 Welcome back. I am Colonel Melinda Eaton. The final session of the morning will feature four speakers, each bringing their own perspective to military medicine and the impact of medical research to delivery of warfighter medical care. Our first speaker, Sergeant First Class retired Luke Shuley, will be introduced by Dr. Melissa Givens, Department of Military and Emergency Medicine, Uniform Services University. Dr. Givens. Good morning everybody. I'm really excited to be up here and introduce you to Luke Shuley. He's going to talk to you about his experiences, but the reason I'm up here introducing him is because I want you to know how insightful this young man is and how much he can contribute to the military health system and to the greater civilian world. I got to know Luke when COVID was raging in New York City. In the early part of the pandemic in New York when things were really at a crisis state, we stood up a field hospital and responded to the pandemic. Luke was one of the first to volunteer and come share his skills. And not only did he show up and volunteer, he really shined as a leader and even after we shut down the field hospital and the rest of us went home, Luke stayed in New York and really advanced medicine along with New York Presbyterian. And so I have seen him just shine in every example and I think you're really fortunate to hear his story, so I hope you enjoy him. Luke, I'm going to go ahead and bring you up. Thanks, Missy. So my name is Luke Shuley. I'm going to give a quick presentation, a little bit of an AAR, a little bit of a clinical evaluation and some hard lessons learned that I went through that I think that a lot of people in this room, especially the people who are still wearing uniform, hopefully some of this hits home with you. We're going to cover some good things. We're going to cover some bad things. I'm going to tell you my story and why I try to continue to tell this story, especially on a government funded platform or there's government entities involved because these are where some of the challenges that I ran into can be addressed. We'll try to bridge the gap on the civilian side as far as how some of the things that we're doing on the civilian side also have extreme impacts on the Department of Defense side. So like you see here is January 18th, 2018, 0305 in the morning. I found myself in a nice neighborhood in the Nalazad district of Helmand Province, Afghanistan and we were conducting a small village clearance operation before we get started. Next slide. So like I mentioned is we're going to cover some things in this discussion. Some of them are going to be uncomfortable. Some of them are going to be our failure. Some of them are going to be my failures and some of them might not be in the best limelight for the organizations that are here. I'm going to say these things on behalf of myself. I don't represent the 10th Special Forces Group, the United States Special Operations Commands, First Special Forces Command, the Department of Defense or the US Army. I also work for Carnegie Mellon University. The things that I'm covering in this lecture separate from my lecture tomorrow is not on behalf of the university as well. The focus of this story for me is to tell you how I went from the guy on the left to the person on the right. I went from becoming a Green Beret to an advocate for the Department of Defense Special Operations Medics and taking care of our wounded warrior population. So like I mentioned earlier is we were conducting a village clearance operation. It was the ODA that I was on, four additional Special Operations ODAs, seven CH-47s, one UH-60, four Special Operations Kandax and one National Introduction Unit. The primary objective of this particular mission was a village clearance operation where we were infilling to the north, clearing to the south at a predetermined limit of advance, which happened to have been a Taliban madrasa. If you spend some time in the military, you can probably understand what happens in the madrasa and why we were using this as our limit of an advance. My position, although I was the senior detachment medic, is I was also the assault leader for our CQB-1 element. My primary objective was to get to that madrasa, set up a stronghold, conduct counterintelligence and then assist in personnel recovery out of the Taliban-controlled prison which was across the street. So team internal medical situation, just to give you a quick rundown, is although I was the senior medic for the team, is I was also in a assault leader position. This was something that me and my boss went back and forth on as far as whether or not it was a good position for me to be in. And the worst case scenario is what we ended up going through. His argument was always that if something happens tactically and we need to get our medics involved, my medic can't be an assault team leader. My argument being a stubborn SF guy is you're always a green beret first and whatever else is secondary. We went round and round through that deployment. Finally, I was in that assault leader position. The junior medic that was there, great dude. It was his first deployment. He was located in the rear of the element with the C2 element and we got separated. Once we got to that prison, which we'll take another look at, is we got into a gunfight, we got split up from the leadership element. And conveniently enough is our warrant officer, who people say don't do anything in the military, was also a prior 18 Delta. And it was good because we needed a second set of medical hands. So leading up to the point of injury, you can see there in the top part of that picture is that's the madrasa where we were held up on. We started having issues with our partner force. We noticed a decrease in the security amongst our Afghan counterparts is they quit wearing their night vision goggles, they were sleeping, they weren't carrying their weapons. On top of that is we started getting shot at from the buildings in the bottom left corner. So we knew that there was a physical threat. There was a tank in the middle between the madrasa and the prison. We went to the tank there. We set up a small drone to take a look into the prison just to make sure or just to see what was going on on the other side. We noticed that there were you know the Taliban knew that we were there they were moving machine guns and stuff to get ready to ambush us coming through the building. So we threw a couple hand grenades over the fence or over the wall which was about 14 feet tall. And we moved to our breach point. So our original breach point was the blue circle in the bottom left corner of the prison. And as soon as me and another guy went around that corner we started getting shot at. The problem with the the well there's a couple of different problems. So one is the plan now changed. So I made a decision to change our breach point from the southwest wall to the west wall. But I had already tied in a six pound C four charge. Anybody familiar with ballistics or explosives? Anybody? How far back should you be from a six pound C four charge? Long ways. So I had a 30 foot initiation system, which is about a third of what it probably should be. And I was already tied into it. So I took a six pound C four charge that was supposed to go through a four foot wall. And I put it on a wooden door. Because at three o'clock in the morning under night vision, you're not going to sit there and try to untie it. And we were getting shot at. So I took the six pound C four charge said this will probably do. And and moved back. So I put the charge on where that lightning bolt was. And I moved back to where those two green circles were at. And I was there with a buddy in mind. His name is Andrew. And that whole morning we had been doing these six pound C four charges was 30 foot initiation systems, because we never double checked what our msd or our minimum safe distance was. So the msd for six one is you probably shouldn't even be using six pounds of C four. But again, like I'm a dumb SF guy, like if you don't know more, just add to it. And that's exactly what I was doing, but with explosives. So if you're taking notes, don't do it. So the whole time, like we were eating these C four charges, like we'd put a charge on a wall and then me and my buddy Andrew, we would take turns hiding behind each other so that his body would shield me from the blast and then my body would shield him from the blast. Who does tbi research? Let's talk after. So you know, we get behind this corner. And although we were getting shot at like I remember I was on a knee. I'm sitting next to Andrew. I'm like, Hey, man, at least we're not going to eat this one. Like this is going to be the best breach we've had yet. At least we have some cover. So I pull the initiators, the breach goes off. So lo and behold, the building that we were building breaching into. So now this is this is from the north to the south looking at the building. That building blows up. So on the inside is the EOD team that was with us and some of the other guys that were there. We had a jet advisor with us. They found some 55 gallon drums, some remnants of HME in the building that we were breaching into almost simultaneously exploded. So whether it was a sympathetic detonation or an over pressure that led to an explosion of another ID, the building blew up. That arrow that's right there is where me and Andrew got stuck. On top of all that. Now we got reengaged in another gunfight. My team sergeant who was on the other side of the tank, he got knocked unconscious. And nobody else was with us. The other guys that were a part of my SF element were out doing their own operations. The Afghans that work with us don't speak English. And it was the only the only way that people really knew what was going on other than the explosion was they came to find my boss. And he was walking around the middle of the street. I mean, there's I mean, it's not a hellacious gunfight, but there's people shooting at each other. And he was walking around in circles in the middle of the street talking nonsense on his radio. Somebody came back and got him. Finally, he kind of comes to and he remembers where we were at. The dust had settled. They moved another SF contingent over to start working the debris off of us. Some of those pieces they look big, but they were extremely heavy. We couldn't all these the guys that were on the ground couldn't even pick them up. So we ended up or we didn't I didn't do anything. They went they found a car took the car jack from the spare tire and lifted up all the pieces of debris with a car jack spare tire manual lift. So initial impressions is I knew that I was pinned under a building. I knew that Andrew was with me. I couldn't reach my comms. I couldn't hear because my comms have been knocked off and I couldn't see because my nods were damaged. I knew that I wasn't dead, but I knew that I was not in a good position. I knew that I was either paralyzed or still pinned and then straight panic. So you might be asking what's the difference between panic and straight panic. So panic is like probably what you're used to. Straight panic is whenever you're doing a operation at three o'clock in the morning where you're supposed to be quiet and instead you're sitting underneath the building screaming for help because you have no idea what's going on. And that's exactly what happened to me is I knew that we were actively engaged in an enemy fight and I could not hold it together because I was terrified about what had just happened. I completely lost sight and visibility on the fact that we were actively engaged in a gunfight and lost my mind based off of what happened. So as I'm screaming for help, the other guys are coming over. And again, I wasn't 100% convinced that I was paralyzed at the time. I thought that whatever happened, everything was on top of me. I couldn't see anything. I really couldn't feel anything. My sensation was all messed up, but I didn't know exactly how bad it was. So point of injury care, the recovery team kind of came over, got everything off of us. We called in close air support. Luckily for me is we still had A-10s. 15 minutes to get me out, 45 minutes to get my buddy out. Initial impression from the medic on the ground was they thought that I either had a broken femur, pelvis, neck and shoulder injury. I was unable to move. They tried to extremity carry me, started pain management via 800 micrograms of fentanyl and then moved me back to the madrasa where we set up a CCP. They got an IV in me, did some improvised immobilization, and then pain medications and sedation with ketamine and versed. Andrew, the guy next to me, had four broken ribs. I don't know how. The story I tell is that my body protected him. And that's why this happened. But so they moved us back to the madrasa and we called in Medevac. Next slide. So the extraction, we called in the Medevac. The nine line went in at 0344. I didn't get Medevac until 0445. So the helicopter, the HLZ that we were in, the Hasty HLZ that we were going to use, it was too small. So the helicopter gets overhead, says they can't land there. There's not enough room for the aircraft. They don't have hoist capability with them. And although we had it on the ground, they didn't have it in the air. So we had to move all the way back to our infill HLZ, which is about 300 meters away. It took us a little bit of time to get there. We had to back clear through all those buildings because now they had been occupied by bad guys. And it took us about an hour to get there. Just a quick little write up from the Medevac crew that was there. The biggest take away from this is I never carried a pelvic splint because they took up too much space. And there was no good way to carry a cervical collar in my aid bag. We were doing a bunch of stuff where we were flying in on helicopters and then we were walking whatever the distance was to wherever we were going. And there's an old saying that ounces make pounds and pounds make pain. And if I didn't have to carry it, I wasn't carrying it. So it was at this point that I finally got a pelvic immobilization device on. Quick little write up from the forward surgical team. I know it's a little bit tough to tell. This was where they did a little bit more of a diagnostic, not necessarily imagery, but they had a little bit more time to take a look at what was going on. And this is for me whenever things started to getting a little bit real. As I started having worsening sensation and range of motion on my left side, they put a C collar on. I started having suspicion of a spinal cord injury. Anal tone has already started to be decreased. And they started preparing to send me to CAF, the Roll 3 for more diagnostic imagery. So if you spend any time in Kandahar, especially as a medical provider, this might look familiar to you. This is the ICU in CAF. CAF Roll 3, just a little bit more of a write up as far as what they found there. So finally now I get confirmation that I broke C4, 5, and 6. I had a right scapular fracture and a left acetabular fracture. And I was pending transport to launch stool for neurosurgery and spinal decompression surgery. This was also the first Asia test I had taken. So it was categorized as an Asia D, state, and Asia D throughout the entirety of the process. So a quick write up from launch stool is C5 vertebral body fracture, C4, 5, lamina fracture, C6 compression fracture, everything else that you guys can see up there. The neurosurgeon that was in Afghanistan at the time didn't have the equipment to be able to do it. Although they wanted to do the decompression surgery faster and sooner, they didn't have the resources available to them to do it in CAF. So he flew with me to launch stool and I had decompression surgery done in launch stool. Not one of my best pitchers. This is the ICU in launch stool. This shows a pretty good picture of the anterior and the posterior fusion C4 through C6 on the anterior and C3 through T1 on the posterior. I stayed in launch stool for about 24 hours. They flew my family to Germany to meet with me just because of the severity of the injury and then we flew back together on a C17 milbird with the CCAT team from launch stool to Walter Reed. Here's a quick little addendum that got added to my records. Basically it says that at one point throughout the transport my ventilator quit working and that they had to continue to manually ventilate me or manually yeah ventilate me with an anbu bag because the two backup generator or ventilators they had also quit working. If you spend any time working with manual ventilation it's hard to do good. We might think that we do it good but you squeezing it five times really fast and then not touching it for 12 seconds is not good ventilation. And this is one of the areas that so I end up getting diagnosed with an anoxic brain injury and I think that this is probably one of the instances or the instance that cost it. So my wife at the time was a nurse was riding in the C17 with me and I'm not putting anybody on blast with this email or with this addendum but she was sitting in the back of a C17 started noticing that alarms and stuff were going off nobody else picked up on the fact that the alarms were going off. I ended up vomiting and she couldn't get a hold of anybody in the aircraft and had to start like manually taking some of that stuff out of my airway before somebody came over and started picking up ventilations. So get to Walter Reed I get issued my first army issued power wheelchair I would have that wheelchair for the next six months. I spent a little bit of time in the ICU at Walter Reed and then I had reconstructive pelvic surgery my chain of command had come down to Walter Reed because we had seven other guys injured within about a two-week period and we did our ceremonies and stuff at the hospital. So had reconstructive pelvic surgery from the less acetabular fracture I was having a little bit issues with my blood pressure in the ICU so the orthopedic surgeon that was there they actually had a cadaver come in where they could break the cadaver break the cadaver's pelvis in a way that was similar to mine they were able to practice the surgery the day before doing the surgery and then the following day I had surgery. So this is what the end product looked like. So I spent about a month and a half at Walter Reed Walter Reed amazing hospital super up-to-date plenty of resources a lot of other wounded warriors at the time doesn't have a spinal cord injury program. So we asked about the spinal cord injury program and apparently there's one in Tampa so the Tampa VA has a spinal cord injury program that they transferred me to and I was there for another from 26 February through 26 May. Some additional diagnosis is outside of the spinal cord injury incomplete tetrapalusia spasticity neurogenic bladder insomnia all the other ones that were up there is I got admitted to that that hospital on an inpatient status spend a couple months there before being discharged. One of the things that we're going to revisit is the resources that are available how they're used and where they're implemented at UPMC. So UPMC is a University of Pittsburgh Medical Center anybody from the Department of Defense have a relationship with UPMC for active duty status we don't have one. So whenever I was in the Tampa VA and they were getting ready to discharge me because I was able to start walking again my unit was ready for me to come back to service and so was I I wanted to go back to service I was at 10 Special Forces Group out of Fort Carson, Colorado and I was dying to come back there is no spinal cord injury program capable of being run at a special operations unit like 10 Special Forces Group. The group is designed to fight wars and to go to war not to be able to deal with highly complex recoveries like spinal cord injuries so I found my own I went through my insurance company while I was in the hospital while still on active duty and found a facility that specialized in spinal cord injuries to what they would take me as an active duty service members for me to do out patient physical therapy because the DOD didn't have something for me available. I spent I still I still go to UPMC for outpatient rehab so I got discharged back to my home a record just outside of Pittsburgh, Pennsylvania and then I had some additional complications I end up getting re-hospitalized back down at the Tampa VA for the post-deployment rehab and evaluation program. Has anybody heard of this program? A couple hands out there so it's a relatively new program it's a great resource for active duty and veteran service members they do everything the focus is primarily on mental health and physical well-being it was I was there for about two and a half months so we're going to go down a hole a little bit and then it'll get better and positive at the end. Everything for me changed on that day absolutely everything. I never would have expected it I loved everything that I did in the Army every single second of it absolutely every single second of it I would have done anything for this country the unit the regiment hands down on any given day I would sacrifice whatever it had to be to do it and I would do it all over again every single time. Do I wish maybe I would use a smaller C4 charge? Maybe. But the reality of it is is that J changed my life forever and just like you see here is I felt like I lost my career my purpose my drive my health my mental health my mind my life and my marriage and the way that you see these things here was part of my problem is these are how I prioritize things then and sometimes even now where I put my career above the things that were way more important like my family and my marriage and all of those things I sacrificed knowingly sacrificed to support what I was doing within the Department of Defense for the US government and I wouldn't change it again but it changed my perspective a lot. One of the things that was super hard for me to deal with is the fact that I put my career above everything else my wife my mom my friends my health everything I put my career above it all and I felt like the Department of Defense and the US government and the army failed me on a lot of different things and we're going to talk about some of those things. I know that some of the people in the crowd are coming from the civilian side this is something that has to be brought up for a number of different reasons and it's not a bad thing it is what it is but we can get better. The first thing I'm going to talk about and there's a lot of medical people in this who might have the ability to influence some of these changes is I called home after I got hurt and they wanted my family to meet me in Germany my wife buys a ticket the army doesn't pay for it she buys a flight to Germany is for 24 hours later for $3,500 and we eat the bill. Award paperwork I don't give I don't care about any single medals that I've received from bronze stars to purple hearts to whatever it is my team started put me in for a valor award during the deployment for something that had happened and six months later I follow up with them after everybody comes home you know the wars won we come back and I asked myself hey man whatever happened to you know some of the paperwork and the answer was the unit had enough valor awards to get the presidential valorous unit award so they quit accepting valor awards so we never process your paperwork I said okay it's okay that's fine it is what it is expectation versus reality at Tampa VA summit Walter Reed somebody case manager comes in and says hey we have this great facility down in Tampa for spinal cord injury patients it's at this VA it's in Tampa it's great they have the best program the DOD it's their premier spinal cord injury center and he gives me a book and he shows me the rooms the rehab equipment and the program that goes into this program so I agree to go down the Tampa the brochures for a different program different rooms different equipment different program it wasn't even the same it wasn't even the same program somebody from the Department of Defense working on behalf of the federal government was giving me information about where to go to spinal cord injury rehab and had never even seen the building lack of SCI resources I brought it up as I had to find my own spinal cord injury rehab program in Pittsburgh Pennsylvania anybody from Colorado know a good hospital in Denver that's got a spinal cord injury program anybody over here at Craig Hospital one of the best spinal cord injury hospitals in the country that would have been 20 minutes from where I was working there was an option for active duty service members to go there to get treatment that nobody knew about PCS to the WTU so my unit two months after getting injured says that we don't know how to deal with you we're going to send you to the Warrior transition unit for you to continue rehab the same unit that I ended up sacrificing relationships with my family marriage health and mind two months after getting injured was trying to send me to somebody else to take care of me because they don't want to do it household goods I get discharged from the hospital all my stuff is still in Colorado the army can't figure out how to get my stuff from Colorado to Pittsburgh I had to have a group of guys that I work with rent a U-Haul and drive across the country together just to get me my own clothes the WTU prep program the prep program was a mental health and evaluation program to get guys back into the force the WTU wanted to kick me out of the army before letting me attend the prep program there was three purple heart recipients in the WTU program out of 220 people there with me and I had two of them there was one other guy that was there another SF guy that was there and they refused to take care of us and some of the other people the way that it should have been done because it was too much work for them to do it was too much facilitation nobody had ever had to go to an inpatient or an outpatient rehab program since they've been in charge there retirement and PA so I find out that the army has a cap on how many days you can be on a profile during rehab and they cut me retirement orders while I'm at home not exactly how I planned on retiring from the army was a piece of paper that says congratulations we've retired you from the U.S. military DD 214 and retirement award same thing they send me my DD 214 it's missing two deployments it's missing three schools it's missing awards my name isn't spelled right on it they still couldn't get it right and I had to pay out of out of pocket to travel to the WTU so the only reason I bring up some of these things nothing on here is super relevant to me none of none of the things that are on here are going to change my life but what it showed me was the amount of time and effort that I put into the regiment I didn't get it back and that's where we have to be better the number in the corner 69,000 or 6,919 killed in the last 40 years of war between Iraq and Afghanistan over 53,000 injured how many other people have similar stories to this that we have failed to take care of all right that's it that's all the bad stuff we're moving on to the good stuff so what do I do now is I work full-time for Carnegie Mellon doing research and development on a couple other projects and we're going to talk about some of them I also have a little consulting company I do some humanitarian work I teach SIG Sauer work for the Special Forces Foundation and fly for StapMetabac one of the projects and one of the ones that's super relevant to this conference is a project that we do with Pitt and Carnegie Mellon developing artificial intelligence and robotics for trauma care I know that one of the other presenters this morning talked about a virtual learning platform that the Army is using as a sustainment program I'm fortunate enough to be one of the subject matter experts on that program there's another spinal cord injury government funded project through John Hopkins for spinal cord injury development that I'm extremely excited to be a part of as Missy was stating is this was another opportunity that I've had post retirement from the military to continue to not only work on the civilian side but to help empower veterans that have got out transitioned out of the military and get them back into a meaningful workforce this was a really good eye-opening experience I think for a lot of people both on the civilian and the military side and it's one of the things that I continue to try to do now as a civilian is to figure out ways to bridge the gap in both employment and credentialing for specifically our special operations medics how to get them more increased educational opportunities in the U.S. and out of the U.S. I spent the probably I got to Ukraine on February 28th I spent probably 130 days there maybe working with the Ukrainian military bringing special operations medics trauma surgeons general surgeons 1 a.m. doc 2 anesthesiologists forward to help with the efforts in Ukraine during this invasion it's been a really good application of seeing how soft medics this is what they were designed to do for and it's bad that this is the only time that we can see them shine why does it take a global pandemic or another invasion of another country for us to start implementing and using some of these guys with this special skill set to the fullest of their ability the last thing that I'll mention about Ukraine is we talked about what's going on previously with the wars on terrorism in Iraq and Afghanistan and what the next war looks like there's a bunch of different thoughts around what's going on in Ukraine and if you're in this room there should be some focus on what's going on in Ukraine both on the medical and the tactical side logistics and operational and I would encourage you that if you've forgotten about the fact that there is a war going on in Ukraine with one of our highest adversaries you should open your eyes to it and try to see what's going on and how you where you're at and whatever your position is can use what's going on there to help us here both on the civilian and on the military side this hospital here that I was working with has treated 13,000 casualties in the last six months that is insane numbers insane numbers unfortunately there's a lot that we can learn from the war in Ukraine with Russia from the US and I'll leave you with this it says it on the website and somebody said it this morning the MHSRS is the Department of Defense foremost scientific meeting it provides a venue for presenting new scientific knowledge resulting from military unique research and development the MSRHRS is the premier military or civilian meeting that focuses specifically on the unique medical needs of the warfighter if you're here on behalf of a college university government organization NGO you're presenting a poster presentation or you were part of an abstract raise your hand keep raising them this is not about you put your hands down this is not about you and this is not about me this is about the flag that you wear on your shoulder the country that we live in and getting us ready to support the warfighter during the next one it's coming there's a reason that there's that much space left on that wall more names are going to go to it it's the things that we do here both private and military sector that make differences in the U.S. and abroad to save American lives on the battlefield and here in the States this isn't about your publication record it's not about your PhD thesis it's not about another article that you're going to write it's about everybody but you including me thank you all right thank you Mr. Shuley all right our next speaker is well known to many in this room Dr. John Holcomb Dr. John Holcomb received his MD from the University of Arkansas Medical School in 1985 and entered the U.S. Army that same year he spent the next decade deployed with the Joint Special Operations Command from 2002 to 2008 Colonel Holcomb served as the commander of the U.S. Army Institute of Surgical Research and trauma consultant for the Army Surgeon General over the years he has had multiple combat deployments he is the recipient of the Lifetime Achievement Award in Trauma Resuscitation Science from the American Heart Association the United States Special Operations Command Medal and the Service Award from the American College of Surgery he is a three-time recipient of the Army's greatest intervention award he has been a member of the Committee on Tactical Combat Casualty Care since 2001 Colonel Holcomb retired from active duty in 2008 in 2019 Dr. Holcomb joined the University of Alabama in Birmingham and serves as a professor of surgery he also holds the title Professor of Surgery for the Uniformed Services University of the Health Sciences Dr. Holcomb is actively involved in clinical medicine, education, research and entrepreneurship he has published over 690 peer reviewed articles it is my honor to present Dr. John Holcomb well good morning everybody it's great to be back at MHSRS previously known as ATAC which I think started in about 1994 the first time I came to that with 200 people in the audience what a great change what I'm going to talk about today is what we do matters and I think you've heard some of that before already in the previous speaker I do have some disclosures none of them have anything to do with what I'm going to talk about today next slide so I'm going to tell a little story that's what I like to do and I give talks I like to tell stories and we're going to weave together six themes all of them you've heard in the in the previous speakers this morning and I'm going to do that probably in the next 25 minutes it's not just whole blood although we're going to talk about whole blood I want you to think about your intervention infectious disease public health whatever your thing is think about how these themes also apply to what you do what we do really matters not only for the military but civilians here and around the world we're going to weave together whole blood as I said as an example first we're going to talk a little bit of history because history is really important so Sam Carmichael is a civilian certain young certain at Wake Forest wrote this paper and really describes the history of blood a chakras hesitation it didn't start last year or a couple years ago actually in 1667 as a treatment of mental health illness with predictably poor results I think it was a wife and a husband this is the real story there whole blood didn't become safe and therapeutic until the early 20th century although it was used in the Civil War in the United States with the advent of reliable equipment sterilization and blood typing all springing from our work in World War I Colonel Pete Churchill in the book Certain to Soldier speaks to whole blood as a really a lifesaving treatment that changed the course of combat casualties because at the beginning of World War II and through the majority of it actually whole blood wasn't used neither were red cells plasma dried plasma was a primary resuscitative fluid and many soldiers died pale from lack of red cells and oxygen delivery Pete Churchill changed that by implementing whole blood in the latter part of the war by the end of World War II dried plasma and whole blood had been used in over 800,000 transfusions and most importantly in the pre-hospital area to include on the beaches of Anzio as depicted in this picture Tom Shires and James Caraco real giants of resuscitation in the 60s and 70s and had funding from at that time the Department of Defense and the federal government started looking at animals and they used controlled resuscitation studies where they used a stopcock to bleed the animals off and on and what they found was by giving crystalloid and a little bit of shed blood that those animals did really well this paper was presented in 1963 in Vietnam because at the beginning of Vietnam whole blood was the primary resuscitation fluid as it was in Korea but during that time because of this work that was done with the controlled resuscitation animal models they changed the crystalloid and components and then this thing called Da Nang Luan started showing up otherwise known as ARDS not previously well described in the resuscitation of combat casualties in World War II or Korea and then you kind of move forward from the mid 70s 20 years and Art Fleming a retired colonel surgeon and along with his folks in Los Angeles Bill Shoemaker were giving 10 to 13 liters of crystalloid instead of the four to five that Shires talked about now they really used excessive crystalloid and were causing substantial complications next slide so Blau now moved forward another decade describes this iatrogenic resuscitation injury were giving excessive crystalloid not what was described by Caraco and his and his partners but now excessive crystalloid up to 20 liters of crystalloid this is where I became into place and general place as well we did this because we were taught to do this and we did it over and over and over again thinking we were helping our patients Zolt published this paper in 2003 showing that this excessive crystalloid cause intraabdominal hypertension abdominal compartment syndrome multi organ failure and death it was an iatrogenic injury from excessive crystalloid next slide so the animal research using controlled hemorrhage models led to many many problems and the problem really as you kind of got went back and unfolded this was the animal models utilize were not the right animal models hindsight course being 2020 next slide so the evolution of resuscitation if you look at this history of resuscitation and this is a U.S. centric slide 1864 is the first documented use of whole blood that I can find done by a Southern surgeon who by the way became chairman of surgery and dean at the University of Arkansas where I went to school didn't know that at the time and anyway whole blood and blood products were really used as the primary resuscitative fluid for 126 of the last 158 years and as you can see in this timeline the Vietnam era in early 2000s we really got confused and gave a lot of crystalloid not many red cells very little plasma and almost no platelets that has come back we've reversed that and it comes from our experience in the war as I'll talk about in just a minute where now we're back to dried plasma and whole blood again pretty interesting back to what was done in world war two next slide so history history is important so whatever it is you are researching right know that history know the history it's extraordinarily important we've all heard the phrase history repeats itself so those who don't know history are doomed to repeat it well that's what happened in resuscitation stories here we must know the evolution of the issue know why it happened know the methods read those methods of those papers and understand the science okay next theme here is experience next slide so in 1993 as a young surgeon right out of training and Dr. Rausch showed a slide from the Black Hawk Down movie which is pretty close to what happened on the battlefield I found myself as a young surgeon taking care of combat casualties clearly for the first time I graduated surgery Red and C-91 and in the middle of this mass cow Colonel Denver Perkins who had a vast experience in Vietnam as an infantry soldier and was the Colonel who was the head of anesthesia in our combat sport hospital in the midst of the mass cow we're running out of blood products patients are still coming in he said why don't we do a whole blood drive well I had never heard of whole blood I'd never heard of a whole blood nobody ever said a word about whole blood I thought we were going to go to jail if we lived through this episode truly but we were running out of blood products Denver was the Colonel I was a young major we gave units of whole blood that were warm from the donor you did need to put it through a blood warmer to hang them on the patients and the coagulopathic bleeding stopped it was an amazing experience at this at this not this venue but in the ATAC meeting in 1994 I described it as a religious experience as a clinician probably not the right way to describe something in an academic forum so you then move forward another decade or so and we started writing about these things from the previous war and the current war writing about damage control resuscitation Phil Spinella really grabbed ahold of this out of the 31st cash database and started writing about risk associated with fresh whole blood compared to red blood cell transfusion and they evened themselves out and then warm fresh whole blood is was independently associated with improved survival in patients with combat injuries really documenting now this experience the growing experience in the war next slide and then move forward another decade to tactical combat casual care and Frank Butler writes for tactical combat casual care for pre-hospital care in 2014 whole blood is the primary recommended fluid a pretty bold statement at the time in 2014 but one that I think has been proven true Jeremy Cannon who's in the audience wrote a really influential paper from the Eastern Association for the Surgery of Trauma guidelines talking about balance resuscitation think back to that timeline before when we gave lots of chrysalloid very little red cells plasma platelets and now are one of our major trauma associations with Jeremy Cannon's first author writing about balance resuscitation is the way to go and then the joint trauma system CPG's clinical practice guidelines and we'll talk about the joint trauma system here in just a second but the damage control resuscitation whole blood or blood product resuscitation was the first guideline written in 2004 has been continuously updated as recently as 2019 is really helping to drive the adoption of these practices next slide so experience on the battlefield no substitute for deployment experience now when you don't have a war it's hard to deploy to war right but there are wars going on all the time and our military healthcare system deploys all over the place all the time leaders must deploy deployed patient care is critical it's different than garrison care I've done lots of garrison care continue to do it today I work in a large university setting the care and the way we do it on the battlefield for our injured soldiers sailors airmen Marines is different than taking care of the garrison care back here in the rear we learn the problems of the battlefield we learn the potential solutions and then we bring those back and iteratively work on them over and over and over again share them with our colleagues who don't deploy and then we deploy again and again and again this is what makes us different this is what makes us different we have to deploy leaders must do it it makes military medicine different from garrison care and everything else that's done in medicine research so general peak at that time a certain general asked me to assume command of the institute of circle research in 2002 and I had the the great honor of helping to lead that organization for six years we did hundreds of animal studies literally hundreds we had animals going up and down all day long all different kind of animal studies going on we did very little randomized human research when people when you look back at your jobs and I've done this in the air that I made as commander of the ISR was not pushing more clinical research high quality clinical research I wish I'd done that we did a lot of work Jill Sandin led a lot of these efforts with Mike Dubic Charlie Wade a whole group of really talented researchers and what I would say that was really fun there was combining clinicians with basic science translational science research experts and forming these multidisciplinary teams as Dr. Woodson said earlier that really brings together everybody's expertise to solve problems Jill wrote about blood pressure which we're bleeding occurs and found the same thing in the animals that was found in World War I and World War II and then went on to talk about the uncontrolled hemorrhage models are different they're fundamentally different than controlled hemorrhage models and published that in shock going back to those Caraco and Shire studies and then as you progress on through the through that decade she wrote about using different combinations of blood products compared to chrysalloid and which one do you think was better was whole blood when you did it in a prospective randomized animal study whole blood was a superior product next slide we deployed research teams out of the institute of surgical research in combination with the joint trauma system we established Laura Broch is in the audience this was really the driving force establishing for the first time a human research program protection program in a combatant command this paper documents how to do that that that human HRP is no longer in place as I understand but this paper describes exactly how to do that and then Jeremy Perkins who's not here because he's working back at Walter Reed talked about these research teams and what they did how to form them how to deploy them how many deployed and the lessons learned from a research point of view on the battlefield extremely important that we continue to put research teams on the battlefield so why do these why is research so important this is a little graphic that from the Institute of Medicine in 2013 and while the graphic doesn't speak to trauma systems I added that tile at the top and the real point of this despite all the words is in a learning health care system research influences practice and practice influences research they go hand in hand you cannot have a high quality learning system without a very active robust research program that informs practice and vice versa now coming back to trauma and this is a slide I've used many times if you're going to improve outcomes in a trauma system you have to work across the entire continuum of care both from simulation injury prevention pre-hospital pain control all the way through the things that we often talk about stopping bleeding transfusion to rehab and outcomes that entire continuum of care needs to be optimized in every one of those areas and then the the sum of those individual parts actually is much greater than the individual parts next slide Peter Rhee wrote a paper entitled increasing trauma deaths in the United States and published in Annals of Surgery in 2014 but in that paper talks about the funding so they do have research and have a robust research program and everybody on the front row knows this you have to have a funding at the federal level trauma funding is woefully unappreciated and not supported they you have cancer and heart disease HIV and trauma is obviously the small little thing down there in the bottom right of your screen but it if you look at if you look at the societal impact the trauma funding is nowhere in relation to the societal impact the trauma is the leading cause of death of 1 to 44 in the United States most of the people in this room are in the 1 to 44 range right the old folks in the room were in the cancer and heart disease range and I always say now who makes all the decisions well it's probably the old folks that's where that's why they're funding cancer and heart disease but for the young folks and for the guys who go down range they would look and then go wait a minute I'm not going to get cancer and heart disease down range I might get injured right let's fund the injury it kind of makes sense what do most funders fund this is a you know when I went to the university once I got out of the army I started looking around and and and running a research organization multidisciplinary research organization and a and a and a division and a really busy academic center most funders fund preclinical research basic science and preclinical research so what are research labs do they do preclinical and basic science research very little clinical what should we fund what should we fund I think we need to reverse those circles and fund more clinical work and less preclinical work that is not disparaging preclinical work it's just we've got to have a balance between those two and in my opinion they're not balanced so in the in the bleeding world we talk about the bloody vicious cycle which is quaggulopathy and bleeding and all that kind of thing but this is a research side and it's data right I love data by the way if you haven't figured that out I'm like this data geek guy as a trauma surgeon so consistent federal funding is not commissurate with societal impact it's not it's been well published and well documented so therefore few investigators go into real injury research because it's like that bank and money thing the money's not there because the investigators aren't interested in injury because there's no funding you lack high quality data low quality data gets published in low impact journals and that in my opinion drives tradition driven suboptimal clinical care once we have if we have high quality papers people kind of follow that they really do the clinicians try to follow high quality data but it's just it's really a sad vicious cycle in my opinion next slide now there are hundreds of non-randomized whole blood papers right now when you go to folks who are not believers who haven't had their religious experience they go well this paper it's retrospective it's single center it's this and that there are two randomized studies so far both funded actually by the DOD both pilot studies both small single centers one by Brian Cotton in 2013 and then a decade later the folks out of Pittsburgh Frank Gayet published their pre-hospital whole blood study now there are two large definitive randomized whole blood trials both DOD and NIH funded that have been funded the paperwork is getting sorted out they're gonna they're gonna start any day now the troop study PI Yon Janssen in Alabama 1,100 patients at 12 centers is an in-hospital study and the lights tower study PI Jason Sperry almost 1,100 patients at 10 centers is a pre-hospital study both have mortality as a primary endpoint so in another couple years these two large funded prospective randomized studies will come to fruition and will have data in both places so the DOD in my opinion should fund more clinical research we need to deploy research teams to the battlefield under the auspices of the joint trauma system every trauma system must have a research arm must have research capability and that's where it should come out of we need to sustain these multi-year programs of clinical research like lights just keep them going do one study right after the other the infrastructure required to put these programs together is substantial and needs to be sustained and we just need to do this clinical research that translates to the battlefield and the civilian area it makes finding sticky especially during the interwar period we do these large clinical studies we find out something that works you implement them in the civilian world and in our MTFs and then the results are sticky when we go back to war next slide dissemination implementation and policy some of my friends on the front row are going Holcomb is talking about policy how about that the joint trauma system first described in its nascent form in 2003 has three components on their website I'm suggesting as I said already research should be the fourth component and supports three committees that really cover the majority of what goes on the battlefield from pre-hospital care in route care and then in the operating rooms as well the whole blood as I said before was the first clinical practice guideline it's been continuously revived revised and improved based upon data from the war both for the performance improvement efforts from the registry and from research efforts describing an improved outcomes and just to reiterate in 2014 actually what is it eight years ago pre-hospital blood is recommended as a primary fluid of choice the number one fluid a lot of logistics tail to that a lot of catching up had to be done next slide so military and civilian adoption which is critically important to making changes that happen in the battlefield stick in the civilian world so that as we do our mill sieve teams and all of our military reserve and civilian folks learn how to utilize these innovations that from a whole blood point of view Colonel Taylor wrote and described the blood program from 2014 and to 2021 and during this paper describes the change where whole blood was almost universally walking blood bank whole blood to low tighter whole blood tested and approved FDA cleared et cetera really an important and huge transformation in the use of whole blood on the battlefield one of the questions that is interesting and we're trying to and others are trying to tease out the data is low tighter whole blood as efficacious from an outcome point of view is the walking blood bank whole blood it may not be it might it's a research question and then the folks in san Antonio and I know there's a number of in the room have really described their implementation of whole blood in the civilian setting they've done a better job than any place that I know of with both pre hospital in the hospital multiple hospitals multiple trauma centers reacting to everyday trauma and also mass casualties a really lovely implementation of whole blood across their entire trauma system next slide colonel gurney is in the audience writes about whole blood at the tip of the spear documenting the experience on the battlefield and the improved outcomes and then Stacy shock for colonel shock words in the audience as well writes this really kind of almost a policy statement if you're not policy but a statement consensus statement from the joint trauma system the defense committee on trauma and the armed services blowed program on whole blood pretty important statement from these three very influential groups and then Phil Spinella and Andy Cap both in the audience right on this paper published now six years ago six or seven different ways whole blood is better than components extremely logical paper and one that I really really like and and hand out to everybody really an excellent summary of the benefits going all the way from economics logistics to a course clinical outcomes and then that policy statement you know coming in 2022 the dodi talking about it's not exclusively on whole blood it's on the armed services blow blood program but also speaking to whole blood walking blood bank and low tide or all whole blood on the civilian side right which is where a lot of care happens a lot of our military folks train on the civilian side certainly the reservists work in civilian hospitals this is data from 2020 so a little dated published last year in transfusion 24.5 percent 25 percent of civilian level one and two trauma centers currently use whole blood so that's two year old data it's gotta be 35 percent now so I would just say around a third of U.S. hospitals likely level one and two trauma centers a third are using whole blood in the hospital in the ED and their ORs but by combining NemSys data which is the pre-hospital data not linked to in-hospital data by the way right so there's no linkage of pre-hospital and hospital data yet only 0.5 percent of eligible pre-hospital patients received any blood product so the standard of care from 2014 in the military recommendation is almost never used in the civilian population from data from two years ago next slide it is the standard care as I said in deployed setting it's not university done but it is the standard that we teach to rarely done in civilian pre-hospital setting although there are absolutely exceptions to that I think the reason there's not more pre-hospital blood is because the pre-hospital providers can't bill for it therefore they can't use it because they can't put something on the rigs both air and ground that they can't bill for because they can't lose money that to me needs a change I don't know if that's a law or regulatory change but just think about this there are actually prospective randomized data not shown in this talk that pre-hospital blood decreases death by 10 percent both military and civilian yet because the civilian medics can't bill those systems can't bill for it you can't have pre-hospital blood in hardly any place that needs a change next slide so dissemination implementation policy I think that when you look at the CPGs on the JTS website they're clearly focused on the deployed setting and yet a lot of them apply into our MTFs it's my opinion they should be implemented as they can be at our MTF so that we train all of those people on all of those concepts all the time this should be taught at all professional military courses TC3 should be a requirement for all military personnel not just medical and the policy shouldn't take 20 years to promulgate on this on this product training next slide so these Milsiv partnerships I didn't put the another slide up and because I have to compress this a little bit but Milsiv partnerships have been described since World War I it's actually books written about it during that era going back to that history lesson earlier Marty Schreiber who's in the audience wrote a paper in 2002 from our experience at Bentob talking about military trauma team training you know in a civilian institution and then Colonel Gurney again writes about team training in this year single surgeon teams multi-disciplinary rotated assigned all these issues have been going around but the idea is that we want our teams before they deploy to have taken care of trauma patients before they go down range and you just asked the question are the selected Milsiv sites using whole blood are they using whole blood they're going to see it down range both pre-hospital and the hospital are they using it the Milsiv sites next slide so I tried to generate some data 9 of 52 of our military MTFs have whole blood they're listed here 9 of 52 in the Milsiv hospitals there's 87 Milsiv kind of team sites that the number is a little bit flexible how many have whole blood I couldn't get that answer it's not really hard it's kind of hard to get that number at level one and two trauma centers civilian centers this is not Milsiv or military 25% or 123 two years ago used whole blood and then in our in our military health care system back here in garrison where is whole blood pre-hospital where is it right remember I saw whole blood for the first time in 1993 in Somalia and I had never heard of it I thought we were going to go to jail I had no idea how it was going to work there was nothing I'd read about it before we shouldn't repeat that again for the for the guys and gals that are going down range both pre-hospital and hospital where do our teams learn about whole blood right I just kind of made that point I think we shouldn't see it for the first time on the battlefield next slide training we need to train as close as possible how we fight currency in all aspects of trauma cares critical to access on the battlefield systems registry PI research the entire team medics to administrators can't see whole blood for the first time in combat next slide worldwide impact next so whole blood is now included on the research agenda for trauma critical care this is an international publication that came out a couple years ago people around the world are talking about whole blood now and how to implement it and how to do it next slide in the UK I've already mentioned the two previous US studies the UK is doing a similar study of 848 patients 10 sides primary endpoint 24 hour mortality with pre-hot with whole blood and starting in the pre-hospital arena next slide in the French are doing a similar study albeit small or 164 patients whose primary endpoint is coagulopathy next slide so much like the previous speakers said Ukraine is a pretty interesting place to go to my first trip in April of this year talked about whole blood as you might expect talked about the benefits of whole blood whole blood was being used in the crane but a little bit under the under the radar if you will because the federal authority had not authorized the use of whole blood next slide two weeks after working with some folks in western Ukraine and their hospital that was receiving casualties not right off the battlefield but days later whole blood was new and it was in the refrigerator in the emergency department next slide and then two months after this through a lot of work by a lot of people the Ukraine Defense Minister of Health decreed at the Minister of Health in their civilian hospitals that whole blood could be used pre-hospital in the hospital and essentially took all of the JTS guidelines and implemented them TXA one to one to one walking blood bank when you read this document next slide so there is worldwide impact people are using whole blood around the world and many centers the United States 50 percent of the whole blood is for non-trauma patients actually maternal fetal bleeding postpartum bleeding and it just just proves this thing that we know over and over and over again that innovations from the war move into the civilian space and become a standard care on PubMed if you put whole blood and military from 1994 to 2000 you can see this exponential increase in papers published about whole blood next slide so in summary whole blood was used for the first time that we know of in 1993 for a U.S. combat casualties the DOD and NOH funded a program of blood research and has changed clinical practice really around the world how will we ensure our clinical teams utilize this practice before deploying they have to use it here in the rear it's really hard to see something for the first time downrange and what we do matters the research the funding policy clinically and keep that laser focus on improving outcomes on the battlefield next slide now this is Don Quixote right was one of my favorite books ever and Don Quixote kind of challenged authority and I say this will all do respect to the general officer sitting there right here in front of me but challenging dogma and authority respectfully and with data is the way to go is admittedly upsetting to the status quo but it's the only way progress is made especially during the interwar period we got to be ready for the next war next slide so it's not just whole blood this is an example of many many other things it's just a vehicle described that what we do matters not only for the military and civilians but here and around the world you all should be very proud of what you do you really should you know pat each other on the back pat each yourself on the back a little bit there's a lot of people that are trying to bring you down but a lot of what you all do is so extraordinarily important and it makes a difference on the battlefield next slide thank you very much all right thank you Dr. Holcomb our next two distinguished speakers are from the robotics institute Carnegie Mellon University Pittsburgh Pennsylvania Dr. Arturo Dubroski is the alumni research professor of computer science at Carnegie Mellon University where he directs the autism lab one of the largest academic research groups focusing on artificial intelligence and its real-world applications he has been pushing the boundaries of science and transitioning results of his research to industry and government practice for more than three decades in addition Dr. Dubroski has led development of machine learning solutions to extract clinically useful information from streams of complex biosignals and applying the resulting tools in clinical and field case scenarios Dr. Howie Chassette is a professor of robotics at Carnegie Mellon University where he serves as the co-director of the biorebotics lab his research program has made contributions to challenging and strategically significant problems in diverse areas such as surgery manufacturing infrastructure inspection and search and rescue he leads multiple PI projects centered on manufacturing and co-led the formation of the advanced robotics for manufacturing institute a 250 million dollar national institute advancing both technology development and education for robotics and manufacturing recently Dr. Chassette's surgical snake robot cleared the FDA and has been in use in the United States in Europe Dr. Chassette received his undergraduate degrees in computer science and business from the University of Pennsylvania and his masters and PhD from Caltech in 2002 the MIT technology review elected Dr. Chassette as one of its top 100 innovators in the world under 35 in 2014 popular science selected Dr. Chassette's medical robotics work as the best of what's new in healthcare Dr. Dubrowski and Dr. Chassette the floor is yours okay thank you for the introduction I'm how we chose that this is Arthur Dubrowski and we're going to tag team our conversation today so I want to thank Jose Salinas Dr. Salinas for inviting Arthur and me to this conference and for the opportunity to speak here without him you know we wouldn't be here right now so so thank you but before we go on I also want to express some gratitude I want to express some gratitude for those of you who sacrificed and serve our country you know Jose asked us to speak about you know the future of robotics we're in our Ivy Tower isolated from the real world and you're out there you know serving our country so we can do our jobs so please thank you I also want to express some other thought about Luke so Luke works with us and when the Taliban went into Afghanistan Luke I don't think he even blinked his eyes before he said he was going to go and try and help his friends and it's amazing how he just continues to want to serve us despite all the things that happened he went to the Ukraine to serve as a medic again to support you know our freedom without even hesitating once without a bitter thought from the past so again I want to express at least our gratitude for that okay so let's go to the next slide so what what Dr. Salinas asked us to do was sort of dream a little bit about what's the future and I'm not sure if we're really qualified to say what would be next but at least I can tell you our aspirations and what we want to do is we want to bring medical care everywhere at any time and can be performed by anyone that is the long-term goal we want to democratize medical care so expert expert care can be delivered anywhere without a moment's notice now next slide now I think we're seeing this already with the advent of minimally invasive surgery I think we all understand that no one wants surgery but if you have to have it you're better off having the thing on the right than the thing on the left next slide so the benefits of minimally invasive surgery it's less pain you also have reduced costs and finally you can bring medical care anywhere we're hoping to not only bring it out of centers of of medical excellence like the university at pittsburgh medical center but we also want to be able to bring it into the office into the field can we do an appendectomy in the field with one of our field technicians instead of having to airlift an injured warfighter to say India or Germany in order to receive the proper care that he needs next slide so with with minimally invasive surgery if you sort of look at how deep you can go into the body and how the kinds of spaces that you may want to see next slide this is where all the current procedures are and you'll see this this next slide there's this open for going deep into the body and be able to navigate the intercavity spaces there isn't technology that does that yet and next slide and that's where a lot of research and development in the academic world has been centered and we're starting to see some of this with some medical device companies as well like intuitive surgical for example so it's that cavity that's where the opportunity is but the next slide but if you look at conventional minimally invasive surgical devices today you're limited to rigid okay you're rigid you're limited to rigid laparoscopes that can only enter a line of sight or you have buckle you have flexible endoscopes which can travel the luminal spaces but they buckle easily what you want is a it's almost like a snake or a steerable laparoscope a surgical snake robot like you see in the right hand side that can go through the intercavity spaces as easily as driving a colonoscope and that's where some of our work is going so if you go to the next slide so what you see here is the first in human operation with this surgical snake robot we're doing an epicardial mapping and then an ablation of the 75 kilogram woman and what you see on the right is the live fluoroscopy of the snake robot going around her heart and then the upper left you see the visualization from the catheter the visualization catheter seeing what the physician hit hit the space bar again one more time what you're seeing is not a surgeon performing this procedure you're seeing a specialist this is someone who didn't have the same level of training that that a surgeon has so we're starting to see this progression from surgeon to non-surgeon and when we do this we're eventually going to have tools that will have perhaps non-doctors non-medical field techs be able to perform these kinds of procedures so next slide so the oh let's go to the next slide let's just skip that we can skip that clear the FDA good next slide so where I'm excited about is this fad that came about about 10 years ago and it's since died off but I want to bring it back called natural orifice transluminal endoscopic surgery and the idea is you enter through a natural orifice reach the luminal space and then go after the target that you want to fix like the pancreas for example this to head of the pancreas you could you know what's the phrase you know God created the pancreas so surgeons can send their kids to college hit now what we want to be able to do is reach this in a minimally invasive way and the person can then go home and have spicy food the next day perhaps but the reason why this brings us out to the field is now you don't need a sterile operating field anymore you can start to do procedures in a more I don't want to say dirty but less clean environment and still have the same efficacy and at the same time hurt the patients a lot less so we actually started this next slide so I know what you're thinking when you woke up this morning you weren't going to see two snake robots shoved up a pig's ass I get that a lot but what you're seeing here is two of our snake robots going in and removing this pig's pancreas and it's doing it in a way where the pig only had this big this much of an incision inside itself and inside its large intestine next slide and that's a close up and these are the benefits or the research that's excuse me these are the areas that we're currently doing our research on next slide so the next thing I want to do in order to have portability is talk about ultrasound so I think we all agree that no one likes looking at ultrasound images unless it's a baby and it's yours okay otherwise when you're looking at ultrasound images it's just impossible to discern it's hard so Arthur and I along with our colleague John Galliotti what we're doing is we're creating new techniques to enable surgeons non-surgeons field medics to be able to inspect and see what's going on just using ultrasound information and if we can do that think of the benefit it's portable it's low cost and also it doesn't affect the patient that much and this work is part of our project next slide that we're doing with the University of Pittsburgh Ron Porpartich whose name I never pronounced correctly he's the leader of this project and we've been very fortunate to have him bring the University of Pittsburgh doctors together with the Carnegie Mellon robotics researchers and we're creating these tools that will allow someone like me or maybe slightly more competent to be able to insert lines saying to your femoral vein in the field using ultrasound to guide the search and there's a lot of details here I'm happy to talk about but what we're excited about is being able to deliver that kind of care out where people you know shouldn't be getting that kind of care next slide so I want to change subjects again and talk a little bit more about one of the another ways we can get accessibility so during the great ventilator shortage rumor of the pandemic we at Carnegie Mellon designed and built a new low cost ventilator using parts that are readily sourced that would be about the same functionality sorry 80 percent of the functionality of a hospital ventilator but the cost and convenience and portability of those manual ventilators that Luke was talking about the ones you have to squeeze so we were happy about this ventilator but the part that that the doctors really liked I thought it would be pretty cool having a low cost ventilator because then people can go home and be able to get treatment without going to the hospital but the part they liked was go to the next slide was that this ventilator talked to the cloud very easily so now you can actually control, monitor and perhaps give advice anywhere in the world to any of these ventilators and what's exciting here is not only can we have this portability this ability to disseminate the care but we can now start tracking the data we can now start figuring out are there a phenomenon occurring en masse and be able to make better predictions as to what's where the disease may be going where it's going next now I realize we have Netsyn that network protocol the part that we're interested in Arthur and I in our research is what is the information that you're going to put on that protocol that would be relevant for the doctors to talk to process I think how am I doing for time I forgot the time okay so I have one more thing I want to present let's go to the next slide and this is what we call expedition robotics and again our goal is to bring medical care out as quickly as possible anywhere and what we saw so Luke one of the things he did when he volunteered during the pandemic was he went to Columbia University in New York and he helped helped with the with the doctors there and he actually took a video of them setting up a field hospital on the on the soccer field at Columbia University it took three days to set up that field hospital and it was three days of really strong and smart people trying to figure out where to put what where as quickly as possible what we want to do is we want to create tools to help automate that process so we're re-envisioning how to deploy these kinds of field hospitals next slide so the reason why oh that's the Columbia slide I'd forgotten about I already said this stuff so just that's the picture from Columbia next slide so in our vision is we have UAVs go out survey an area next slide we then have containers driven by trucks to the remote location these containers are specially designed for medical use they're deployed next slide and then they're assembled into a hospital now they could be assembled using forklifts that people drive or we can have additional autonomous robots going around helping people helping the people as well assemble these structures next slide and here's some of our concepts as to what these containers should look like in a sense what we want to be able to do is build a hospital as if we're putting Lego together using as few tools as possible and that's important because when we go and deploy these hospitals it's not like if you forget something you can go to Home Depot and pick it up we may be in austere environments and again time is of the essence next slide and this is another picture of what we had in mind next slide so these are the ideas at least that we're starting to think about on the robotic side of things and you know robotics that the actual if you want to think of Archer and me as I'm the brawn and he's the brains if you want to think of that way so go to the next slide and that's where the artificial intelligence more comes in I spoke with the robotics and Archer will take over and talk about the AI aspect of our work please you know how some physicians maybe in some specialties such as psychology or maybe psychiatry choose to study that field because they may be lacking something that's also my mode of operation I work on artificial intelligence guess why yeah so thank you very much I just missed the beginning of the presentation by Luke Shulley and I don't know if he mentioned that he's role in special operations was of combat medic and you heard that we work on automating Luke Shulley and now you know why because he's not in service anymore so we need to replace him we need to scale his capabilities as well as many of his peers so that we can treat more people in need not just in military but also in civilian situations involving traffic accidents and such so this is an important technology that requires not only strong clinical and field care expertise but also robotics and luckily for me artificial intelligence so that it all can work together one more thing before we switch to the next slide is I may look young but I'm not I started working in the field of AI almost to the day 32 years ago and by the time when I was joining this field as a young researcher I thought that I may be late to the chase because I watched all the science fiction movies seeing robots roaming around and doing wonderful things and smart systems making decisions for humans and whatnot luckily I was wrong fast forward 32 years and where we are my goal just like Howie's is to make sure that everyone uses my tools everywhere at all times you can check what is the penetration in healthcare of AI solutions everybody's talking about it but how many successful deployments did we see so in my part of the talk I'll try to expose a few limits few roadblocks that prevent us from achieving that final goal and I hint you on some ways that we on our side think about overcoming them of course the list of these roadblocks will be incomplete next slide please so we have actually a few examples of great potential of applying artificial intelligence to various aspects of healthcare the first one is actually I'm gonna only focus on things that I participated in so that I don't mislead people when I explain them is an example of a discovery made with AI we applied AI to a bunch of data collected at the bedside settings in intensive care with the idea of predict episodes of a cardiorespiratory insufficiency so all sorts of waveform vital signs were involved and such and we build those predictors and then we discovered that these predictors do not seem to follow the principle of one size fits all so we suspected some heterogeneity of the patient population and indeed what this graph shows is our discovery horizontal axis in this graph is time not yet is time vertical axis is a relative risk for developing cardiorespiratory insufficiency CRI episode you see that some of the patients follow the yellow line that they are always at high risk they are very unlikely to be messed by healthcare professionals who are watching them in ICUs so they are not really in trouble of being messed but about two thirds of these patients green blue and red lines follow very different trajectory they escalate rather rapidly only about a half an hour before the episode be before the onset so we were worried about those so that's interesting discovery we didn't know that this heterogeneous structure existed unless until we applied the AI the next one very quickly is an example of already fairly mature project that looks at emerging epidemies of hospital acquired infections in hospitals in facilities but they should apply almost without any changes those solutions to navy ships military installations and such apparently hospital acquired infections are responsible for more deaths in the US every year than prostate and breast cancers combined I didn't know the magnitude of this problem so we have the system that uses AI whole genome sequencing electronic health records not only detect the emerging outbreaks very early but also to probabilistically assess the most likely pathways of transmission of the bug between individuals so this gives infection control teams strong weapon to counter those events next one and this is perhaps the most related to the theme of this conference and the details of this work will be presented in more detail tomorrow in the morning oral session this is about monitoring sufficiency of resuscitation using only non-invasive vital signs so with the eye to deploy it in the field medicine the idea here is to not only discover where the patient should be put under aggressive fluid resuscitation but when to stop that aggressive protocol because we don't know we don't want to overshoot next slide so I come from Carnegie Mellon University we think that we know a little bit about artificial intelligence and at some point somebody decided that we need to put this knowledge into some sort of systematic view and this is the diagram that you're going to see when you come to visit us and you speak with anyone about anything related to AI this is a must this is called AI stack of capabilities and it's not very revealing except that it accumulates all the layers of different components that need to be present and implemented and made to work together for anyone who wants their AI systems to be successful in practice so you cannot ignore any of those layers next slide please coincidentally when you think about applying AI in practice but any other technology in practice you need to think about also project maturity project cycles so we have data capturing we have data assembly and curation we have data making data ready for AI these are the necessary steps without completing them we cannot expect that our modeling endeavors will be successful so if you are from academia and you work in the area of machine learning or AI you typically focus on the green little arrow modeling and summarization of data because this is where most of research fund is very few of us academicians venture to the left or to the right of that green arrow in the bottom to the right is very important if we develop a solution that works and we can prove it in the lab then it would be natural to move ahead and deploy it but there are so many challenges there most of the applications of AI that make any sense to humans are important and usually those fields of application have already developed culture standard solutions and the whole processes around and healthcare is one of those examples we have pretty strong rules of practice of care and other things that are already there and now an AI guy shows up with this black box solution and tells you the clinicians to please start using this instead of everything else that you've been using so far how is that going to work? you're going to be very skeptical about all this so that's one of the issues that slow us down in this idealistically speaking massive deployment of this good technology to practice and then integration how to integrate those things so that they don't disrupt good existing processes but instead enhance them it's not a trivial task so if you want to look at this comprehensively you have these two dimensions the different capabilities and the different stages of maturity of your projects and next slide that creates a grid the grid of capabilities and next slide is very busy I'm not going to read through all these boxes obviously but basically you can think of them as a mix of existing capabilities and capability gaps there are plenty gaps in this picture and the AI research community is working hard on filling them in and I'm going to now talk about the key limitations that are selected or extracted from this complicated grid only a few of them to focus our attention next slide please so what are the key limitations we are going to talk next but where they are coming from they're usually caused by data and by the models so you can if you see that AI doesn't work you can blame either or both typically next slide so one big challenge is that our current AI models they are very hungry for training data they typically want to be fed with lots of examples of healthcare records that reflect health and other examples that reflect illness so we need to collect this data to train those models of course this process is not cheap it's not easy it's mundane it's cumbersome and what's worse we need to put many of you doctors and nurses on the rear ends and ask them to sit down and point by point label big amounts of data to train those data hungry models so this is one of the aspects of this whole process that prevent us from achieving what we want everywhere at all times everywhere everyone right so we have some tricks in our sleeves one of them goes under the code name of semi-supervised learning it's nice because you only need some training data a little bit and then hopefully you have a lot of data that is not actually labeled by any human expert sitting around electronic health records bedside monitoring data are examples of this situations so if you have some initial amount of data you see a plot of this data a scatter plot of this data on some projection of the future space you may see something like this red unhealthy people blue dots healthy people next slide and so if you wanted to apply common sense you may see an opportunity to draw a line that separates these two groups of data and you would achieve perfect classification and this line may look like this so if all that you had was this data maybe that was a good solution but then next slide we may see that there is a bunch of unlabeled data that comes together with some of this labeled data and when you see this distribution you may actually think that we should probably change our opinion of how this line should be plotted next slide and it may look like this right so this is the essence of one of the tricks in our sleep semi-supervised learning which can reduce the need for manually annotated data next slide and how we do it is probably less important next slide but what's important is this approach does not require any additional effort from human data annotators as soon as you have this initial sample of labeled data so that's very cheap in terms of human effort next slide still on the same challenge another idea that has been popular for the past couple of decades is called active learning here the system is introverted and introspects its own inabilities or lack of confidence and peaks the unlabeled data points very specifically to address those deficiency so it's very selective about how much and what exactly data should be labeled by humans the system itself becomes proactive in choosing what the human should provide advice on and so we can expect the system to pick the most controversial cases or the most uncertain cases and then human with their expertise can help resolve that another radical solution to the lack of labeling resources and the hunger for lots of training data in AI systems is currently popular it goes under the code name of weak supervision and this is instead of going to to our healthcare partners and asking them to go point by point through the data we ask them if you were to perform that task of labeling data what kind of rules would you use to decide whether this case is healthy and this case is unhealthy how would you approach this and we take notes we collect those rules we put them together in a consistent probabilistic system and then we apply the sets of rules that they provide that we harvest from the experts and automatically label perhaps very large amounts of data now this is very cheap but it's also interesting and exposes other opportunities that I'm going to mention in a moment but I mentioned just three tricks there are many more there is self-supervision there is n or a k-shot learning or one-shot learning there are also approaches that look at generating artificial training data by data augmentation when you get your actual examples from the field and then you modify them slightly by noisifying or rotating images and such to artificially inflate the size of the training data sets so they are all good and the number of different approaches and the amount of work being done in AI research community is indicative that the community has recognized this as a problem and is working hard to resolve it so that we cannot be limited as much as we are right now by the lack of training data next slide another challenge that pops up is but can we trust the training data that we collected in the field from experts we assume it's grand truth and we often just go forward with our projects without thinking much about it so one example to just give you a hint about this imagine you were tasked with a somewhat strange task of assessing biological age of a person based on their facial image when you look at this picture of this gentleman if you are asked to exactly pinpoint their age with a resolution of one year that's going to be a difficult task and maybe you won't feel very confident giving that the answer to this question because at certain range of ages it's kind of hard to pinpoint this so the result will be frustration from the data not data and also the quality of the resulting grand truth data will be limited it will be carrying lots of noise and therefore the downstream models that you want to build from that data will be subjected to the same kind of challenges of noise however if we reformulate this question next slide and instead of showing just one picture show pictures of two people and ask who looks older that question is actually much more easy to answer confidently and faster and with less frustration and also introduces less noise to the training data so this is just one trick on how we can increase the quality of the labels next slide another big challenge is so okay we train these models we want to hand over these models for you to use in practice do you trust them so one example is a black box decision support system that hypothetically advises the surgeon to amputate the right leg of a patient any person would come on sense would question that recommendation and ask why with some of our techniques that we use these days for AI we can actually somewhat answer that question but with the mainstream that's not an easy task for them they have hard time answering questions of why even more difficult question for some of those systems is how confident you are about this recommendation in most cases the best we can do these days is to answer this probabilistically say we are 95% confident a really quizzical doctor and nurse would then ask how about the remaining 5% and because our field is so dependent on statistical machine learning these days we honestly don't have a good answer other than statistical if you look at the roots of AI on the hand it was all about mathematical logic it was not about statistics we use statistics later on when we realize that this mathematical logic makes it difficult to use in applications of our solutions so we kind of switched to statistics for confidence for convenience now you are facing this challenge so now some of us are bringing back the logic to the front and allow us to apply this this is very important because our systems tend to make very big errors in response to tiny changes of the input features they sometimes defy common sense once you see the system solve a very difficult for human problem very well and the next minute you expose it to a problem that perhaps human being would consider trivial and they will fail miserably I don't like that we need to fix this so basically this is the field of trustworthy AI and the good news is we are working on it next slide and then of course they are humans humans can also screw things up if you let them right and we want humans to be part of this of this process next slide but the biggest problem with humans these days is we don't have enough of them we don't have enough experts that will build AI solutions so that the US remains competitive remains the leader of this field worldwide but we also don't have people at the bottom of this pyramid the people who will be informed users who will know how to use these capabilities to benefit our society the good news is militaries are aware of this this is a slide borrowed from army AI integration center part of the army futures command and so they are working to to solve those deficiencies in stuffing next slide they are also approach attempts to automate some of the more mundane processes in AI so that we can scale our capabilities without necessarily bringing more people to the table and I think this two initiatives should go together because we have so many gaps that prevent us from deploying everything everywhere at all times next slide and this is my final slide I wanted to leave you with this I think nicely summarizes the topics we covered today between how we chose it and I because you will see a robot there you will see AI there of course not with your own eyes but it's there and you see a human and we want all those three together so behind this is actually a movie I'm gonna ask to run in a second but this movie is about 10 years old now it's an example of a successful application of brain computer interfaces technology so this lady was paralyzed completely legs and hands not working and her own neural network was wired with some electrodes you see those little boxes sticking from her head on the top and her neural network was wired to the controller of this robotic arm and she learned how to control this robotic arm and she's feeding herself a bar of chocolate paraphrasing famous quote one small nibble for a woman one giant bite for BCI arm can we please run a movie and that will be eight for me that's the same yeah now I can drop it on my hand drop it there we go okay we can stop thank you very much it's not a big bite that's three squares there's a woman one small nibble for a woman one giant bite for BCI arm thank you we can stop thank you very much