 Live from Washington D.C., it's theCUBE, covering .conf 2017, brought to you by Splunk. Welcome back to Washington D.C., nation's capital here, .conf 2017, as theCUBE continues our coverage, the flagship broadcast of SiliconANGLE TV. Along with Dave Vellante, I'm John Wall, I was glad to have you with us after we've had a little lunch break. Feeling good? Feeling great? Yeah. Good conversation with some customers, dug into the pricing model. Yep. Yeah, got some good information. What'd you learn at lunch? Any good? We'll talk about the end of the day. All right, good. Look forward to it. Let's talk healthcare right now. Derek Merck is with us now. He is the Director of Computer Vision and Imaging Analytics at the Rhode Island Hospital, which is the teaching hospital for Brown University. Derek, thanks for joining us here on theCUBE. Good to see you. Absolutely, very excited to be here. Yeah, so, well, and as are we to have you, Director of Computer Vision and Image Analytics. So let's talk about that. What falls under your portfolio? And then tell us where does Splunk come into that picture? Well, it's been an interesting journey. Rhode Island Hospital is a huge clinical service, takes really good care of the people of Rhode Island. I'm in diagnostic imaging, so I work with all of the CT scans, the MRs, radiography, ultrasoundography. And what I try to do is automate the data that's coming off all of these machines as much as possible. So, you know, typically a patient will come in, they'll get imaged for some reason. The physician will take a look at that image and make a diagnosis. And then that image goes into an archive and it may be used again later if the patient comes back. But other than that, it's not really used at all. And you know, with the sort of emergence of computer vision, access to training images, sets of data has become really critical and diagnostic imaging has become really interested in taking better account of what imaging they have. So that they can try to answer questions like, what's alike about these images? What's different about these images? And automate diagnosis. What's similar about all of the images from patients who have cancer versus patients who don't have cancer? Which is basically what a radiologist's job is, is to go and look at this patient's image and figure out, does this patient have cancer or not? So that's the way you would teach a computer how to do it in an automated fashion. So I spent a lot of time trying to figure out how do you keep better track of what's available and be able to ask these sort of population-based questions about what we have in our portfolio of data, our data portfolio. And I spent a lot of time writing systems by hand in Python or other kinds of scripting tools. I spent a lot of time trying to interface with the hospital informatics systems, the electronic medical record. The electronic medical records, again, really meant for taking care of patients. It's not meant for population analytics. And so we ended up basically building our own healthcare analytics system just to keep track of what we had, what were the doctors saying about different cases? Show me all the cases where the doctors think that some particular thing happened and be able to ask these questions in real time, generate huge data sets, anonymize them, run them through computer vision algorithms, train classifiers. Diagnostic imaging's really excited about this kind of technology. And then there's been a lot of interesting side projects as well. One of the things that administration is most interested in is we're keeping, because of these kinds of systems, we're keeping a lot better track of radiation exposure per image. So the CT scanners will tell you how much radiation was used for an individual study. But again, our analytics systems, historically, you have no way of saying what's the average, what's high, what's low. And there's months of latency, six months of latency between when you run a scan and when maybe the American College of Radiology comes back and says, some of your scans were a little high in radiation exposure. Whereas now, because we keep track of all this data, we have these real-time dashboards. And that's the kind of thing that we use Splunk for. So we keep track of all the data that we're collecting and then we create these dashboards and give them to people who haven't had access to this kind of analytics before, for looking at utilization, optimizing workflow, things like that. Yeah, I'm just kind of curious, when you mentioned it, like X-rays, maybe Dave, you know more about this than I do. But it seems like that it's kind of a standard practice. You have a certain amount of exposure for a certain amount of tests and that data, I don't know how, but it sounds like it's more critical to have that kind of data than a layman might think. That I was curious about the analytics of that. Like what are you using to determine there in terms of that exposure and? There's always a trade-off with radiation-based imaging. So there's a lot of non-radiation-based imaging, like you may have heard of magnetic resonance imaging or MR, and those are thought to be perfectly safe. Like you could get MRs all day long. And in fact, they do give MRs to people all day long for research purposes sometimes. You climb in the tube, I don't want to climb in the tube. You get a little claustrophobic. Expensive. Yes, yeah. That's the thing that we don't have very many of them. They're very slow, but they're safe. Ultrasounds, very safe. We give ultrasounds to pregnant women all the time, very safe. But they don't give you great quality images back. They give you a very small field of view and things are wiggling around. A CT scan is super fast and it gives a physician all the information that they need in a snapshot. CT scanners are so fast now they can freeze your beating heart, right? They can make a revolution around your body of a thickness so that they can capture your heart while it's in motion. You know, like with anything, if you have a camera and you take a picture of somebody running across the screen, you don't see the person. You just see this kind of blur, right? But now with modern fast aperture cameras, you can take a picture of neutrinos and things that are impossibly fast. I don't know that that's actually true. You might want to add it back. But a CT scan is the same sort of thing where you can take your heart's beating all the time, your lungs are moving all the time, your bowels are moving all the time, your blood's coursing through your veins all the time and it's so fast it can freeze it and give you this volumetric data back. And they use that for all kinds of different things that they're not able to do with other kinds of imaging modalities. The downside is they're potentially somewhat dangerous, right? People have known since the 1890s when X-rays were first discovered by Wilhelm Rinken that if you put somebody under an X-ray beam for too long, you know, your hair will fall out, you'll get skin burns, all kinds of things that these early pioneers of X-ray did to themselves without realizing and documenting all of these problems that could happen. And a CT scan uses ionizing radiation. If you get too many CT scans, you'll get skin reactions or other kinds of things. So it's really important to keep track of the risk-benefit ratio there, right? And people give you a CT scan. If you fall down, you hurt your head. They give you a CT scan because they're worried that you're gonna die if you don't get a CT scan, and so along with that is this idea of like, how do you track how many CT scans an individual patient gets in a year? Right now, the hospital has a hard time keeping track. If somebody comes into the emergency room of automatically identifying, oh, this patient's already had six CTs. Should we put them in line for an MR instead of another CT? But again, these are the kinds of things that we're able to get at through using, through better management of our data, better organization of our data. You mentioned you're doing more of this real-time analysis. Splunk obviously is a tool that helps do that. Other tooling, are you using cloud-based tools? We have to be really careful about cloud-based stuff because there's this protected health information that everyone's really concerned about. Working with data at the hospital is really walking a fine line. You need to be very conscious of security. They're really reluctant to let non-anonymized data out onto cloud sources for storage. There are some ways of getting around that, but basically we run all of our servers in-house. So there's a couple of big data centers down in the basement of the hospital. Mostly they have clinical duties, but we have a number of research servers that are installed down there as well. And they're managed by the same IT staff in this kind of hardened architecture. And I actually can't do any work from home, which is an unusual kind of experience for, you know, I'm used to being able to log in remotely. Oh, Jarn. Well, no, or you spend too much time on the job, sometimes you'd like to. I'm ambivalent about it. There's goods and bads about it, but you know that. Okay, so how do you deal with that streaming infrastructure and real-time analysis? You guys sort of build your own? Yeah. With open source tools or whatever. I use a lot of open source tools. Traditionally the hospital wants to pay for everything because they feel like if they pay for things it comes with uptime guarantees. That's all I am. So when I build my systems though, because I'm working on shoestring budgets and things. And because I believe in open source. I use open source wherever I can. We have, I wanted to mention we're actually, for a lot of the work that we do, supported through Splunk for Good. So I don't pay for a full Splunk license. Corey Marshall, who runs Splunk for Good has sort of recognized the value of some of the stuff that we're doing with dealing with non-traditional data. It's not the sort of standard things that the other people who are working in the healthcare space with Splunk are working with. You know, we're working with imaging data. We're working with patient bedside telemetry data. You know, the EKG signals and the heart rate signals and aggregating all of this stuff in one place to make more sensible alerts and alarms. Oh, you know, this patient set off an alarm three times in the last hour. I should send a page to the nurse who's taking care of this person. And it's different than the kind of business optimization that I think a lot of people in the healthcare space are using Splunk for. So you have your core mission around diagnostic imaging but then as we sort of touched on, you have all these other peripheral factors in your industry. The Affordable Care Act, obviously there's HIPAA, there's EMR, there's meaningful use. How much does that affect your mission? Does it sort of get in the way? Is it something you have to be cognizant of? Like constantly, obviously HIPAA. Oh yeah. But other factors. I try to just be cognizant. I don't let anything get in my way because almost all of these things that you talk about, I mean they're really meant to protect the patient and make sure that everything that I do as I'm working with data that we're anonymizing things, we're using data securely and we're trying to help the patients. And so I think I just have this kind of moral check in my head of like is what I'm doing right now good for my department, good for my institution, good for my patient. And then because I'm aware of all these other rules, they're very complicated and difficult to navigate but I at the end of the day can say, I understood that rule, I followed that rule and what I did was the appropriate thing to do. And it's like house rules. Yeah, yeah. Okay, talk a little bit more about Splunk, how are you using it, what it does for your mission, for your operation? So what I came to the conference this year to talk about was this dose management system that we built that I think is really important. We've had vendors coming in and telling us that Medicare isn't gonna pay hospitals or is gonna reduce reimbursement to hospitals who can't prove that they're using ionizing radiation imaging appropriately. So what does that mean? Nobody quite knows exactly what that means. How do I tell whether my hospital is adhering to these rules that are ill-defined? And these vendors are coming in, they're trying to sell us solutions that are like $100,000 a year licenses. And so administrations taking this seriously, they're trying to figure out which of these vendors are we gonna give money to. And in the meantime, a bunch of the CT technology staff and I basically put together a system that answers all these questions for them using Splunk. So we use Splunk to collect meta information about how all the scanners system-wide are being used. We have 12 CT scanners. They shoot 90,000 different studies every year. Each one of those studies may be hundreds or even thousands of individual slices of data in these volumetric data sets. It's a huge amount of data to keep track of. And you're not using Splunk to keep track of the imaging per se. You're using Splunk to keep track of what imaging you collected. So it's a small fraction. It's just the meta data about each one of the studies. And that meta data comes with a bunch of interesting information about what the radiation exposure for each one of those studies was. And Splunk has these wonderfully adaptable, easy to use tools that once we convert our strange DICOM device independent communications in medicine data, we flatten it, normalize it, turn it into generic data. It's JSON, it's dictionary files. And then Splunk has these great tools that can be applied instead of to business analytics and optimization to image analytics and optimization. And so we build our dashboards on top of Splunk to show per institution what was the average dose per protocol, per body type. You can track which technologists have the lower doses and higher doses. We found all kinds of interesting things. My favorite story that the chief technologist was just telling me, I was putting together my slides for this presentation that I did here about this. And I said, we need an example of a dose outlier sometime when we had a higher than expected radiation event. And so he, and we never have dangerously high radiation events, right? Which is- Good caveat, thank you. That's all the machine's track. All the machines care about is whether you're harming someone and we never harm anyone. The machines don't track, this one's a little higher than what you would expect it. So that you can say, well, why is that? What happened there, right? But now we do, using our Splunk dashboards, right? So I asked him, hey, can you give me an example for my slide deck? And he literally just looked over at the monitor that he had open. And he says, oh, right here. Here's a patient who had a 69. These numbers are irrelevant, right? They're supposed to be 50. He knows what the numbers are supposed to be. To me, numbers are just numbers. This patient had a 69. And he picks up the phone. This was five minutes ago. He calls down to the control room and he says, I'm not blaming anyone, but why did Mrs. So-and-So have a little bit higher radiation dose? 69 is not dangerous, by the way. The alarms don't go off until like 75 or 80 or something like that, right? So he just called and he asked like, hey, what was going on with this patient? Oh, she had a dislocated arm. Okay, I understand. This was a head scan. I'm like, Scott, what does a dislocated arm have to do with a head scan? He said, well, she went through the CT bore with her arm up over her head, which is not the way that, but that's the only way she would tolerate. And so the CT thought she was this big and it had to raise the amount of radiation that it was putting into her to go through a larger object, right? And so he documented that. He put it down and again, we use Splunk for ticketing for outlier identification. So he put this one into the outlier identification database that we have. He picked, you know, other for the reason because we don't have a dropdown menu with dislocated arm and marked it as closed. It's justified. So when the, when JCO, the Joint Commission on Hospital Accreditation comes through and they say, what do you do to manage your higher than expected radiation exposures? We can both say, well, we never have unsafe radiation exposures. It's all documented right here. And when it is higher than usual, this is the way we document it. And here's examples of, you know, 10 or 20 of these odd instances where, you know, something happened and either it was completely justified like this lady where machines were used appropriately, that was appropriate, or very occasionally we'll find something strange, like an improper head holder was being used at one site for a while. And it was resulting, you know, these head CTs should usually be around 45 or 50. And instead they were 55 or 60. And so they went and they just took the, what a metal head holder that they were using and replaced it with a carbon fiber head holder that they should have been using. And then all of a sudden our doses came down and we could document it. You just tell me, it was a dislocated arm. Let's leave it at that, all right? And we're happy with that. Derek, thanks for being with us. Absolutely. Appreciate the time here on theCUBE and glad to have you here and continued good luck with your work at Rhode Island. Thank you very much. You guys have a great day. Very good, thank you. All right, Derek Merck joining us here on theCUBE will continue live from Washington DC right after this.