With CHD often discovered in low-risk pregnancies and associated with neurodevelopmental deficits, the need for more screening and advanced techniques is clear. In this two-part video, specialists from the UCSF Fetal Treatment Center show the potential for improved outcomes with early detection, provide indications for screening, explain how CHD impacts brain development, and explore in utero therapeutic options.
Refer to Fetal Treatment Center we're going to speak today about innovations in care delivery for a funeral congenital heart disease and just as a forward with some background, um kind of the y so can we actually make a difference in survival? In short term morbidity, long term morbidity mortality for these Children? And can we make a global difference in costs and disease prevalence and other are there other aspects in which we can make a difference to put congenital heart disease in perspective? Congenital heart disease, meeting intervention by 12 months of age is estimated at three per 1000 live births. So quite quite common. And in fact, congenital heart disease is the most common cause of infant death due to birth defects in the United States And the CDC 10 years ago now estimated that health care costs for treatment of these conditions was over $1.9 billion. Um So quite quite a feat to try to improve for these Children. And what about the rest of the world? This is a nice paper, seminal paper really from one of my colleagues who recently passed away Julian Hoffman, looking at the incidence of congenital heart disease and multiple populations and different socioeconomic groups. And the bottom line is there was no evidence for differences in that incidents three per 1000 live births in different countries across the globe. So what is the potential impact of prenatal diagnosis of congenital heart disease? Well, of course, informed choices for the pregnant patient preparation of the family and the health care team. The potential for reductions in costs uh in transport of sick infants and so forth, and then improvements in perinatal outcome and dr Bandy will we'll talk more about this critical neonatal heart disease and cardiovascular pathologies associated with human dynamic compromise before birth. And finally, from a research perspective, prenatal diagnosis gives us the opportunity to have some insight into the spectrum of disease and the potential for intervening in Euro, as well as for altering the spectrum of more severe disease at birth. I'm not going to talk about fetal intervention as an innovation that's sort of a different group of people, but just to to mention what we're able to do now um with relatively good success is balloon dilate the aortic valve in aortic stenosis and evolving hypoplastic left heart syndrome through an entirely per cutaneous approach fatal arrhythmias are something that I spent quite a bit of time thinking about both tacky arrhythmias and brady arrhythmias and their potential treatment. So, prenatal diagnosis of these gives us the opportunity to treat and dramatically improve outcomes for these patients, including for super ventricular tachycardia, where maternal transplant essential therapy with anti arrhythmic has brought 50% perinatal mortality down to less than 10%. And we UCSF are currently enrolling in three related randomized controlled trials of Jackson Black and I'd and soda lol in the international fast therapy trial. So if prenatal diagnosis gives us so many opportunities and we've had ultrasound now for decades, how are we doing with prenatal diagnosis of major congenital heart disease? Well, this is a 22 companion papers that were published in 2009, looking at the California population, in northern California, the major surgical centers and in southern California as well, asked all of the parents of Children coming in for intervention, either intervention or surgery At less than 12 months of life. One simple question, did you know that your baby had this heart problem before the baby was born? And the answer was yes, in only about 30%. This question was asked again on a national level, this time using the society through essex surgeons database, looking at prenatal detection almost a decade later And sadly, California was still the same. So we haven't made huge strides in prenatal diagnosis over the last 10 years or so Now, why would that be? Um both of those studies were in the era before the outflow tracts recommendations of 2013 were made. And so most of the mistakes, as far as low prenatal diagnosis were in patients with tetralogy of fellow T. G. A. And term guests as well as double of the right ventricle. All of these have a normal four chamber. Um whereas the ones with abnormal four chamber view were more likely to be detected but still covering around 50 even a single ventricle patients being detected prenatally. So the question is can we do better? And the answer is well we certainly should be able to. An abnormal four chamber view should be present in 60 of congenital heart disease. So it should be easily identifiable. And if we add outflow track views up to 40 additional patients should be detectable. So that's sort of my uh my introduction to why should we care? And you know, we're really not as good as we think we are. Um but I want to switch gears now and talk about the topic of the day, which is what are we doing to to innovate? And the first thing I I wanted to highlight was the concept of an early fetal echocardiogram. So what is early? I don't mean five weeks. The hearts like a beating tube at this point. But by 10 weeks um LMP the heart is completely formed into a miniature version of what we would consider a normal heart. So by about this time, by uh sorry, that's 10 weeks post conception. Um we should be able to identify normal and abnormal structures And the heart is essentially just smaller than it is at 17 or 18 weeks in the traditional screening setting. How has this evolved? Uh early foetal echo idea. Well, the early obese can at 11 to 13 weeks for confirmation of interview. Current location, number of fetuses and gestational age assessment as well as detection of an employee's has been around for quite a while and really exploded with The institution of Universal Nukul translucency screening between 11 and 14 weeks. Although this was for risk stratification for an employee, a very soon became evident that we could also pick up some congenital heart disease using local transistor NC screening. At the same time, cardiac screening um started to be practical with uh improvements in imaging technology. So in the early 80s and into the 1990s, trans vaginal approach was really necessary, so there wasn't a huge amount of uptake. But as the trans abdominal scanners started to get better uh in the early 1990s we were able to demonstrate normal cardiac abnormality and some cardiac uh normal cardiac anatomy and some abnormal anatomy with both trans abdominal and trans vaginal approaches and since that time with really increased experience, expertise and technology, as well as what I mentioned, the nuclear translucency. Earlier screening, several case series have been published and correlations between early foetal echo and later foetal echo. Um We're done here at UCSF showing that this really is feasible. So, our current indications for Uh for Fetal Echo in the 18-22 week group can be sort of boiled down to a suspected cardiac abnormality on obstetric ultrasound, maternal risk factors, fetal risk factors, familial risk factors and so called family history and other increased risk, including those knuckle translucency patients. Those are usually located uh indications we see those all the time. What about early? Well they're going to be similar but a little bit different in order to maximize the impact of this and keep our diagnostic percentage is where we want them. So for early foetal echo, we like to see patients with significantly increased anti uh fetuses already identified to have extra cardiac abnormalities or chromosome abnormalities and families with a history of a previous child with serious congenital heart disease, essentially Uh similar indications. But we really have moved the risk needle from about 1.5 to 3 up to 10 so that we're screening fewer people and have a better chance of of having true positives and and very few false negatives. I think I'll skip the new cool translucency idea for for the moment. I think this most of the audience is going to be Uh familiar with that. And just point out that the American Heart Association scientific statement that came out in 2014 already was talking about using some of these fetal factors and non cardiac abnormalities or chromosome abnormality To say that they should really have an echo at detection. And so if that abnormality is detected at 12-14 weeks, then an early foetal echo is recommended by this group. What about trans vaginal versus trans abdominal? I think I used to have one of these in my clinic and it it struck fear into the hearts of my Of my pediatric cardiology colleagues, both male and female and uh thank goodness. Uh we don't really use these anymore. The trans vaginal has been shown to be superior up till about 12-13 weeks. But at 14 weeks in both this kinberg study and in our own experience, the trans vaginal and trans abdominal imaging is comparable and trans abdominal is usually sufficient. If not, we tend to, rather than going to trans vaginal just bring them back in a week and can get better images. We have some equipment tricks and foetal echo that I'm not going to talk about. But I do want to show you uh some pictures to try to convince you that this particular innovation is actually ready for prime time. So here's a 12 weeks venous an axial four chamber of the heart. I don't think anybody would argue that this is not completely normal. Different ultrasound platform. Same basic idea. The four chamber view allows us to evaluate site is position access. And this looks essentially the same as the 18 to 22 week Screening ultrasound of the axial four chamber of the fetal heart. We can also look at outflow tracts. Three vessel view. Three vessel tricky of you. This is at 14 weeks on yet another ultrasound platforms. Um and this is a twin with an anterior placenta. Uh So really with high frequency transducers or even with linear transducers, we can get some phenomenal images even on these very small structures, including a mitral valve leaflet in those short axis at 13 weeks. Then we can see the normal anatomy And addition of color Doppler helps us in difficult situations. Here's 13 weeks a fetus looking at a normal ductile arch, normal aortic arch using color Doppler and Color Doppler is also helpful when the two D. Structures are not as well seen to show that both of the valves here in the four chamber are patent and that the arches are also patent with anti grade flow. So this is a normal screen and exactly what you would do As part of a detailed and Tomic scan or 768-11 At 18 weeks. So this screening views can be done. We're actually doing full echocardiograms. So we do use Doppler. It's indicated in these patients because they all have a suspected cardiac abnormality. Power Doppler these lower velocity uh structure is easier to fill a little bit with directional power Dubler. So if your ultrasound platform can do those, we recommend it. And here's an abnormality seen at 13. I think this is 13.5 weeks very clearly with color Doppler vascular ring U shaped structure instead of the V. Shaped structure in the three vessel. And tricky of you. We can also do detailed function and rhythm assessments. So here we're using pulse wave or spectral Doppler sparingly. Of course, as I said, um ultrasound safety is key. But these are indicated in these patients that have a suspected abnormality. Um but sometimes we don't even need that. This is a 12 week fetus with obviously a single ventricle and color Doppler showing a single outflow in the setting of a right sided stomach, left sided apex. This patient was correctly diagnosed with a severe single ventricle hetero taxi and was able to interrupt the pregnancy soon thereafter. As I said, we published our work here at UCSF and this now is actually almost 10 years ago. But we were already at 85 full visualization of all of the structures that we would see on an echo, not just the screening views but an echo at 14 weeks. So that seems to be kind of the sweet spot. We have all of these patients come back for the 18-22 weeks. Can because some things can be progressive and we can miss small holes. But for the most part we're able to offer this now to all patients. And so um uh to all patients at high risk. So what else could we potentially learn from this early foetal echo? I think these are mostly in the research realm but very promising and allowing for additional therapies. The second innovation that I'm really very excited about uh is going to be on the next just couple of slides before I before I turn it over to Dr P. Vondie. And that is a problem that we identified a few years ago that I showed in the first few slides. And that is we can do all of the first trimester scanning and great second trimester diagnosis. We can plan deliveries, we can do fatal intervention but the patients have to get to the cardiologist first. And as I shown, the rate of Mrs exceeds the rate of uh correctly picked up babies when they're screened at the community level. These are mostly low risk pregnancies that congenital heart disease is happening in and so we need to do something better with screening. And so I've been working with a group here at UCSF fund Development and application of using artificial intelligence or deep learning for disease diagnosis in the fetal heart, hoping to improve congenital heart disease pickups with deep learning. So we have we've now published uh a few abstracts that coming out very soon. Full paper on this and I'll show you some of the early data we initially studied 18 to 24 weeks gestation singleton's with images from multiple ultrasound systems all at U. C. S. F. Or in boston. And um of the roughly 29,000 images, 493 patients were normal 87 tetralogy of fellow and 100 and five had hypoplastic left heart syndrome. We excluded extra cardiac anomalies and all of the studies were validated by expert over readers and with a postnatal diagnosis, confirmation and step one. We used to supervise learning approach to train a convolutional neural network classifier to find five axial cardiac views. And then in step two for each view, we used the same convolutional neural network to classify whether the view was normal or abnormal. And step three we created a composite diagnostic score from each classifier in step two to calculate the sensitivity and specificity for congenital heart disease. And um there's way to plug the Euclidean distance of this in space so that it can tell the difference between H. L. H. S. And tetralogy of fellow. So this is the the kind of flow diagram. We have the whole ultrasound unlabeled. It's put into this diagnostic classifier. The classifier spits out the five different views and then the classifier looks at each view and says, is that a normal three that's vessel trachea or abnormal? And with that yes and no pattern will also classify the disease. Um So this part of it is as far as we want to get with screening for sure. It can also do biometric calculation and segmentation. So cardiac access and size. So these are our results by step. Our view classifier initially was greater than 90 in its uh detection, correct detection of the five views with a very very high um sorry, very low false positive rate. The AU to see if, if you can't see it if you're on your phone you may not see it but these curves are actually so close to one. Um That the test was sort of nearly perfect In step two where we looked by view um the way you see looks a little bit worse. But if you look at the legend here for for instance uh these are abnormals tetralogy or hypoplastic left heart syndrome. It was a coin flip. Using the pink. But the pinkish abdomen you wouldn't expect the abdomen to be abnormal in H. L. H. S. Or tetralogy. So I think that that sort of shows that it's working um that it wasn't making an assignment based on something that shouldn't be shouldn't be diagnostic. And so we were doing pretty well with this. Our model sensitivity and specificity Uh is shown in the first two columns for just abnormal. Uh almost 90%. And for tetralogy and hypoplastic left heart syndrome. Quite good, especially when compared to the reported sensitivity from those earlier papers that I showed you. And then finally the biometrics uh here the machine did quite well where technology of below expected to have an abnormal axis when the machine measures actually did have abnormal access. So we've concluded by uh using this deep learning and convolutional neural networks with supervised learning approach and models to apply to clinically perform secondary MR ultrasounds that we can get pretty good improvement in detection of congenital heart disease at least. Um Conceptually where in the process of applying this to a larger real world screening data set with a known true and false positives and negatives. And that'll allow comparison of the model to the current standard of care communities screening. Once these data are available, we'll be able to do cost analyses and at this point I just have to say scalability is the main strength of this work in prenatal diagnosis as uh A. I is being used more and more in the medical field. Uh This is a big bonus that we could potentially scale this to the entire population as opposed to teaching and training more humans to do it, combining clinical insights with ensemble learning methods, can facilitate data efficient strategies and target clinically relevant use cases. And we're currently applying this methodology to other use cases. And with that I'll turn it over to dr sha'ab Bondi, my friend and colleague to talk about the heart brain connection in fetal congenital heart disease. Thank you so much. Doctor Moon Grady. Um That's such exciting work that's being done and such an important step in increasing prenatal detection of congenital heart disease. Um we're going to shift in the next 20 minutes and talk about once we do make a prenatal diagnosis of congenital heart disease. Um Our approach here at UCSF, which is really taking a holistic approach to not only trying to understand the fetal heart, but also how it might be impacting other developing organs in the fetus, specifically the brain. Um And that is because we know that human cardiac development is largely completed by the end of the first trimester, whereas human brain development occurs over a much longer period of time, with morphological events occurring in the first two trimesters and then a refinement of connections and myelin nation occurring in the third trimester and after birth. And that third trimester of fetal life is marked by a period of accelerated brain growth, increased metabolism and increased demand for oxygen And the brain and the heart responds by increasing blood flow to the brain. Such that approximately 25 of the cardiac output is directed towards the brain. So there is clearly an important relationship between the heart and the brain that begins in Utero and could be altered in the setting of structural congenital heart disease. We also know from literature from several investigators around the world that neuro development is impaired when a child has congenital heart disease. Although there is a spectrum, this is just looking at one domain IQ cognition. And we can see that compared to the general population. On average, Children with congenital heart disease have slightly lower IQ. And in the setting of a genetic abnormality can be much lower than that, but this is really just one domain. Um, and as we've learned more over the years, Children with congenital heart disease can have uh impact many domains of neuro development, sometimes subtle yet pervasive, um and can have a real impact on the child's quality of life and ability to gain independence. We also know from work that was initially done here at UCSF by Stephen Miller when he was here at the time, along with Patrick McQuillan, that newborns with congenital heart disease, even though they are born at full term for the most part have evidence of delayed brain development. And this is looking at a particular measure of brain development. N acetyl asper Tate to Colin ratios, which is an important metabolite in the brain that increases with increasing maturity. And in their study, what they found was that in a full term baby with congenital heart disease at 40 weeks. This is those with complex heart disease such as transposition or single ventricle lesions. Levels of metabolites in the brain were similar to levels typically seen 34 weeks in premature babies. So although these babies are being born at full term, their brains are approximately 4 to 6 weeks delayed second. We've also from the MRI literature also know that brain injury is common in these newborns with complex congenital heart disease. This is also work from U. C. S. F. Where they made the observation that about a third of babies with either transposition or single ventricle heart disease have evidence of pre operative brain injury even before they go to the operating room and that's usually in the form of white matter injury. A scene in panel A here or small focal stroke, A scene in panels, B and C. Here. And the reason why this is important is that those studies that observed these brain injuries were done as part of a research study and there were no overt neurologic findings that would prompt the provider to obtain an MRI. So they were usually termed clinically silent brain injuries. But in this study we found that although they were clinically silent, they were clinically relevant. Those patients that exhibited white matter injury on the moderate to severe end of the spectrum had worse motor outcomes at 2.5 years of age. So clearly there with that background information, there is a really important relationship between the heart and the brain that can be altered in the setting of congenital heart disease. And given that a lot of the findings we are seeing are in the newborn period, this relationship appears to begin in utero. And so our program here is trying to use advanced fetal imaging to help us understand this heart brain relationship in utero, to understand fetal brain development, physiologic mechanisms that can explain our observations and I think importantly also whether we are able to predict neurodevelopmental outcome prenatally in patients with complex congenital heart disease. Um certainly we can't forget about what we've learned about the brain using ultrasound technology um and fetal echocardiography. And there have been observations that brain blood flow patterns can change in the setting of global hypoxia, specifically in fetuses that are growth restricted with chronic hypoxia. There observed to have this so called brain sparing pattern in the middle cerebral artery in the brain, which essentially is a lower impedance or resistance within that vessel to allow for more blood flow to the brain. And it's thought to be an auto regulatory mechanism in response to this chronic hypoxic state. Um And this same pattern has been observed in fetuses with congenital heart disease as well, specifically those with very severe disease such as hypoplastic left heart syndrome. But where I think we've learned the most about the fetal brain is from fetal brain MRI. And this is work that was originally described by Kathy Lymberopoulos and she was at boston and is now at Children's National, where she observed a fetal onset of delayed brain development. Um and what we're seeing here is total brain volume was observed to be significantly lower in fetuses with congenital heart disease is seen by this red line here compared to controls. And what's interesting is that the divergence between these two groups begins in the third trimester of fetal life, which is really in line with what we understand about normal brain development in utero. There's also since those original observations, there are now other groups who are trying to understand what is the physiologic mechanism of delayed brain development in utero. Because we know that these different forms of congenital heart disease are very different physiologically, but all appear to be impacted with regard to brain development. And this work has largely been uh pioneered by the group in Toronto from mike seat and his group where they used advanced imaging in the form of fetal cardiovascular MRI magnetic resonance oximeter tree to quantify oxygen delivery to various in various regions of the fetal heart, including oxygen delivery to the brain. So they also looked at oxygen delivery to the fetus as a whole. And they found that in the fetus with congenital heart disease there is lower umbilical venous flow and lower umbilical venous oxygen saturation um to the degree of a 17 reduction in oxygen delivery and oxygen consumption compared to a fetus without congenital heart disease. They also looked at specifically oxygen delivery to the brain and found that fetuses with complex congenital heart disease. This is a mixed bag of of different forms of congenital heart disease have significantly lower cerebral oxygen delivery and oxygen consumption compared to controls, and that there is a direct correlation between brain size and cerebral oxygen consumption. So these findings really led to the resurgence of the conversation of whether providing the fetus with supplemental oxygen could increase oxygen delivery to the brain and thus optimize brain development in utero. And this is not a new concept, maternal hypoxia, as a way of providing supplemental oxygen to the fetus was first described in the 1980s in the setting of the growth restricted Fetus, where there was there was an observation that if you give mothers oxygen, um it does indeed cross the placenta with acute increases in cord blood oxygen saturation in the umbilical vein and the umbilical artery. And this was followed by noninvasive data using fetal echocardiography that also showed with the administration of maternal hypoxia in the third trimester. In particular, you see distinct physiologic changes in the fetus with a decline in pulmonary vascular impedance and an increase in pulmonary blood flow. Importantly, the physiologic changes are only seen in late gestation and they tend to normalize after cessation of maternal hypoxia. So what we've started here at U. C. S. F. A few years ago with many of my colleagues is the fetal heart and brain study. And the overall goal of this research program is to understand the causes of delayed brain development and congenital heart disease prenatally to identify fetal predictors of neonatal brain injury and neurodevelopmental outcome and importantly what's to identify possible neuro protective agents in fetal life. one of the studies that we recently published from this research program was our attempt to identify an imaging tool that can efficiently give us quantitative information about cerebral oxygenation um that can also allow us to test possible neuro protective agents in utero. We used a tool called T. Two Star MRI, which really was just apply to the fetus in this situation. It's widely used in the pediatric and adult population as an MRI sequence. Um And in this case, along with collaborators and biomedical engineering, we were able to apply this particular sequence to the to the fetus, and the basic idea is that Teachers Star MRI measures relative levels of dioxin hemoglobin in the tissue of interest, which in this case is the brain and that the dioxin hemoglobin content is inversely related to T. Two stars signal decay. So the shorter the decay time, if you see this blue line here denotes more dioxide hemoglobin or less oxygenated tissue. So this is um was not possible without collaboration with my colleagues here in um in radiology and biomedical engineering, namely Dwan shoe and jing liu. Um where they were able to um apply semiautomatic whole fetal brain segmentation to sort of get around the limitations of fetal brain imaging, which are limited by fetal movement as well as movement from maternal breathing. And this just allowed for a three D. Segmentation of the fetal brain. That then allowed us to measure the T. Two star value at multiple time points in this, in this case being eight distinct time periods and then fit the T. Two star decay time based on the average signal intensities within the segmented brain at these eight individual time points. Uh and what we ultimately get is what we want to know is how long does it take for this T. Two star signal to decay. And so first of all, what our data demonstrated was that similar to what's been reported in the literature fetuses with congenital heart disease in the third trimester. These are those with neonatal congenital heart disease that would require an operation in the newborn period. Have significantly smaller brains compared to control fetuses. Mhm. Um And secondly when we looked at this T. Two star value as our measurement of cerebral tissue oxygenation, we found that fetuses with congenital heart disease have significantly lower cerebral tissue oxygenation compared to controls. And we were able to look at individual subgroups of congenital heart disease. Um small numbers but allowed us to just describe our findings um including fetuses with D. Transposition of the great arteries and left sided obstructive lesions. Um And we found that regardless of the underlying physiology um T. Two star values or cerebral oxygenation was similarly low. Whether the fetus that hypoplastic left heart syndrome or D transposition of the great arteries compared to controls as an exploratory aim. We also look to see what would happen to this T. Two star measurement if we were to give mothers oxygen while they were in the MRI scanner. So what this figure is showing us is the T. Two star value at baseline. Um And then we provided the mothers with oxygen through a non rebreather face mask. Yeah. And then repeated the measurement at approximately 10 minutes. And what we found is in the control group, as seen by this thick black line here on average, there was really minimal change with maternal hypoxia in that group, which really made sense to us because it is likely that in the normal situation the fetus is able to auto regulate cerebral oxygenation and blood flow to the brain. We found similarly in the left side obstructive lesion group that there was also minimal change with maternal hypoxia. However, in the group with transposition of the great arteries, there appeared to be a significant increase in T. two star from baseline to after 10 minutes of supplemental oxygen. So these were really just preliminary exploratory findings that allowed us to identify T two star MRI as a efficient imaging tool that could give us a lot of information about cerebral vascular physiology. But it also gave us information that clearly underlying cardiovascular physiology is important in response to neuro protective agents, including maternal hypoxia and must be taken into account in future neuro protective trials. So um for our future directions for this particular study, um we are hoping to use T. Two star MRI um to continue to give us information about the acute maternal high proxy a response, but also allow for longitudinal safety monitoring given its efficiency and secondly, use this particular imaging tool um to look at the association between fetal cerebral tissue oxygenation and post natal neurologic outcomes, including risk of brain injury and neurodevelopmental outcome. We're also through a fetal heart and brain study group. We are very interested in investigating other factors that can influence brain health and feel congenital heart disease, not only including cardiovascular physiology but also genetic abnormalities that could have led to both the heart defect but also brain abnormalities that may explain ultimate neurodevelopmental outcomes. And also very importantly understanding the maternal fetal environment and what role does psychological stress and environmental exposures play on the developing fetus? I'm going to end with this slide, which I get this question a lot. These imaging tools are really exciting and innovative but can this information help us predict outcome or more specifically can prenatal imaging of the brain help us predict neurodevelopmental outcome so that we can better counsel families about what to expect. And this is our first attempt at trying to answer that question. This is a multi center study we did with our colleagues in Toronto, which was a longitudinal brain MRI study um including patients with hypoplastic left heart syndrome and D transposition of the great arteries. And they had brain mris performed first in fetal life in the third trimester, followed by a second MRI in the neonatal time period. But before they went to the operating room for their neonatal operation. And what we found as you can see in this figure here, this is looking at total brain volume as our measure of brain maturity and specifically in those patients with transposition of the great arteries. We found that those patients that went on to develop significant white matter injury in that newborn period, as seen by this orange line here had smaller brains beginning in utero. So perhaps fetal brain MRI with quantification of global and regional brain volumes may be a useful biomarker to help us risk stratify fetuses at increased risk for brain injury and ultimately worse neurodevelopmental outcome. So with that I want to thank our multidisciplinary fetal heart and brain research group here at UCSF. This is really one of the advantages of being at this institution is we have a lot of great people across departments uh that we sort of all came together to answer these questions, including individuals in um radiology such as Rory Glen and Jim Markovich in biomedical engineering, uh neonatal neurology as well as my colleagues and neuropsychology and our wonderful nurses in the fetal treatment center. Um and we and these are our funding sources. Um So I wanted to end by showing everybody are wonderful fetal cardiovascular team in the fetal treatment center. Everything is pretty much done by zoom these days. And in addition to the wonderful physicians we have wonderful nurse Kristen Ghazni L. And nurse practitioner and a tran that really run the program along with Anita, who is the director of our director of our program. This is our whole team, including the some of the providers who are not on that zoom call. Um And um this these are the fetal treatment center physician liaisons. This uh lecture is recorded as our previous FTC webinars. So you can go to this website if you need to access any of the past webinars. Um and then if you do need to contact us and refer a patient to UCSF, here is the contact information and how to reach us. Uh If you if you have any questions, thank you very much.