Thank you for joining. My name is Mariam Aboian. I'm an Assistant Professor at the Department of Radiology and Biomedical Imaging at the Yale School of Medicine. I will be presenting a demo of PACS and electronic medical record. A lot of times, when we think of medical record, particularly those of us who started medical school a long time ago, we think of a chart of medical records in a stack. This chart was being delivered to the office where the patient is being seen or to the hospital floor. We would open the chart, write notes in it, look at the different values. We constantly would be leafing through the chart or taking you to different places. Then people would be trying to find us to get the chart. Or sometimes we would be waiting a while to get the chart from the medical library. In addition, we also had an imaging folder where a lot of the imaging scans were located. We would have to find a view box, a light box where we would hang up the image and look at the image. This was very slow and very cumbersome. When electronic medical record and PACS came, it was very widely accepted and well received. Because it allowed us to put all of this information in easily accessible location and have it be very organized. This is an example of electronic medical record that is at Yale. It's an epic software. You can see that we have an ability to search by a patient name or their medical record number, or date of birth. Here, I pulled up a specific case of a patient Info Serv Zzztest. If you pull this patient up, you can see that their demographics are displayed on the sidebar and you can actually get a little bit of their information. You can even get the preferred language the patient wants to be speaking. When you go to their detailed information, there's a couple of different tabs, such as SnapShot tab, which provides you basic very quick information. In the SnapShot, you can see if they have the COVID vaccination, what their problem list is, what their significant history and details are. You can also find what their team is like and who's taking care of this patient, and if you have an emergent finding, who do you need to reach. You can also get information on allergies, medications, immunizations, pretty much a lot of the basic information that you want to know when you're evaluating a patient. Now, if you need more detailed information, you can go to another tab, which is the chart review. The chart review has a couple different tabs that we can go to. One of them is the Notes tab where doctors and nurses put in their information in a standard medical record format. Here's an example of a progress note or a telephone encounter that will provide me information about the patient, why they're being imaged, who is ordering the study. You can also search this record, which is much more difficult to do if you have a paper chart. In the Search feature, you can identify information based on encounter date, or based on status of the order, or the service that ordered the exam. If somebody has a headache and you're trying to image their MRI brain for the headache, you want to know, did the patient come in with facial droop? Did the patient come in, was his headache instant or was the headache happening for a couple months? You can get all of this information from the electronic medical record through doing these search analyses. In addition, there's other tabs such as Labs where you could have laboratory values displayed or pathology. For this patient, there was no path report or imaging. Here, for this particular patient, he had multiple imaging studies performed. We're going to look at a specific imaging study, which is a PET CT with a tracer Amyvid. You can see there's going to be a report there, but there's also a link to where we're going to look at the images. If you click on that link, it will pull up an additional set of software, which is our PAC software, our company use Visage Imaging. It will allow you to view the images. Now you can't just view these images on any computer, you actually need to have very specialized hardware to view these images. This is our PACS workstation where we view these images. This is actually my personal workstation. You can see that I have to medical grade monitors in the middle. Here is where I make the diagnosis. Here is where all the images of my patients are displayed. I scrutinize the images, I reorient them, I analyze them. These monitors have very specific regulations for what they can be, I can't just buy any monitor off of Amazon. Here's another two monitors on the sides. These can be any monitor that you can purchase. This is where my electronic medical record is displayed and my dictation software is displayed. Here's a dictaphone that I use for dictating their reports and we have the keyboard and the mouse to get around. How does PACS looks when we pull it up? Here's an example of what comes up when you open research PACS or clinical PACS. You see a Study Browser, and you can search for the patient or directly pull up your patient from Epic. You can also look for studies from based on the date of the study or modality of the study. You can also look based on the accession number of the study because each patient's study has a unique individual accession number and each patient may have multiple imaging studies so for each medical record number, you can have multiple accession numbers. You can also search different databases. Here's an example of me looking at recent studies or being able to reach out and pull up another database from archive or if a database of images is being stored in another server. Now, let's go back to the original VISAGE research database and here's an example of a patient. This is a patient with a very large brain tumor and they presented with a headache to the hospital. I can view the information on how they presented and how long it took them to present and what their symptoms are from electronic medical record, and I can analyze their images on the research PACS platform. As you can see, we have a display of different sequences of this MRI in the top row, and the protocol that I'm using here actually reconstructs the sequences into a coronal plane in a bottom view. I'm actually able to look at this tumor in three dimensions, and I can look at it with many different protocols of image acquisition, which you can see all the protocols in a ribbon here on the bottom as we scroll through. Now, most of the radiology image analysis is done qualitatively, and that has been historically, since introduction of PACS and since use of film because we have focused on the expertise of the human eye. We are starting to enter a new era in radiology where we more and more are starting to use semi-quantitative image analysis for our diagnosis. Here's an example of what we're doing in my laboratory. We have developed a couple of different algorithms that are plugged in into the PACS software. This is actually a published algorithm PyRadiomics, and this is our own developed auto segmentation tool for brain tumors. When you press this button, the auto segmentation tool, you'll see these different contours that are drawn around the tumor, that are able to contour the tumor edema, the tumor core, and the necrotic portion of the tumor so it's able to tell you very different portions of the tumor. Why did we do this integrated approach to patient care? Well, it turns out in medicine, we don't just see a patient and form a diagnosis, we actually our inner circle of service when we take care of patients, we don't just sit in our little rooms. If you think about it, we provide service to our patient by providing these personalized medicine reports to them with detailed reports that have meaning to our patients, with language that is understandable by a patient. In addition, we're providing service to our referring providers where they generate detailed reports that actually help with care. They don't just form a simple diagnosis, they can actually track how a tumor can grow over time and that will provide the oncologist, or a neuro-oncologist, or endocrinologist, how they should be treating the patient and when they should initiate new therapy. We also serve our advanced students such as neuroradiology fellows with tools that save time and provide image interpretation assistance. As you can imagine, radiologists are specialized in multiple different areas and we have to cover diseases through the entire body. It is not possible for a single person to know detailed information about many, many different diseases in the body at the same time, so we're able to generate image interpretation assistance tools. If somebody is an expert in pancreas imaging, they may not be an expert in brain tumor imaging, but they may have an image assistance interpretation tool that will help them diagnose complicated brain tumors, even though they're not an expert in that. In the end, we also provide service to our students with teaching of medicine, teaching of research methods, promoting careers, and opening up doors. A lot of these approaches are geared toward making understanding of medicine easier through a multimodal approach of understanding patients. I'm very excited about this field as we're incorporating novel AI tools and quantitative image analysis into clinical practice and it's really going to change the way we practice neuroradiology. Thank you.
