Thank you so much for this wonderful introduction. I'm very happy to be here to talk about something very near and dear to my heart. So, let's see. All right, so my part of this is basically it's the, ultrasound quality assurance is the holy grail, really. We need to achieve program perfection. So objectives, I'll be talking about the idea standards of applying ultrasound and I'll link it to the I-AIM framework, which I'll explain what it is. And then we'll talk about image archiving and workflow and then quality assurance process and touch a little bit about billing. So we all know that ultrasound is here to stay and ultrasound will replace sound. This is a fact. It's becoming very small, portable. It can really replace the ultrasound. And the public knows about this. So this just published in New Yorker a few days ago. So it's showing that the, could ultrasound replace stethoscope? So you're expected at one point that you would perform an ultrasound on your patient by the family. But you know, training for ultrasound, there are varied courses. It's a lot of times if you can't hold a probe, you can do the ultrasound and there are too many ways to train physicians and nurse practitioners and everybody in doing ultrasound and there's a lot of variation. There is the Society of Critical Care and Medicine course, for instance. It's two days, very extensive program. And there are other courses that are available. There are three hours to get a taste of what's ultrasound. So there's a lot of variation. But I'd like to introduce the I-AIM framework for focus, which is the I-AIM is an acronym for Indication, Acquisition, Interpretation and Medical Decision Making. It's a framework that's published to help in education and to serve as a clinical model for focus. So for us, for in the critical care, the scope of practices application is basically, you need to answer a question. You cannot just apply ultrasound without having a question in mind. So you should know that what is the question? Is my patient hypotensive? Why is my patient hypotensive? Is he dry or not? Is he hypovolemic or not? During code situation, for instance. And then the second thing is, how can I get the best information? What cardiac and non-cardiac views that I need to get? To interpret the images, you need to look at the image quality first, and then identify the structures, and then you go with pattern recognition after that. And then for medical decision making, you wanna know whether what you found answers the question that you have initially, and then what am I going to do about it? So there's management application. For this, the problem you need to know a lot to apply what you found. It's you're reducing 22 or more images of echocardiography to one or two windows, and then you need to apply what you recruit from this to the clinical situation. So an example, if your patient's blood pressure low, then we know that blood pressure is cardiac output times SVR. Cardiac output's heart rate times stroke volume. So with ultrasound, we can really quickly rule out severe hypovolemia, severe pump failure. We look for patterns for obstructive physiology, and then if everything else doesn't show, then we can blame it on SVR. But the problem is there is something called underlying chronic disease. So for instance, what if your patient is hypovolemic and started with dilated cardiomyopathy? In this situation, you can see this patient has a pump failure. This is an apical four-chamber view. This is the left ventricle. The contractility here is really severe, severely depressed. So the initial diagnosis was cardiogenic shock, and the treatment was minimize fluids and start epinephrine. And then the clinical pictures worsen, and then someone looked at the chart and found that the ejection fraction base was less than 30%, and patient evolved sepsis, and how given fluid now improved. There's also the possibility of having a little bit of subtle finding in your ultrasound. So for instance, in this patient with aspiration pneumonia, ARDS, presented for X-Lab, history of COPD and sleep apnea, which raised the concern for right ventricular dysfunction. But from this subcostal view, for instance, first, it's a subcostal because I can see a liver here. This is the falciporum ligament, so there is some, not effusion, there's some asides. The right heart here may be equal size of left heart. I don't know. The right atrium may be dilated. You cannot really infer a lot of basically finding from this view. You need more assessment, because RV dysfunction, for instance, the point-of-care ultrasound will help only during extremes, when the patient is really in very bad hemodynamic disturbance situation, or if it's new. So this was passed as no issues with the right heart, and the IVC was flat, so patient was hypovolemic, giving a liter of fluid very quickly, induced with anesthesia, and the result is severe hypotension. So it turned out to have acute and chronic corpulmonality. You can see here the right atrium is really dilated, as a sign of chronic disease. The RV is thick. There's an obstructive shock. The IVC became very plethoric, and then finally, here you go, he arrested. So the problem here is you shouldn't give fluid very quickly to someone with RV dysfunction. So when you have issues with underlying chronic disease, if there's mixed etiology of shock, rare kind of shock, unexplainable findings, or if there's lack of agreement what you found with the ultrasound, you need to call for help. And the problem is a lot of times is you don't know what you don't know. So for this reason, it was the J-group named this as the second health technology hazard in 2020. And this is because they felt that the safeguards for using these ultrasounds really lingered behind the speed of adoption. And there is lack of sufficient oversight, obviously, for use or misuse. So I'm gonna share with you a case. This patient, this is an angry email I got last week, or actually forwarded to me, a patient who had a really mishap. This patient went to the OR to have a Plorix, which mistakenly placed in the liver, so ended up coming to the surgical ICU with systolic blood pressures in the 50s to 60s on phenylephrine, norepinephrine, vasopressin, multiple boluses, and the question was, this patient was for a few hours in PACU, why he didn't get point-of-care ultrasound? So thankfully, the ultrasound was given. But here's the text I got last week when I was in a mission in Cambodia. The patient cramped in PACU, significantly hypotensive. The resident thought that the IVC was full. His dialysis patient, but got dialyzed the same day. So the first question, did we save the images? So this is the only thing will protect the resident from making this horrible mistake if he made a mistake. So my reply was, I think maybe there is very commonly, you can mistake the aorta for IVC. So probably that's what happened. If he's severely hemorrhagic, then the IVC should be flat. So, and then I remember when I was training my residents, using even artificial intelligence software to train the machine. These are basically machines that guide you to do the image, and then once you have a good image, it's auto-save it for you. So we tricked the machine here to let it think that the aorta was in the IVC, because here you see the hepatic vein that looks like an IVC, but it wasn't. This is the real IVC. So it's very easily to miss the IVC, especially when it's flat, and confuse it with aorta. Can you move next slide? Next slide. Okay, so thankfully the residents saved their images, and here is the subcostal IVC. It looks like full amplithoric, so he was right. Then the subcostal for chamber here. From here, you cannot really assess the left ventricular function. It seems not so bad, but this patient turned out to have some tachysuba, where the base contracts very well, but the body and apex is not. So this patient's IVC, which was full, didn't indicate the hypovolemia, because patients with diastolic dysfunction will have at baseline dilated IVC, and probably it was a little bit too early in the course to have a real flat, but that shows you that saving the images and having an expert review is important. We're getting an M&M this Thursday. Once I come back, we're gonna review the images and try to improve on performance. Next. So the structures to build a good POCUS service, really, again, I'm gonna go over. It's patient selection. When is, what is the question you wanna answer? And then you need to use adequate equipment by a trained provider or supervised provider to get the minimal images that you require to answer that question, and then there are a few steps that needs to be done. You need to record the study, assist the image quality, and then report to findings in clear and concise, clinical-oriented way, archive the images, and then we need to do quality assurance. And then when medical decision-making, that's where you need some support with the experts. What about courses? So the problem with the ultrasound courses or introductory courses, this is a survey done to the physicians who attended the first six pediatric point-of-care ultrasound courses. They looked at how many studies they did after they went home, and 10% were able to get more than 25 studies after they went home. The SSCM criteria for competency, what they think that you can achieve some level of competency is doing it, but maybe to get the emergency-level initial assessment is at least having 30 performed studies and 50 reads. So clearly, the majority of these folks who did the courses didn't follow through. What we did when we started our program, we hosted the SSCM course. We were able to convince SSCM to give us access to the videos, so we structured a four-day course for our residents based on these videos. So basically, the residents in the morning with one of the attendings who knows point-of-care ultrasound and critical care echocardiography, we review these topics in the morning. For instance, day one, basic physics, nebulology, how to obtain windows. And then the afternoon, we go and scan real patients, five patients in the ICU. The second day was for left ventricular assessment, vascular, intervascular assessment, then obstructive physiology in day three, and then day four, putting algorithms together. Also, we review 50 cases with pathology. By the way, this is something available. You can do it in your institution. You can talk to SSCM and get access to their hosting SSCM course. You can talk to me, I can get this, help you set this up. Then, as you see here, we review two hours of video lectures over four hours. After we watch the video together, we get the resident or fellow to make sure that he understands the topic. So I pause, I let him go up to the front and try to get him to talk about what he understood from the lectures. And then in the afternoon, we go and we scan real patients in the ICU, five patients. So at the end of the fourth day, they already done 20 scans. And then we did testing for them. SSCM designed the 30 multiple choice questions that we applied, and the results was really impressing. So the passing score for SSCM was 70% on these knowledge questions. And you can see here, pre-course didn't do well, and then we compared them to a one-day course group, which are the more senior and junior residents. The ones who did the four-day course actually did very well, and then six months later, they were able to show that they retained their knowledge. We took this multicenter, these are all the centers that are applying the SSCM curriculum right now. We apply the curriculum, we meet every three months via phone calls, and we make sure we discuss the problems that can arise. And you can see here, the fellows did even better. So the passing rate for the fellows was really low after the course. And then after six months, basically 97% pass of the fellows in these institutions. That's the impact of just four-day extensive course. So with the fellows, what we did actually, we did the course, and we did two days in one week, and then we give one week time for the fellows to apply the, to get the 10 studies in the ICU, and then the following week also, they did another 10 studies. We give them one month to achieve 20 supervised studies. We test them, we did OSCE exam. Unfortunately, this video, for some reason, didn't wanna work, but we apply modified Delphi methods, so when they're done with all this, at the end of the rotation, we get them to a simulation center, and we test them to make sure that they know how to identify structures and to procure good images. And then simulation, this is an example of a simulation study. Seven, six, five, four, three, two, one, one. Time for pulse check. Pulse check. Looking at our rhythm. Looks like a nonchalant rhythm. We have a pulse. Pulse. Pulse. Countdown, countdown, please. Five, four, three, two, one, one. We do have a question for you. Yes, yes. Yes. Let's hear it, let's hear it. All right, guys, there's no sign of cardiac infusion, so it's not cardiac cardiac epinephine. The right ventricle is small in size and hyperdynamic, so no increased risk of pulmonary embolism. Looks like we're underfilled. We need some volume, guys, guys. Okay. One milligram, milligram of epinephine. Some surgical bleeding, hemorrhagic shock. Okay, we've got surgery soon on the way. We're getting pressure fluid. Those are good quality, quality care. So after this, do you think that these residents did well on the simulation? They were very happy, but I told them they really missed one. Can you move to the next slide? One important point that they paused for 17 seconds, so they shouldn't really not extend the pause more than 10 seconds per the guidelines because perfusion pressure happens with good compressions. The moment you stop, the perfusion goes to zero and then take time to get better. There are two studies in 2017 showed that when you apply ultrasound during cardiac arrest, there's eight seconds longer pause when you apply ultrasound, but all these studies done where the trainees didn't perform this with a protocol and didn't get training with simulation. Again, just to focus that it's less than 10 minutes, so we showed that with more simulation, we were able to shorten the pauses and be able to stick to the less than 10 seconds. Now, we involve nurses in these simulations and basically we're giving the simulation every year for our residents. Okay, so now image acquisition. First thing is you need to save your images. To make it easy, we were able to apply a scanner where you can scan the patient's information, just the MR number, and then scan your ID. We made sure that we have this early on. It made a huge difference. So image starts with saving, and then after that, next slide. Okay, go ahead, go ahead. Okay, go ahead. Okay, go ahead, go ahead. I'm just gonna leave this out here. So then acquisition time. We don't do an interpretation during acquisition. It's usually the residents, in this case, it was a medical student. We get them to procure images and give them some feedback on how to get the image, but we don't do any interpretation at this time. And we never do cardiac ultrasound alone. It's cardiopulmonary. So always we do an abbreviated heart exam followed by lungs. Okay, so now reporting. Now, outside the room, we review these saved images. Right ventricular cavity was normal in size. Structure, right? Structure 100% normal. Normal septum. Left ventricular cavity was normal in size, but there was hypertrophy. We had very good image quality. So this is probably the, this is the external type of reading, reading of left ventricular hypertrophy. We also had the corridor IBCs, and then. So as you see, we go over, first of all, quality of images, structure. We go from top to bottom. For instance, for the subcostal view, pericardium, RV, septum, LV. So we go systematically. It's not about how many views you get, but actually going systematically over the views. And then after that, we report the findings to the team. The IBCs. So these are the three steps that needs to be done. Next. All our images are saved in the machine when we started. We used to take these images every month and put them in a secure hard drive that located in my office. I only access it. And then, but after that, we're able to get the workflow solution by third party. So basically, it's a cloud where all the studies go via cloud to the secured cloud. Nobody can see it except the sonographers and our department. Now, we review these images periodically, and if there is an indication to the study, and then we found that the image quality was good, we upload it to the PACS system where everyone else can see it. Otherwise, we just keep it, archive it, and use it for credentialing and use it for feedback. So here you see we're reviewing all the images. For the trainees, we even publish what image quality they were able to get. And then this is the privileging process in our institution where we give privilege for physicians who finish training and meet the requirements to do ultrasound. So the quality assurance, if you're a trainee, we look at all the images, but if you finish your training, we just sample some images to look. And this was all done because there was a buy-in from the chair, and I was able to get some non-clinical time built in so I can go. So it's very important to apply, to establish program, a POCUS program, is to have a good mentor or leader or director with buy-in from the institution. It's not, because otherwise, you need to apply some quality control. I'll talk very quickly about billings, why we need to care about money. I noticed that the folks in the emergency department gets their ultrasound renewed every few years. They have the best gadgets, and I was begging for a probe, and it took me a year to get $12,000 to buy a probe when I started. So the key word here is billing, that you can convince your administrators to give you some money so you can get the ultrasound. So billing, here are some basics about billing. ICD Code 10 is basically the language that the WHO use to communicate why the exam is indicated. The patient is hypotensive, I don't know why, for instance. So, and then, you will do the examination, needs to be, images needs to be saved, and it should be saved in the PAC system, where images should be viewed by everyone. And then, you need to document your findings, and then you apply CPT code. CPT code is a code for the procedure you've done. And then, claims, payer, and then you get reimbursed. So TrueScan is saved image, physician documented. What is the CPT code? It's the code for the procedure done. All our codes are for limited exams, and the professional fees is how much you get paid for interpreting the images, which is not a lot, $45 average, but technical charge is what the hospital like. It's basically charging for using the ultrasound, and that's how you can convince the administrators to get you ultrasound. Now, if you do not, if there is no saved images, that's called blind scan, you're not gonna get paid. If you do it without indication, you shouldn't get paid. Illiterate scan is you save the images, but you didn't do documentation. And then, miscoded scans, if you apply the wrong code, you're not gonna get paid. So phantom scan, obviously, is if you did not save anything. If you just did a scan and say, hey, but yeah, that's what I found. So I found this actually billing. We start billing for applying ultrasound and cardiac arrest, and it helps us get in, make sure to apply all the standards. So clinic integrations, we integrated this during COVID. This is a surgical resident who gets one day training on applying just a simple subcostal view, but all these images are reviewed by the attending who's critical care ECHO certified during rounds. So pre-rounds, they go in to the patients. They apply a very quick cardiopulmonary exam based on the subcostal view. So here, she looks, this patient's hypotensive, by the way. She looks at the subcostal. She see that both ventricles are hyperdynamic. Now, she's just saving the images. I'm not sure where it started, but anyways. And then, after that, she got a quick look at the IVC, and then the aorta and the upper lungs. Next. This patient was having, because the IVC was full and because of all the overall statement, the overall findings, we concluded during rounds that the patient has distributive shock, and we didn't give fluids because the patient was not fluid-tolerant. We started norepinephrine. These are B lines in the lungs. All patients with ARDS had that. So the time consumed doing this exam was less than five minutes, and basically, what we did is, if I am not happy with the exam or if there are signs of chronic disease, then I go inside the room and do the exam myself, and I needed to do that only 16% of the time, and 84% of the time, I was satisfied with just the images. Next. We published these hemodynamic phenotypes to help guides. We used this tool to discuss the findings with our residents, NPs, and get everybody on the same page, and that's about it. Just a quick glimpse of how to apply this or how we were able to apply this in our institution, but the key thing is having supervision the entire time. Images are saved. Quality assurance is the key.

Ultrasound quality assurance

I-AIM framework

Medical training courses

Ultrasound image archiving

Billing codes

Integration into clinical practice