymph nodes has already been discussed as being expanded. It's now a separate section by anatomic region and then defined as normal or abnormal. For MRI specifically, normal features include circumscribed nodes, reniform shape, homogeneous enhancement, increased T2 signal, and symmetry, both for ipsilateral and contralateral nodes. Lobulated nodes are fine if they have a thin cortex. And large dimensions can also be normal as long as the morphology is normal. For MRI, small lymph nodes, it may be difficult to see the fatty hilum, and that can still be normal. And kinetics are usually fast initial phase uptake and delayed phase washout. So what are you looking for for abnormal? Well, this is a subjective assessment. So you're looking for subjectively asymmetric or enlarged nodes, either compared to other ipsilateral or contralateral nodes, focal or diffuse cortical thickening, an interval change from priors, and then in a larger node where you would expect to see the fatty hilum effacement or absence of that hilum. As was mentioned, there's no size or cortical thickness for MRI to distinguish benign and malignant. The 3 millimeter cortex that's used for ultrasound is not validated for MRI. And MRI is a different situation because you have three orthogonal planes to look at, and you will have partial volume averaging. This is a normal intramammary node on a T2-weighted image. These are usually upper outer quadrant and follow the venous drainage pathway. For axillary nodes, we have the levels here. So again, pectoralis minor is the defining anatomic structure. Lateral and inferior is level 1. Posterior to pec minor and interpectoral nodes are level 2. And superior medial is level 3. And the recommendation is to report not only the level, but also the approximate number of abnormal nodes to assign to help with clinical decision making. This is a nice example of multiple abnormal levels for axillary lymph nodes. In blue, the blue arrows show multiple abnormal level 1 lymph nodes. Pec minor is outlined in yellow. And in the far right image, you have abnormal level 2 nodes posterior to the pec minor, and then level 3 media. For lymph node assessment, so again, if there's no known breast malignancy, an MRI is normal for the breast. So say you're in the screening setting, or you're doing staging, but you're looking at the contralateral breast and the known malignancy, and it doesn't have a suspicious finding. Then the abnormal lymph node is given a BI-RADS category, and you may include a description such as this. Abnormal level 1 right axillary lymph node, BI-RADS category 4b, moderate, suspicious for malignancy, recommend targeted ultrasound and ultrasound-guided tissue sampling. But specifically, in the known malignancy setting, so you're looking at the ipsilateral breast of a known malignancy in the staging setting, and there are abnormal lymph nodes on the MRI that are not known to be malignant, either intramammary or axillary, then no separate BI-RADS category. Instead, just describe the nodes and recommend management based on clinical decision making. So example, for abnormal level 1 axillary lymph nodes, if it would change clinical management, recommend targeted ultrasound and ultrasound-guided tissue sampling. And this allows for flexibility for the providers to decide which course of action is best. Internal mammary nodes on MRI are normal with current MRI techniques, so you will see them. They can be normal up to 9 to 10 millimeters in the literature. So what's abnormal? Well, again, it's a subjective assessment looking at asymmetry, and data for size threshold is very limited. In a small study of current cancer patients, greater than 5 millimeters was most predictive of involvement. And if you are going to describe them, then reporting the intercostal space is helpful. So this is a normal internal mammary lymph node. You see it sitting next to the vasculature. It's small. It has fat around it and a tiny fatty hilum. This is an abnormal internal mammary lymph node on the side of a previously treated breast cancer. It's enlarged and expanded. OK, we're going to wrap up with the reporting system, meaning the BI-RADS categories. So starting with BI-RADS category 3, compared to the previous edition, there is now more data, quite a bit more data for BI-RADS 3. But these studies are retrospective, single site with varying designs, clinical indications, and finding types. Data still remain limited compared with mammography and ultrasound. However, data do support that a malignancy rate less than 2% is achievable, with a caution in non-baseline exams. So in this particular study in 2020, the malignancy rate for BI-RADS 3 assessments on non-baseline exams was 9% versus 2% in baseline exams. Frequency of use is going to be updated to be less than 5% of cases compared to the 10% that's in BI-RADS edition 5. And it's important for all of us to audit our practices for all of our findings, but particularly for BI-RADS 3, because that informs our clinical practice and also contributes to the literature. So the best data at this point is for masses with an oval shape, circumscribed margin, and homogeneous internal enhancement or dark internal septations that are also T2 hyper-intense and not new or increasing. As for other findings, it should be infrequent based on individual and practice experience. For non-mass enhancement specifically, there has been no combination of published features that reliably get to a malignancy rate less than 2%. BI-RADS 3 should also not be used in the setting of known cancer staging, which is often while we're doing MRI, right? So in that setting, whatever the finding is needs a definitive diagnosis for clinical management. And treatment also affects the ability to assess findings in follow-up. So at that point, you would give it a BI-RADS category 4A and perform tissue sampling if it would change clinical management. So as a probably benign finding, so you have the early phase T1 post-contrast in the left and subtraction in the middle showing an oval-shaped mass with a circumscribed margin and homogeneous internal enhancement that's also T2 hyper-intense. So it's given a BI-RADS category 3 and was stable with two years of follow-up and then assessed as benign, probably a fibroadenoma. BI-RADS category 4 subdivisions have been added for MRI because there's data showing now that the MRI likelihood for malignancy for 4A, 4B, and 4C is within ranges for mammography and ultrasound. It's not required to be used, but can be really helpful for a more meaningful audit, assisting with RADPATH concordance and patient and provider expectations. And we're going to finish up with known malignancy. And these are findings that are the biopsy-proven breast cancer prior to surgery and also includes when there is no enhancement at the site, either before treatment or during or after neoadjuvant therapy. There's also new guidance for reporting additional ipsilateral findings to the known malignancy. So we're going to go over a few of those scenarios to follow up. So more extensive findings. So these are those that are the same morphology as the known malignancy. It's just bigger and contiguous. Thus they have a high certainty for malignancy, and this is a case where there were 28 millimeters of calcifications on mammography, biopsy-proven malignant. There is corresponding segmental heterogeneous non-mass enhancement. It's larger. It's 44 millimeters. This is BI-RADS category 6. Also BI-RADS category 6 are additional close findings, which is a new term that's going to be introduced. These are those close findings that are not thought to significantly change clinical management, not expected to, because they're less than 2 centimeters from the known cancer and do not increase the extent greater than 2 centimeters. Here's an example. You have a mass beneath it by 6 millimeters is an additional 8-millimeter mass. So it does not increase the extent greater than 2 centimeters, but you want to fully describe the size of these findings, relationship to the primary malignancy, and total extent. The use of the word satellite is recommended to be avoided due to heterogeneous meaning to different people. Those additional close findings are different than additional ipsilateral findings that may change clinical management or expected to potentially change clinical management because they're greater than 2 centimeters away and or increase the extent greater than 2 centimeters. So here's a case of a known malignant mass with contiguous but extensive non-mass enhancement measuring 55 millimeters, contiguous but a different morphology, and tissue diagnosis is recommended if it would change management. In other words, if the patient's not planning a mastectomy. Another similar case of a known malignancy of a rim-enhancing mass here. An additional rim-enhancing mass with a similar appearance, which is 2 centimeters away, more than 2 centimeters away, that would increase the extent greater than 2 centimeters and thus could significantly change clinical management. Recommendation is tissue diagnosis if it would change clinical management. And that's it. I'm going to hand it over to Dr. Peter Eby. So I am really proud to work with this stellar group of radiologists who has contributed extensive amount of volunteer time and effort and expertise to crafting the revision of the auditing and outcome monitoring section. And what you will notice first is that the previous version called it follow-up and outcome monitoring, so that when you specifically had questions about the audit and you went to look for the audit section, you could not find it. Well, we're going to make it easier for you by putting audit in the title. It is the same mission, though. We are aiming to provide a universal recipe for you to, for all practices and individuals to evaluate and adjust your own performance for the sake of your patients. So our updates are going to include, of course, new benchmarks for mammography, tomosynthesis, ultrasound, and MRI for everybody to strive for. We have a revised section of frequently asked questions. We have a revised guidance section. We have a new section called what not to audit, because there have often been questions about how to handle certain examinations. We are going to make it very clear for you. For example, BI-RAD 6 examinations are not audited. And as you have heard already, we are striving for modality neutrality, and these indications will apply regardless of whether or not the examination was acquired with mammo, ultrasound, or any modality that you might future imagine. We are also clarifying the difference between or what constitutes a screening versus a diagnostic examination. And this is important for the audit, because the screening audit should be based primarily or only on screening examinations. And we know that in the future, there may be new examinations that are used for screening. Probably the most likely in the near future is contrast-enhanced mammography, which is currently only approved for diagnostic use, but may be approved for screening use. In that case, we would still apply the exact same definitions to screening and diagnostic, and whatever else we may choose to incorporate into our breast radiology armamentarium. So to that end, screening is referred to as an examination, regardless of modality, applied to patients without any reported symptoms, clinically detected signs, or imaging findings. And a diagnostic examination is applied to patients to further characterize a symptom that they may report, something that their physician may find, or something that you have recalled from their screening examination that may or may not be malignant. And what that means is that modality-neutral assessments will be applied. A negative examination on screening is a BioRADS 1 or 2, meaning that you are not recommending any additional action at all until the next screening round. But if you give it a 0, 3, 4, or 5 at the time of screening, that is positive, because you are recommending additional imaging, or follow-up, or a procedure before the next screening. Diagnostic is slightly different. It is based on whether or not you recommend a tissue diagnosis. So a 1, 2, or 3 on a diagnostic examination is considered negative for the purposes of the audit. And if you recommend a tissue diagnosis, a 4 or 5 is considered positive for the purposes of the audit. In defining what constitutes a screening or diagnostic examination, this has become more important because we are seeing more and more varied practice of working up patients at the time of their screening, keeping them doing their diagnostic if they need to on the same day. This makes it a little bit more complicated for the audit, whether or not you choose to report all of those examinations together or separately to appropriately classify these examinations, you must separate them. This also applies to patients who might have two screening examinations on the same day, for example. So again, the sole binary decision of screening is whether or not to recommend additional action before the next round, and the recommendation for or the acquisition of additional images really creates a diagnostic examination. And so it is positive for the purposes of the audit, again, if you recall them or acquire additional images. And I'm going to give you some examples. And that diagnostic examination is created when additional evaluation is performed regardless of modality. So this is just an algorithm for how screening might ultimately be coded. And you can imagine that if a patient has arrived for a screening mammogram, and she stays for her results, and the radiologist decides that they want two additional spot views for a potential asymmetry, the patient has then had a screening and a diagnostic examination for the purposes of the audit. Now, you might issue a single report that summarizes this, but for the purposes of your record keeping and to understand your own performance, you need to classify these separately. As a screening mammogram, zero, because acquisition is recommended, and then however you ultimately handled the spot. Magnification views, if your asymmetry disappears and that ends up being a negative, again, that's a screening plus a diagnostic. So this is a screening ABUS. If the patient is recalled for additional evaluation, and you might, if they wait and you do it on the same day or a different day and do a separate handheld directed ultrasound, that is a screening plus a diagnostic. Exactly the same as the example I just gave you with mammography. With a handheld ultrasound, I'm going to give you two examples for clarity. This is a handheld ultrasound, and the patient does the standard protocol according to ACR parameters. Also acquires a couple of images of classically benign simple cysts. The radiologist sees this, calls it a benign. That's just a screening. It's just a benign screening, period. But if the technologist takes some pictures of some potential shadowing and the radiologist evaluates additionally, acquires more images to show that it is either suspicious or non-suspicious, then a diagnostic examination has been created by that additional evaluation above and beyond the standard screening, okay, for clarity. Moving on to the next topic, there are some high-risk histology updates. So this is the typical or the prior list of high-risk histologies. Atopic LCIS and florid LCIS are now being moved from high-risk, which is a benign category, to malignant. Peripheral duct papilloma is going to be considered benign, just removed from the high-risk list as is radial scar, based on all of the evolving data regarding very, very low risks of upgrade and no additional contribution to or very little contribution to additional lifetime risk. We are adding a BI-RADS 3 basic audit to the standard audit. The purpose is because we have very little data, even after a few decades of Dr. Sickle's initial robust and well-designed and well-executed study that established our less than 2% threshold for the three mammography findings of clustered punctate and round calcifications, asymmetry, and non-calcified oval mass. We have almost nothing in the literature to provide benchmarks for how often this is used and should ultimately prove to be positive. So the audit for BI-RADS 3 is created to help us gather that data, but also to help everybody who uses the ATLAS and uses the audit to see how they are doing in their own practice. It is really time for us to have a closer look at how we are adapting this not only to mammography in the digital and tomosynthesis age, but also in ultrasound and MRI. This is an example. We are providing flowcharts in the section of how to do this with extensive description. We are also providing a diagnostic audit flowchart. These flowcharts did not exist in the prior edition. This is the BI-RADS 3 audit flowchart example. You don't need to read it now, but this will be in beautiful large print detail when you get your audit and sit down in front of the fire to read. Tough crowd. We are also adding an optional audit for the MRI extensive disease. This is an unusual scenario, but it is for the same reason as we are adding the BI-RADS 3 audit, is that we have little consistent benchmark data about the PPV and the recall rate for MRI for extensive disease, despite the fact that it represents, along with screening, a very large segment of our indications for doing these examinations. The challenge is that these patients already come with a BI-RADS 6 assessment because they have cancer. And I told you at the beginning of my talk that you do not audit BI-RADS 6. You do not audit those cases. So the difference is that all of those traditional audits are done at the level of the patient. For this, we have to perform the audit at the level of the finding. This is optional, but it provides a consistent framework for facilities that want to audit this and want to be able to contribute to our national understanding of our performance or our international understanding of our performance, as it were, so that we have benchmarks to compare with each other and to see how we are using it. Is our recall rate very high? Is our PPV very low? How often are these things that we find at the time of extensive disease actually cancer in the ipsilateral breast or the contralateral breast? So the true positive, false positive, and false negatives will be defined by a tissue diagnosis. And we will have the same PPV categories. But again, this will be optional. And we are hoping that some facilities and some people will participate so that we can understand our global performance. Finally, we are adding a new section on the initial method of detection. This is defined as the single first test or sign or symptom that triggered the workup that led to a biopsy recommendation. There are standardized categories that have been created and adopted by the National Mammography Database, which is also maintained by the ACR. These have also been now shared with the North American Association of Sanctuary Cancer Registries that are adopting them. There is going to be a rationale and guidance chapter in this atlas. And currently, it is optional. But we are expecting it to become mandatory as it will be incorporated into national registries. The ask is for radiologists to put a single sentence at the end of their reports when they are recommending a biopsy. The method of detection is blank. Could be screening tomo. Could be screening mammography. Could be screening ultrasound. It could be patient detected or provider detected. This is the list, the complete list of options. The majority of them are screening. The two most commonly, these are either going to be a screening modality or a patient or provider detected. There are some rare scenarios in which a patient may come to our attention some other way. For example, has totally asymptomatic and has a screening abdominal ultrasound and has a liver mass that turns out to be a breast cancer metastasis. That would fit into category and is expected to be few and far between. I have a couple of examples for you. This is a patient who had a 2D screening, no tomo. You can see that she's got calcifications in the left breast, posterior depth recalled. Biopsy was recommended. The MOD here is SMA, okay? Screening mammography. Here's another one. This patient has a BB right here on the right side. You don't see. Oh, there we go. Now I see. Yes. So she came for a palpable lump. She's extremely dense. As you can tell, it's very difficult to see anything in this mammogram. She has an ultrasound and she has a, let me get this right, Jessica. She has a mixed solid and cystic mass. Thank you. Biopsy was recommended and the MOD is patient, okay? I'm going to give you one more. This one, patient presented with a lump on the left side. This is the triangle, the skin marker indicating the area of clinical concern. The left side is negative. The right side has an irregular high-density mass with indistinct and speculated margins. This is also a PAT. This patient was not asymptomatic. This was not a screening examination. Despite the fact that her cancer was in the opposite breast, what triggered this examination was her palpating something, okay? We do not want to give screening credit for this case. Thank you. So auditing really improves our individual performance and patient care at a facility and individual and national international level. Looking at our own numbers helps us be better. It also helps us collect data to see how we were performing and if we can establish additional benchmarks for, for example, BIORADS 3. Thank you so much.

BI-RADS Atlas

breast imaging

mammography

ultrasound

MRI

mass descriptors

breast density

lymph node assessment

modality-neutral

structured reporting