Automating Care Operations
Episode

Mudit Garg, CEO, and Co-Founder of Qventus

Automating Care Operations

World-class healthcare is impossible without world-class operations.

 

In this episode, Saul Marquez spoke with Mudit Garg, CEO, and Co-Founder of Qventus, about how the automation of care operations can be a vital part of the solution to many problems in the healthcare industry. He explains how Qventus built a platform that automates and coordinates care operational activities to make physicians flow through health systems more efficiently. Mudit talks about how AI and behavioral science come together in their platform and tackle different processes with several approaches, as not all need the same automation. He breaks down two examples, the availability and use of operating rooms and the excess days of patients with a hospital stay, to illustrate what Qventus offers to the healthcare industry.

 

Tune in to learn how automating care operations can lead to world-class medicine across healthcare!

Want to start your own podcast or offload the busywork of your current podcast to the pros?

Smooth Podcasting is the producer of our podcast. They help us deliver high quality audio, show notes, transcripts, podcast marketing, and so much more. We totally recommend them!

Check out Smooth Podcasting!

Get The Latest In Your Inbox

SUBSCRIBE

Automating Care Operations

About Mudit Garg:

Mudit Garg is the CEO and Co-Founder of Qventus. The Qventus platform uses machine learning and behavioral science to deliver real-time operations for leading health systems across the country, including academic, community, and safety net hospitals.

Prior to Qventus, Mudit co-founded multiple technology companies including Vdopia and Hive. He also spent time in McKinsey & Company’s healthcare practice helping large providers with organizational transformation and performance improvement.

Mudit has been recognized for leadership as one of the Silicon Valley Business Journal 40 Under 40. He is a Stanford-StartX mentor. He earned his Master’s in Business Administration and Electrical Engineering from Stanford University and a Bachelor’s degree from the Indian Institute of Technology.

 

Outcomes Rocket Podcast_Mudit Garg: Audio automatically transcribed by Sonix

Outcomes Rocket Podcast_Mudit Garg: this mp3 audio file was automatically transcribed by Sonix with the best speech-to-text algorithms. This transcript may contain errors.

Saul Marquez:
Hey everybody! Welcome back to the Outcomes Rocket, Saul Marquez here, and today I have the privilege of hosting the outstanding Mudit Garg. He is the co-founder and CEO of Qventus. The Qventus platform uses machine learning and behavioral science to deliver real-time operations for leading health systems across the country, including academic community and safety net hospitals. Prior to Qventus, Mudit co-founded multiple technology companies, including Vdopia and Hive. He also spent time at McKinsey and Company’s healthcare practice helping large providers with organizational transformation and performance improvement. He’s been recognized for leadership as one of Silicon Valley’s business journal, 40 under 40. He’s a Stanford Start-X mentor, and he earned his master’s in business admin and electrical engineering from Stanford. Such a pleasure to have you here, Mudit, thank you for joining us.

Mudit Garg:
Yeah, thank you, Saul. Very excited to have to be here as well, and thank you for having me.

Saul Marquez:
Absolutely! Now, Mudit, I know you guys are creating some major waves with Qventus and the work you’re doing. We’re going to have plenty of time to chat about that. Before we do, I’d love if you could just share with me what inspires your work in healthcare.

Mudit Garg:
Yeah, my inspiration for working in healthcare actually came from the work I was doing when I was at McKinsey and Company, as you mentioned. And the very first time, my work that I started with, helping a rural hospital improve its core operations and it was probably about almost 15 years ago at this point in time, and that experience really struck me. I saw a really interesting dichotomy in that very first experience, that really captures what we see in our healthcare system. On the one hand, that, even this rural hospital had absolutely world-class physicians, world-class treatments, world-class equipment therapies available, yet the experience of the average patient was far from world-class. And that was really puzzling, like how could those things happen together? And that matched. I grew up in India, it matched, sort of my view of the US healthcare system in that if you wanted the rarest of rare diseases to be treated, this is the best healthcare system in the world to be. And yet for majority of the patients, it fell far short. And what was really striking was that this was happening in spite of what I was seeing day in and day out, heroic efforts from staff and clinicians, and managers who would do whatever it took to take care of patients. So I saw a doctor who said, oh, this patient is waiting for the MRI, there’s no one to transport them, I will wield them. Oh, there is this patient that never got discharge instructions, let me run over and get that done. So in spite of that kind of a mindset of doing whatever it takes, in spite of all the resources, the world-class, clinicians, therapies, equipment, it was really odd to see us fall so far short of what we would need to deliver for our patients. And the staff that were doing these superhuman efforts were pretty quickly burning up. And I quickly came to the realization that it wasn’t the medicine that was behind the times as much as everything else that surrounds it was behind the times. That was sort of the first realization that world-class healthcare is impossible without world-class operations. And that’s sort of what I set out to solve. This is also about the same time when the Affordable Care Act was being discussed, and I sort of kept seeing that if the only change incentives, and we’re not able to drive a reliable system, we’ll just find new ways of wasting money, and we would not fundamentally solve the problem. That is both burning our people out and sort of preventing the full use of all the amazing resources we have in our healthcare systems. So that’s what I set out to solve, to create Qventus in a way that it could help health systems transform their operations. It was clear, much like, if you took systems thinking and step back and looked at the different operational challenges, you could actually do that in a much, much more automated fashion by taking modern innovations in AI and machine learning, using behavior science to shape the organization’s behavior and some of the fundamental operations management principles together. And we built a team with expertise in clinical operations, performance improvement, change management, and started applying them to long-standing operational challenges. We have many of that along the way of doing things that certainly turned out to not be the right way of doing things and had to adapt and adjust from there, but that core mission of delivering world-class operations has stayed true the entire time. And we’re really proud that through that journey we’ve been able to partner with leading organizations across the country. That range, as you were saying, from community hospitals, academic medical systems, large health systems such as Boston Medical Center, Honor Health, and …, Fairview, St Luke’s, Better Care, to name a few.

Saul Marquez:
That’s awesome. Well, Mudit, you know, this operations problem is definitely a big one, and making sure that it’s addressed is key. Talk to us a little bit about Qventus. You know, you mentioned, and maybe later we could cover some of the setbacks and learnings. Those are always, to me, the most interesting.

Mudit Garg:
Yeah, definitely, for sure.

Saul Marquez:
Yeah, that’s where the gold is right there. But how is, what Qventus doing, adding value to the healthcare ecosystem right now?

Mudit Garg:
Yeah, so I’ll give you a couple of examples. So firstly, overall Qventus is enabling hospitals to automate their care operations. So all those operational activities that are surrounding the delivery of care, things like how to get patients access, how to get help, the, all the coordination of care that goes on, how to make sure that the patients flowing through the health systems are able to get through in a reasonably efficient way. All of these sort of operational processes that are intertwined with the clinical care, delivery of care is what we help automate and make much more reliable. So, for example, take, for example, the challenge around the operating rooms, right? It’s really surprising that on the one hand, you will see patients who often have to wait months to get access to the operating room and surgeries. Similarly, surgeons who are saying, I really would love to get an ability to get into the operating room and are waiting quite a long time. Yet from the health system standpoint, when you see the same thing. You see that even if you look at the primetime hours, i.e. only Monday through Friday, only like 7 to 4, or 7 to 5, only about 70% of that time is utilized. So those, that time when you had staff, you had the equipment, you had everything there, but that dichotomy is really hard, like, how could those two things be happening simultaneously? And that’s an example that is actually, that improving our access is one of the top priorities for health systems. Over 75% of the health system executives said, this is one of my top priorities. So what’s happening today is oftentimes, if the, if you have a manager who’s absolutely excellent, who has a lot of time on their hand and is able to then look ahead and say, actually, that surgeon has time two Thursdays from now, and I have an inkling they won’t use it. They have a football game for their son or something else is going on and it doesn’t look like they’ll use it, and they have a block on that time. Typically, that’s how it works, that, the surgeons also have the right. Yeah, it absolutely is sacred. So how do you take that very sacred time and do, in a way, facilitate the release of that time and help another surgeon who actually needs it and their patients who need it be able to get? So our machine learning is looking at every individual surgeon, every local market learning on that on a personalized basis, right? Just like we have our Netflix use, learn on our personalized behavior patterns and pick up and say, okay, well, this plastic surgeon is always booking significantly in advance. For them to have not enough cases booked to towards this from now is a meaningful indicator that they perhaps are not going to be there in the afternoon that day, and so it picks up things like that. It proactively reaches out then to those surgeons’ offices and helps them understand behaviorally what’s the need for them. Hey, in this you get this upside if you release this. And so we get their schedulers to release that time, typically about 20 to 25 days in advance, so when, that would have that surgery would have happened. Now, instead of releasing it typically 3 to 5 days in advance, it’s 25 days in advance, you have time to get another patient access to that. And then our system actually similarly applies matching to figure out which surgeons most will need this, whose patients will most need this time, and proactively reaches out to them to get that time. And it’s remarkable what you can do with that. We’ve seen about, being able to do two additional cases per room per month, right? So if a hospital has 10 ORs, you’re doing 20 more cases per month, additionally, that those patients would have had to wait two months or more to get that case. And so that’s, those results have been transformative, like hundreds of hours of block capacity, at least 50 to 60% of that release time being used for these cases. And as I said, about two cases in an OR per month coming through that, so that’s one example. We have similarly examples on patients waiting an unnecessarily long amount of time passed when they needed to in the hospital. Like, nobody wants to stay in the hospital past when it’s necessary. And last year across our partners we eliminated about 200 years’ worth of total patient time in the hospital that didn’t need to happen. So those are examples of how we help the health systems and the executives.

Saul Marquez:
Yeah, and, Mudit, I love it, right, and I mean, two cases per room, 20 cases per month, that’s significant pickup there, of revenue, of throughput, you know, deferred care that doesn’t need to be deferred. So definitely some opportunities to scale. And you know, it’s hard to, to, to get a surgeon to give up that time. What kinds of incentives are we talking about?

Mudit Garg:
Yeah, it’s a really good question. So actually we have behavioral scientists, PhDs in behavioral science on the team specifically for this reason, because, you have to think about much like it’s been used in many areas of technology, getting users to do certain things that are for their good and sometimes not for their good in other areas. But here there’s a good we can drive through that and at the end of the day it comes to prompt, like helping someone understand there’s something that they can take action upon, what the action is, what the reward is, and how to sort of reinforce that. And so that’s what we do. So instead of just telling the doctor like, hey, do you want to relieve this time? …, we have the models trained on, at what point do we have enough confidence that we can reach out without overwhelming the surgeon’s scheduler. Second thing, from an incentive standpoint, we think about what are things that are going to that surgeon’s scheduler’s mind before they can release it. Well, firstly, what if I end up using it, right? So maybe I should wait a little bit more. So we actually have models that help them understand what are the chances that this current course and speed using the historical patterns they would use it. So that gives some reinforcement in their guards, if they feel like they are looking like it’s not going to be used, it reinforces that, yeah, model mathematics also match your gun. The second thing is, well, if in the off chance you ended up needing it, we actually give you the first priority for any additional time that comes available in …, right? So that, then there is that comfort as well that in the off chance that that happens, then we’re going to get reprioritized as the top of the range to get time back in the OR. So that’s the kind of stuff that we do that helps drive that incentive. And similarly, there’s also sometimes a room is available, but not the right equipment. Like you might be working on a room with a robot and you did your first case on the robot. Your second case is following, but it doesn’t need the robot. And so you’re continuing in the same room. But I could really use the robot, right? So helping explain the why that switch might be helpful is also sort of the kind of incentive that people always really value.

Saul Marquez:
Very cool, yeah, and I mean, in that particular situation, sounds like there’s capital optimization opportunities as well, right?

Mudit Garg:
Huge capital investment, these are massive, massive equipment, very expensive, very beneficial, but very expensive. So how do you maximize the use of them? Like we’ve seen, typically about 30 to 40% of cases happening in robotic rooms are often non-robotic, right? So that’s criminal after spending that much money on that on that piece of equipment to not be able to use it.

Saul Marquez:
It’s expensive marketing, make use of it. I love it. Mudit, so what makes what you do different and better than what’s out there today?

Mudit Garg:
Saul, you know, as I was talking about our, what we drive, which is care operations automation. So maybe I’ll first describe at the altitude of automation and what’s unique about solving automation in this space. And then we can talk a little bit about specific examples that I was giving, what’s unique and how that applies to that. So one, I would say this, of course, naturally a lot of focus on automation in healthcare and that spans back office operations, clinical decision-making. So on the one hand, clearly there’s a set of tasks that are back office tasks, mostly rote tasks that are separate from care delivery, and these are tasks that humans should not even be doing in the first place. There are good candidates for RPA, for example, and relatively straightforward, operational tasks. On the opposite end of the spectrum, you could take clinical tasks, they could automatically read an image and treat the patient. That’s, actually making clinical decisions and … High degree of human judgment and oversight required, human-led automation with intelligence support to inform clinical decisions. So those are two common examples that come to people’s minds. But these, things in care operations actually really meaningfully impact our ability and have a unique set of challenges within them. So things, as I was talking about access and patients’ care coordination, and patients flowing through the hospital, these are all high ROI areas and you see massive fluctuations in performance and outcomes depending on who is managing them. So how do you drive automation reliability so it’s not dependent on who’s managing them? Requires some very specific kinds of characteristics, so I’ll give some examples there. So one, of course, even though you’re driving operational decisions, because of the care operations, you actually need to understand deep amount of clinical data still, right? So you still need to understand, so the back office type of solutions don’t directly apply to this middle space, because you need to understand a lot of clinical data to understand, this patient can go home in three days, or they cannot go home, and things of that type. Second is, you often, these are human-in-the-loop automation, but you still have a human in the loop, but they’re kind of running. And for that, you need to have machine learning models that don’t just predict what will happen, but probabilistically, in certain situations you fully automate, in certain situations, you may only partially automate, and in certain situations, it’s a high downside case, you may just give the heads-up to a user. So the infrastructure, the machine learning required is also very different. It requires to be able to not only do the full automation, partial automation, and all that stuff but also decide when to apply which of those levels. And then lastly, when it’s not full automation, you need to have all those behavior science things come to life so that you can actually shape the human behavior as well to drive those. So that’s like infrastructure, and at a platform level, what you need to solve this place. And then now let’s look at like an example, like what’s different in how we solve the problem in the perioperative example, as we were just talking about, is historically, people may have looked at that problem in the rearview mirror and said, how will you block utilization? Well, doctor, your block utilization is not good enough, we should take some block away or we should give you more block. That’s a very different conversation and actually a much more frustrating conversation for the surgeon to have than to say, hey, you’re doing a great job. Thank you for releasing the time these two times proactively. Your block utilization is not good. If actually there were two more times you could have released it when the system prompted, if you did that, you’d be all the way in a good stewardship of the block, right? It’s a very different thing. There are people who’ve done sort of that work with rules as well, where you can say every time a doctor hits below 60%, we should get, we should remind them. Well, that sort of very crude method of doing it either ends up resulting in lots of messes going to the surgeon schedulers and then the tune it out, or you’re actually missing opportunities of truly actually getting that time released early enough because you’re waiting till it actually hits a 60% point, right? So that’s the kind of stuff that is very unique, first of all, in how we drive it. Secondly, all the rewards and incentives you document, they are very thoughtfully designed on how to maximize the improvements here. And lastly, this sort of closing of the loop to say, with the users is super critical as well to keep this as a fully functioning loop as well. So that’s sort of, both at a platform level, and then how we design our solutions with AI automation and behavior science ends up making us very different than, than other solutions.

Saul Marquez:
Yeah, it makes a lot of sense. Makes a lot of sense, Mudit. There’s customizable solutions for what the machine learning and AI does because there’s not a one size fits all. There’s the behavioral science piece that helps people do the right things so that they’re not overwhelmed and spammed, and you’re optimizing the opportunities overall, makes a lot of sense, and very useful, by the way. So what about those areas like ICUs where there are not cost centers? What are you doing there?

Mudit Garg:
Yes, actually, ICUs.

Saul Marquez:
I meant profit centers, right, there, cost centers, not …

Mudit Garg:
Cost centers, not profit centers, right. So actually, a lot of the work with you on the inpatient side, the excess days, the extra time the patient spends in the hospital, looks at all that steps of the journey. You know, you as a patient came in, and like a day or two before your discharge, we looked at the fact that you might need a sniff and it might now take us a while to get the sniff authorization, and for you to figure out which sniff is appropriate for your family and all of those things, and now you suddenly have to spend two or three extra days in the hospital. So we are looking, we’ve started, basically, about a half a million discharges across the country, and we found some operational practices when applied, consistently to reduce the excess days. And it’s not actually surprising, it’s, if you have a plan for the patient within the first day or so of where they will go eventually and when they could leave. Just the fact that you’re plan … around that upfront, actually creates a lot of downstream benefits because everyone is working around that plan and modifying the plan as needed as well. And we see about 75 to 80% fewer excess days than when that happens. So we have models that are looking at every individual patient, like, okay, this patient is on liquid diet, they’re now moving to oral diet. They’ve started moving around a little bit, they’ve been able to walk around, all of those things. They have line or they have a drain or they have a catheter, all of those things about each individual patient and about twice the accuracy of care team, it can predict early on, it’s hard to predict perfectly, but it can predict early on what’s a good, reasonable plan. And then we are to … For the team so teams don’t have to document it, we are the … where the model has high confidence, but it doesn’t have high confidence, it actually suggests to the team what’s an appropriate plan to have here and also suggest to them to modify the plan as the care goes on. And that’s an example of how we take what’s traditionally a cost center, like the cost of the patient’s excess stay, and make sure that the unnecessary parts of that start getting shaved off, what was clinically unnecessary. And that’s sort of been how we’ve done that. We actually, similarly, very specifically applied that to the ICU as well, especially in the pandemic. As you can imagine, ICU capacity was at a huge premium, right? And it was, there were a lot of these sort of centralized war rooms that each of the health systems had trying to figure out where they would need to create ICU capacity if they ran out of it. Now, traditionally, when health systems have been managed to touch and feel, there’s someone in a unit walking around and saying, okay, which patient could go to the step-down unit if you really need it some capacity? But that’s hard to do centrally if you have 12 or 15 hospitals. So we actually created a model that looked at which patients are most ready for stepdown, so we can proactively start planning them for them to leave the ICU. And we then actually tested it blind. We had three physicians, two intensivists, and one hospitalist, evaluate each patient in the ICU for their readiness to step down in the next 24 hours, and then we had the model do the same, and we saw that they agreed with each other about the same amount of times, that they agreed with the model. So about 80% times they all agreed with each other and they agreed with the model, but … So it became a really cool vehicle to be able to centrally say, if I had to pick five, six patients who can, who are most clinically appropriate to be stepping down, how do we start planning for that? And we’ve seen that you can shave off about a point three days of the ICU stay, so about 8 hours of the ICU stay, that typically ends up getting lost to coordination, right? So when someone says, okay, this patient looks good to go, then you have to find a transport, you do, all of those things can happen proactively, upfront.

Saul Marquez:
Man, that’s so cool, Mudit, and it’s the compound effect from there, right? I mean, little here, little there, and then you’re optimized. It’s the difference between losing $1,000,000,000 in a quarter or being profitable, right?

Mudit Garg:
That problem is about $37 billion of excess days in the country, right? So just the problem of excess days. So it’s just mind-boggling.

Saul Marquez:
It’s mind-boggling for sure. Hey, so at the beginning of the, of our chat, you talked about setbacks and tweaks. Talk to us about that. What’s been one of the biggest setbacks you’ve had and a key learning that came out of that?

Mudit Garg:
Yeah, so I’ll share a couple of examples and maybe talk about from the beginning point what we saw. So you remember the first piece was around seeing how to drive care operations improvement. Understood data is going to be pretty key to it. And in the beginning, applied machine learning to these problems to start sort of predicting what might happen. And one of the early setbacks was actually like we used to predict how many babies you would have delivered in a day in the mother-baby unit. So that the nurse understood how much staff she might need. And we used to predict every day, I would go every day to the charge nurse, doctor, like how they used it and how good it was, and they, think I remember one day I knew we had pretty much exactly nailed how many people would be in the unit that day. And I went to talk to her and I was like, Hey, like we predicted, what did you think? She’s like, Oh, yeah, I think it was it was fine. How’s it? What do you mean it was fine? Like it was spot on. Like, how come you didn’t? How come you’re not more excited? Like it was exactly what you asked me, and she’s like, Oh, yeah, Like, you’re right. Like, it was accurate, but, you know, honestly, I just had so much going on, I didn’t have time to think about, what if it was right? What if it is wrong? What should I do? And that was a really key lesson that the machine learning alone wasn’t enough. If you couldn’t help the user take the action, it was pointless. We were just kind of going and having a lot of fancy technology without actually being able to drive any outcome. That was one early setback, then sort of drove a lot of the behavior science and how to build these things together for us. So then we built the platform to be able to apply that kind of thinking to almost anything you might need to apply to. So we built a really flexible platform and we would actually work with our health system partners in a very flexible way. So we would go and say, Oh, you have this area of problem, like let’s get the data, and you know how health systems often have a lean improvement mindset that you would try a new thing. And so we basically would accelerate that massively. In days, you could, we would have a new predictive real-time intervention live by talking to the frontline staff and seeing what they want to do. And we saw massive innovation sort of going through health systems with that and it was very, very cool. And now sort of how we started scaling from what we were doing into patient falls and patient medications, IV to oral and all kinds of areas very, very rapidly. What I started seeing, and this was sort of the setback there was, as we went deeper in the market, I started seeing and hearing two things. One, I think for health systems, I remember actually a health system CEO saying, well, this is a Ferrari, and I don’t know if my people even can drive it. And I was like, What do you mean? Like, why not? She’s like, Look, we have so much going on. We don’t know if we have time to figure out what is it that we should, what intervention we should do, what intervention we shouldn’t do. What if they pick an intervention that takes a lot of effort but actually doesn’t drive enough results, like this is just, requires too much for us. And I started seeing that more and more as we went deeper in the market, and that was a big setback because it changed how we approach the problem, but it ended up being one of our biggest strengths as we changed our approach. So we changed our approach then to say, you know what, while we have this flexible platform, we’re no longer going to manifest ourselves as a flexible platform to our end users. We are going to study a couple of high-value problems that they have, much like I was describing in the inpatient side. We’re going to then study what are those things that excellent managers do that if we can do them consistently, we can actually drive higher results. And then we can apply the full force of AI and automation and behavioral science on those couple of leverage points, and we’ll design these very, very specifically crafted solutions around these problems, like in the old example, the inpatient example, that can then drive consistent results no matter what the operational environment might be. And that became our approach and that actually stood us in really good stead, especially as the pandemic hit and things like that. That approach, we call it now the Solution Factory, where we can basically take the flexible platform and the scientific approach of designing solutions, and together they’ve been very, very powerful in driving the outcomes that I was describing to you.

Saul Marquez:
That’s fantastic, Mudit, and having these focus areas being really damn good at a couple of things that matter a lot, gave you guys that traction that you were looking for.

Mudit Garg:
Absolutely.

Saul Marquez:
That’s great. And so if you had to say one thing that you’re most excited about today, what would you say that is?

Mudit Garg:
You know, we are at such a time of a perfect storm right now. Health systems have emerged from several years of just absolutely heroic efforts for fighting the pandemic. Our labor workforce is tired and there’s massive, both inflation as well as reimbursement pressures that are driving some very, very significant losses across health systems right now. That’s a very tough situation to be in. But what’s exciting for me is that’s driving action that is going to create long-term change in what we can drive to truly deliver world-class care to our patients. So back to what I started with, we have world-class clinicians, we have world-class physicians, world-class staff, world-class treatments, world-class care, but often creating this operational infrastructure around it to deliver world-class medicine, truly, it’s been hard. But now we have a burning platform. We have a case for doing so, and I’m seeing health systems take that long-term view and really come to us and say, look, we can’t just build more ORs to drive more surgical access. We have to figure out how to drive that. We can’t staff enough beds to take care of the patients we need. So this is not only a cost problem, this is a truly patient care problem to help figure out how to take out excess days. And I think that’s going to actually force us to make some changes and create the kind of care operations automation that can systematically bend the cost curve going forward. And that, I’m really excited about seeing that happen on the backs of sort of this crisis, and I think that will both help solve the crisis, but also help set us in a much stronger position long term as well.

Saul Marquez:
That’s fantastic. Yeah, the pandemic has certainly changed a lot from how we work to how we think. And so I think that’s well said, Mudit, exciting times ahead. So, listen, this has been incredible, and I can imagine only the tip of the iceberg of what you guys do. So I’d love if you could just, help me end this session today, Mudit, with a closing thought, and then the best place for the listeners could go learn more about you and Qventus.

Mudit Garg:
Absolutely, so maybe a closing thought would be, at Qventus we help automate care operations and we are just super excited about what that can do to truly drive world-class medicine across healthcare. To find us, you can find us at on LinkedIn, searching for Qventus, or you can find me at Mudit, M U D I T @Qventus.com.

Saul Marquez:
Beautiful. Listen, I appreciate it, Mudit. And folks, check out the show notes where we will leave links to Qventus, Mudit’s profile in LinkedIn, different ways to get in touch if the opportunity is there. Take it, don’t just listen, take action. Thank you for jumping on with us. This has been fun.

Mudit Garg:
Thank you, Saul, really excited. Had a great conversation. Thank you for having me.

Sonix is the world’s most advanced automated transcription, translation, and subtitling platform. Fast, accurate, and affordable.

Automatically convert your mp3 files to text (txt file), Microsoft Word (docx file), and SubRip Subtitle (srt file) in minutes.

Sonix has many features that you’d love including share transcripts, automated translation, transcribe multiple languages, automatic transcription software, and easily transcribe your Zoom meetings. Try Sonix for free today.

 

Things You’ll Learn:

  • Qventus’s platform uses machine learning and behavioral science to deliver real-time operations for leading health systems across the country, including the academic community and safety net hospitals.
  • Only about 70% of operating rooms’ primetime hours are utilized.
  • 30 to 40% of cases in robotic rooms are often non-robotic, even after spending a lot of money on the equipment.
  • Machine learning models can determine which operations need to be fully automated and which can be partially automated.
  • A plan for a patient within the first day of their hospital stay avoids excess days as the care team can have the plan to work around. 
  • Excess days are currently a 37 billion-dollar issue in the industry.

Resources: