A Practical Framework for Evaluating AI in Healthcare

Episode Contents

• 0:00 NovaNav Introduction
• 1:00 Guest introduction: Dr. Mohammed Quadri
• 2:50 Reducing operational burden with AI via a Transdisciplinary Model of Care
• 7:02 CMS TEAM emphasis on technology
• 8:40 Delivering measurable operational value with AI
• 14:13 Data quality and readiness
• 16:50 Innovation pilots and scale deployments
• 20:50 Balancing governance and speed of adoption
• 22:58 Framing the value of AI in healthcare
• 27:50 Measuring "prevented" metrics such as Avoidable Readmissions
• 29:09 CMS TEAM: Opportunities in technology and AI for 30 days post-op
• 30:20 CMS TEAM overview
• 32:00 Where AI creates value under CMS TEAM
• 37:20 Discussion of AI tools
• 39:10 Wrap-up

Key Takeaways

He shares where AI is already reducing operational burden, including ambient clinical documentation, revenue cycle automation, and the emerging opportunity in perioperative care coordination. The conversation also explores why many AI initiatives fail to scale, and what conditions must be in place for measurable results.

Transcript

Hi, Dr. Quadri. Hi, Lisa, how are you? I'm good, how are you?

Good, good, good. It's nice to meet you. Amrit has talked about you so much, kind of feel like I know you. Yeah, Lisa, I think I know you as well.

I don't know if we ever met, but I think we did speak once. I don't know. Yeah, we did. You know, yeah, I think we did speak once.

And I know if you've been obviously on the advisor calls, but yes, I think we did speak once. So I wanted to, you know, so your questions are okay, because Amrit really was very much speaking about your AI and your, you know, your strength in that area. Those questions felt good to you. Was there anything you think I missed or will you just add it in the conversation?

No, I think you didn't miss anything. In my opinion, you know, the questions are straightforward. And certainly I'm going to give more detailed explanation. Yes, please.

I will also explain like a framework that I use that will allow both providers and payers from patient's perspective, how a platform should be evaluated and what does measure of success look like while implementing AI solutions. Yeah, I'm really looking forward to the conversation. I want to, I have to pull down your LinkedIn profile. I had it up.

You had to bear with me. I want to make sure that I speak properly about your credentials and your background. So give me one second. This is the second podcast I've done today.

So I'm a little. I understand. If you want, I can actually share like a brief bio in an email. Could you do that?

Would that be better? Because I was just going to sign into LinkedIn. That would be better if you don't mind. That would be better.

Thank you. You are in my old stomping grounds. I had, I owned a, it's a founder and CEO of a company called Thigh Healthcare. And you're at RGVJ, right?

Hackensack Meridian Health. Hackensack Meridian Health. You've done work at Hackensack. I've done work at RGVJ.

I remember that from our conversation. Yes. Yes. Lisa, I have your cell phone number, right?

I can put it in the chat. Yeah, that way I can just text you. Perfect. Or you prefer email.

No, whatever. I prefer whatever's easiest for you. Yeah. I can actually just copy and email it.

Or I can text you. You can copy whatever's easiest for you. Okay. 7, 3, 2.

3 1 9. 5 7 0 0. That's correct. Great.

Thank you. Awesome. Okay. In real time.

Yeah. Real time. That was quick. So we're not on.

This is not going to be a video. So we will. You know, we can just, even though you look great. This is just going to be for audio.

That's really it. So I'm going to, I'll start. We're recording and they'll pick up from, from where we begin. So if you're ready.

I'm ready. Whenever you are. Okay, great. Welcome to the surgical journey podcast.

Sponsored by Nova nav. I'm Lisa Miller. Chief strategy officer. And really excited to have today.

Dr. Quadri. Dr. Thank you for being on our podcast.

Thank you, Lisa. I'm honored to be. Participating in this podcast and looking forward to it. Yeah, that's great.

Just for the audience. I'd like to let them know who you are. So Dr. Muhammad quadri.

Is VP of strategic. Our strategy for academic research. Innovation at Hackensack Meridian health. I'm going to take two on that though.

Sorry about that. All right. Dr. Take three.

Dr. Mohammad Quadri VP of strategy for academics, research and innovation at Hackensack Meridian health is a leader with a diverse background in medicine, business and healthcare administration. He is his key accomplishing includes establishing. The H M H research Institute and unifying its operations, creating the academics research and innovation executive council and developing strategic plans for the health system and research Institute.

Dr. Quadri also led the development of a virtual health strategy created implemented ambulatory expansion strategies, conducted feasibility studies for service line growth initiatives. Implemented. Epidemiological strategies and develop value-based care model and published numerous research studies in clinical areas.

With over 18 years of experience in healthcare leadership roles. Dr. Quadri is a passionate advocate. For innovation.

In healthcare committed to improving overall patient health and wellbeing by addressing unmet. Needs in real time. So. I am honored to be here speaking with you.

No, it is a lot, but you know, it, I think. It speaks to, you know I think the innovation, the entrepreneurial spirit, the, the ability to really address like all these problems in healthcare, I think, you know, which is why I'm so excited to talk with you, but let's talk about a little bit about AI. Health systems are inundated now with AI tools, right? They all promise efficiency.

From your perspective. Where do you think AI genuinely reduces operational burden right now? Today? Thanks.

That's a great question, Lisa. You know, I'm going to introduce a concept as well to the audience of this podcast, which is called as a transdisciplinary model of care. And it is a continuity of care model. When I think about where AI is genuinely moving the needle, I find it helpful to consider three dimensions of care, continuity, information, management, and relationships.

The tools delivering real value today address at least one of these, the most impactful address all three. On the information continuity side, clinical documentation, AI is the clear winner. Of all 2024 survey of 43 major health systems by Scottsdale and Steele found that every single respondent a hundred percent reported active engagement with ambient documentation tools and 53% reported high success. That kind of consensus is remarkable in healthcare.

Physicians are reclaiming meaningful time. Studies show four to six hours weekly time that was spent translating clinical knowledge into EHR entries now available for actual patient care. For management continuity, revenue cycle AI is producing hard dollar returns organizations are reporting claims. Review times dropping by over 60%.

Denial rates falling significantly. But I'm particularly excited about what's emerging in perioperative care coordination. AI system that ensure patients are optimized before surgery. That automate pre-admission testing workflows that coordinate across the entire surgical episode.

One platform recently announced results showing up to 40% in surgery cancellations and 50% productivity gains for pre-admission teams. Under the new team model, where hospitals are financially accountable for the full surgical episode, that kind of coordination isn't optional. It's survival. Lastly, the relational continuity dimension is where I see the most untapped potential.

Multimodal patient engagement, meeting patients where they are through voice, text, apps, whatever works for them, can extend therapeutic relationship beyond hospital walls. The goal isn't replacing human connections. It's ensuring patients feel supported throughout their care journey, which paradoxically often requires AI to make that level of personalization scalable. What I'd caution against is investing in AI just for the sake of AI.

The tools that succeed solve real workflow patterns. The same Scottsdale Study Institute survey showed that imaging and radiology AI, despite significant adoption, had only 19% of organizations reporting high success. Technology sophistication doesn't predict value. Workflow integration does.

Hopefully that answers your question. No, it really does. I love the thinking, the model you presented. And I think it's so relevant because it brings AI into practical terms, the here and the now, right?

So we can understand those three aspects of the model, which I love. And you brought up something I just want to mention before you go to the next question is, for teams, the new team model really does say they want technology to be used. I mean, it's pretty explicit about the use of technology. I loved how you said supporting the care throughout the journey.

So this isn't meant to use in one specific incident, if we could have the technology to be there to support the care throughout. I think that's what we're looking for, right? Absolutely, absolutely. And just to add on to that, like a hospital facing team losses, heart penalties or quality score reduction can see cumulative margin erosion of approximately like five to 6% or more on Medicare surgical patients.

So the AI-powered solutions that reduce cancellations, prevent readmissions and automate prompts collection directly address all three revenue threats. Yeah, I love that. I agree. And it's so challenging.

Like you talked about revenue cycle. Revenue cycle sometimes is, you know, it's about capture, but many times it's going back and trying to recapture the money. And so if really AI can, like you said, prevent that revenue leak, it's a great opportunity. I'm a big fan of that used to.

So in large health systems like Hackensack Meridian, you know, sometimes the biggest constraints are often, you know, behavior, right? It's change, it's workflow. It's sometimes it's not the technology, although, you know, getting technology through IT sometimes can be challenging, right? But it's usually the people and wanting to adapt or learn.

And then how do you make it work in the workflow? So what do you think are the conditions that have to be in place for AI so that they can deliver, AI can actually deliver a measurable operational value? Another great question. I'll tell you like to your point, health systems are certainly inundated with AI promising efficiency tools, right?

From what I think about where AI is genuinely moving the needle, I find it helpful to consider, again, going back, I'm going to link the entire conversation during the podcast to the continuity of care model that I actually described, the information and management relationships aspects of it. The tools delivering real value today address at least one of these. The most important is like you need to understand that they address all three, right? I already mentioned that to you.

But this question about the biggest constraint in often behaviors and workflow, not technology. I think this question gets to the heart of why somewhere around 80% of healthcare AI initiatives don't deliver expected outcomes. The technology works. Yeah.

The technology works, right? The implementation fails. And you're right that workflow and behavior are usually the limiting factors. And I usually say this to everyone, technology grows exponentially, but the adoption rate remains pretty much flat.

So how do we bridge that gap? And I think that's where Hackensack Median Health is actually excelling. The first condition is problem clarity. Organization need to start with what outcome are we trying to change, not how can we use AI.

In perioperative care, that might be reducing day of surgery cancellations, improving protocol adherence, or, and this is critical under team, decreasing 30-day episode cost. This metric has to be defined before technology is selected. Secondly, workflow integration has to be designed from day one. The most successful implementations don't bolt AI onto existing processes.

They reimagine workflow around what AI enables. If you are, let's think of it this way. If your pre-admission testing team is drowning on phone calls and manual chart reviews, AI can help. But only if you redesign the workflows so staff are working at the top of their license while AI handles the administrative coordination.

And third is data readiness, which is non-negotiable. This isn't about having an EHR. It's about data quality, interoperability, and the infrastructure to feed AI systems and act on their outputs in real time. I always am a big proponent that if you want to meet an unmet need, it has to be in real time, rapidly, and accurately.

Those are the three big buzzwords for me. If you do not address an unmet need in real time, rapidly, and accurately, then you are not successful. The organizations succeeding with AI have invested in their data foundation first. And lastly, there needs to be a clinical championship.

Technology can be mandated. Adoption cannot. You need physicians, nurses, front-line staff who see AI as serving them rather than surveilling them. They should be involved in shaping the implementation, not just receiving it.

Finally, this is often overlooked. The deployment timeline matters. In surgical services especially, the best implementations are rapid, operational in days, not months. Long implementation cycles allow momentum to fade and workflow to drift.

The new paradigm is configurable, no-code platforms that can be tuned to local protocols without extended IT projects. This is like I've taken more notes in the first few minutes, Dr. Quadri. This is great.

I'm on overload. I have to ask you, have you written a book yet? I'm sorry. I think you need to read a book.

I write a book. Sorry. This is amazing. So I want to go back to a couple of things you said about what are we going, what are we trying to change?

I think that that's a great question. I think every initiative should lead with that question. What outcome are we trying to change? I think that should be a headline for everybody listening because if everyone has that North Star, then think that's how change probably stays the course.

That's the focal point. I think we forget about the one outcome or what outcome we're trying to change. I love that. And I want to talk about data readiness because I think the data quality is so key, right?

So you've got to have great data in order to have the output. Can I ask you a little bit more about that? Because you must be concerned at times about the data quality. Yeah, absolutely.

Sometimes data quality certainly is not at the level to which we desire, but we have teams in place that have taken initiatives and implemented processes and protocols that will ensure that our data quality is up to par and there is no compromise there. Yeah, I think that's really important. I think a lot of times physicians get concerned, rightfully so, like they question the data. So now they're being given data about themselves and they're really unsure, is this even right?

And they're probably right because at the heart of it, you're scientists too, right? So you look at data. And one more point and then we can move on because the workflow from day one is really important. There's a book, I believe, called How Big Things Get Done.

And it's a really great book about these projects, whether they're bridges or tunnels and, you know, like why do some succeed and why do some fail? And the author was a part of a lot of these projects, but the key, which sounds going to sound overly simplistic, this whole book was written about it, but he goes into some great details. It's all about preparation. Literally, like even if you slow down and you're preparing with structure, you will beat timelines, you will come in under budget, you will succeed with your outcomes, but it's all about preparation, the right preparation.

Absolutely. Absolutely. Totally agree with you. Go ahead.

No, you see that. I was going to say, do you see that in your work about like, well, you must because you said workflow from day one, it has to be, everything has to be built on the front end is what I heard. So it sounds like you believe that it's so important that any kind of tool, you know, it has to be part of that workflow from day one. Absolutely.

And, you know, regardless of healthcare or not, like in general, preparedness matters a lot. Yeah, exactly. Yeah. Overly simplistic, but I don't know, like sometimes I feel like we're all rushing to get something done, you know, and I don't know, like, you know, it's, I think it can't be overstated.

So you've seen both innovation pilots and scale deployments. I mean, just with, you know, with all the things you've implemented at your health system, what differentiates AI initiatives that move beyond a pilot and become embedded in day-to-day operations? Because there's lots of pilots. I think that's, I think hospitals are feeling comfortable, like, well, before we bring it in, let's do a pilot, you know, and even the vendors are selling pilots, right?

But how does that get moved? Like, what's the decision matrix for you from a pilot to get it embedded? You know, certainly, you know, we all have watched many pilots that delivered impressive results in controlled settings, but never graduated to operational reality. The pattern of success versus failure is remarkably consistent.

First, successful initiatives demonstrate measurable ROI early and clearly. The mantra is pilot, prove, and scale. The proofs phase has to include hard numbers. In perioperative care, for example, this might be reductions in cancellation rates, improvement in patient readiness, staff time captured or recaptured.

If you can't quantify impact within the first 45 to 90 days, scaling becomes a hard sell. Under team, those numbers translate directly to margin, making the ROI case clearer than ever. Second, they solve for all three dimensions of continuity of care simultaneously. A tool that improves information flow but creates more work for staff will be abandoned.

A tool that automates management tasks but doesn't give providers visibility will be mistrusted. The solution that scale address information, management, and relational continuity together. They provide real-time dashboards for providers, automate their administrative coordination, and improve the patient experience. Thirdly, they fit existing workflows rather than demanding wholesale change.

The best implementations integrate with existing EHRs, communication channels, and care protocols. They meet providers and patients where they already are. Multimodal engagement, app, text, email, voice, they all matter because different patients prefer different channels of engagement. EHR integration matters because that's where clinicians live.

Lastly, and this is a very important aspect, that is governance. Governance is built in from the start. That's very important. Scaled solutions have clear accountability, monitoring processes, and feedback mechanisms.

They are not point solutions that operate in isolation. They are part of an institutional capability with oversight and continuous improvement. I'll add one more observation. The technical sophistication of AI often doesn't predict success.

Some of the most successfully scaled tools are technically straightforward but brilliantly integrated. The complexity is in understanding clinical workflows, change management, and ongoing optimization, not in the algorithm itself. Go ahead. Sorry.

No, go ahead. I was going to jump a little bit, only because you spoke about governance, and we can circle back to a couple of other questions. Sometimes people hear governance, they freeze. It is meant to do, you need governance.

You need very good governance. What's your feeling about the balance between that and slow adoption? You need oversight, you need vision, you need leadership, but you also need to maybe move fast. Is it fast enough for AI to matter operationally?

AI is moving so quickly now. I mean, look in the past couple of months, even how far we've come. No, I think that's a great observation. The tension between AI adoption speed and governance in healthcare is one you're navigating, you know, many folks are navigating across multiple fronts.

So I think the core challenge is that moving too slowly means missing clinical and operational value. So I think while moving too fast, if you move too fast, you actually risk patients' safety, and the incidence goes up. Regulatory backlash, sometimes that's a key consideration, and then erosion of clinician trust that can set adoption back like years. So I would say like, you know, when you are actually establishing a governance structure, linking it back to information management and relationship aspects, you've got to take into consideration all those different aspects of continuity of care, but at the same time, ensure that you have the right buy-in at the right time to be able to deploy it in a timely manner.

That's where you will be able to address those unmet needs. And real-time rapidity and accuracy. Yeah. No, I love that.

I love those three. Can you talk a little bit about, you know, and you just talked about ROI, right? So there's so many benefits of AI, and they often show up as, right, avoided work, reduced friction, faster decisions. But how do you think about, or how does your team and other executives think about the value of AI when maybe the gains aren't strictly, you know, they're not like a financial dollar, but maybe they're other, we like to call them soft costs, but I think they're not really soft costs.

They're just, I don't think we take into the account, you know, avoided work, reduced friction. So how do you take that into account when you look at AI? You know, this is one of the most important questions in healthcare AI right now. Traditional financial frameworks, they aren't really designed for investments where much of the value is in work that didn't have to happen, complications that were prevented or staff who stayed instead of leaving.

I'd suggest thinking about value across three horizons. You know, the first is going to be like a direct operational impact. Some AI delivers like clear hard dollar ROI, like we talked about revenue cycle AI that reduces denials and accelerates collections. And another, you know, going back to perioperative care, perioperative coordination tools that reduce data of cancellations by 10 to 12% have immediately quantifiable value when you consider the cost of MTOR time.

Under team, preventing a single readmission saves the full cost of that admission against your episode target price. These metrics should be tracked and reported. The second horizon is avoided harm and risk mitigation. Like what's the value of readmission that didn't happen?

A complication identified early through patient-reported outcomes before it became emergent. Under bundle payment models like team, this risk mitigation has direct financial implications. And as you know, hospitals are accountable for 30 days post-discharge. So the value-based care market is projected to reach $43.4 billion, if I'm correct, by 2031.

And it's growing at a significant pace of like almost 15% annually. That growth reflects the reality that risk mitigation is becoming core to healthcare economics. And lastly, you know, workforce sustainability and patient loyalty. So when clinicians say they have more time with patients, when staff burden decreases by 50%, when patient satisfaction scores improve by 7%, these translate to retention.

Absolutely. Recruitment. And, you know, advantages, other advantages, including it creates a market differentiation for a health system. I agree.

It's competitive advantage. Absolutely. And replacing a physician can cost over a million dollars. What's it worth to keep them?

Like patient satisfaction drives referrals and reputation. HCAP scores both clinically and competitively. You know, my advice is establish baseline metrics across all three horizons that I mentioned before implementation. Track both tangible and intangible benefits.

And then give AI time to demonstrate compound value. Initial ROI can be very modest, but year two, year three, the returns often exceed expectations. And as workflows mature and AI models improve. You know, I never, I never thought of it.

Compound value. You make a really great point because sometimes we probably might evaluate ROI too soon, or it's good to evaluate it, but it does compound. It compounds over the AI getting better. It actually probably compounds because the users are getting better.

Absolutely. Their understanding, they're getting used to having real time. Accurate rapid data like that. There's like people have to get used to that too.

They're not used to either getting that information so quickly or, you know, what do you do versus the lag? So that's a great point. The compounding value. You mentioned teams and.

Oh, I want to go back to one point. I'm just going to reiterate it. Someone else said this and I think it's so, so important. You know, Whether it's AI or some kind of platform could be something like we have at Nova nav.

Or any other use in any other technology. We should be measuring avoided readmissions. Right. Those readmissions.

That, you know, really didn't need, not, not for the sake of. Not having a readmission, but that should, they shouldn't have been for lots of reasons. I mean, it's not. It's not good for the patient and their families to, to go in unnecessarily to the ER, but I love.

Preventable and avoidable. Yeah, that's a good point. But that should be measured. Like, do we have it on a dashboard that says, okay, you know, This was a, this patient, or this was a preventable readmission.

I think those. That's a great measurement for hospitals to look at, but I don't know. I haven't heard of anybody that, that does that now actually tracks that. I don't know if you track it or thinking about tracking it eventually, but I think it's a great idea.

So I have a. One last business question. I'm going to ask you a kind of personal, not personal question, but your, some of your favorite AI tool questions. You mentioned team earlier.

And so now accountability stretches to 30 days. You know, post-op. When the patient leaves the hospital. And so.

Where technology or AI. Do you think has the greatest potential to, to not only just reduce administrative burden, but actually. It was Dr. Himes that said yesterday.

He called it active care. I love that because it really is active care until. You know, 30 days, but. Where do you think the best use of technology AI is through the whole.

Coordination of, of communication and, you know, being able to care for the patient. I think if we are going to talk about. I think it's, it's also important for us to also like. State it one more time for those who are not familiar.

That's a good point. Yeah. The transforming episode accountability model. It's a, it's a mandatory Medicare bundle payment program that began.

This year, as you all know. And it certainly affects like. More than 700 acute care hospitals. Across 188 metropolitan areas.

And it's, it's currently getting. Cable services for five surgical categories. Joint replacement hip fractures, surgeries, spinal fusion. CABG or so-called as cabbage and major bowel procedures.

Hospitals. Now receive a target price covering the surgery plus 30 days of post-discharge here. If your costs exceed the target, you lose money. If you deliver efficient, high-quality care, you can earn a bonus.

So the financial stakes are significant. Analysis show two-thirds of the hospitals will face losses, averaging up to like $1,350 per episode based on current spending. CMS applies a 1.5 to 2% discount to target prices. This isn't theoretical.

It's happening now. Yeah, it's here now. Yeah, it's here now. And the team fundamentally changes the accountability equation in ways that make the transdisciplinary continuity model that we talked about at the beginning of this podcast not just helpful but essential.

I think let's do this. Let me actually map where AI creates value in each of these dimensions under the team. So for information continuity under the team, AI-driven risk scoring and real-time dashboards that you just talked about actively monitoring, right, that give providers visibility into patient recovery status across the entire episode. When you are accountable for five complex surgical procedures or five complex surgical categories that we just described, procedures with very different post-acute care patterns or post-acute patterns, you can't rely on manual tracking.

AI can flag patients who are deviating from expected recovery curves. And before those deviations become complications, and readmissions eventually. And from management continuity on the team, I think like automated pre-operative optimization workflows that ensure patients are ready for surgery, labs are complete, clearances are obtained, medications managed, education is delivered, AI that handles that administrative coordination of pre-admission testing, escalating to staff only when intervention is needed. Automated post-discharge protocols that trigger appropriate outreach.

Industry data now shows up to like 40 percent reduction in surgery cancellation and 50 percent productivity gains for pre-admission teams with right tools. And then lastly, from relational continuity on the team, I think the multimodal patient engagement that means, that actually meets patients where they are, right? So making it convenient for the patients, improving access, app, text, email, voice, personalized, and then procedure-specific education is important, both pre-operatively and post-operatively. Yes, exactly.

Right, and then digital check-ins that make patients feel supported while providing the data needed for proactive interventions. Teams' quality measures include patient-reported outcomes, tools that systematically capture prompts while providing patient support, so both compliance and care goals. I think the key insight is that team structure creates conditions where AI is not just helpful but necessary for success. I think that's well said.

And then manual processes simply cannot deliver the coordination complexity required when you're accountable for every patient across 30 days, multiple settings, and five procedure categories. The hospitals that thrive, in my opinion, will be those under team, will be those that use AI to extend the reach and effectiveness of their care teams. I couldn't agree more. I love how you put in every single aspect of where AI could be used, because that really did break down the real-life opportunities within a real-life current bundle.

I think that's great. And I think, to your point, I think that's what hospitals have to do. They've got to see where they could put that into these bundles or into other problems that need to be solved. But it's really become specific from analytics to really understanding where the patient is in their recovery, everything in between.

Absolutely. And one last comment I'll make is, AI is here to stay. Yeah, definitely. And the faster we embrace it, the sooner we'll create operational efficiencies, we'll improve clinical decision support, we'll improve patient outcomes.

And at the same time, we maintain and create an experience for the patient. Think of it like what Disney does. They are not focused on satisfaction. They are focused on creating an experience for the patient.

It's a totally different mindset. It's an experience. When you think about what some of the children's hospitals do, it's an experience. Like even when they go down and have to get a CT or MRI, they've created this experience for the kids to think they were in a movie setting or something.

And in this very challenging time for a child, they're thinking experience. But if you think about it from what we could do in the home, how could we make it a really great experience? I think technology affords that. I don't see patients putting a VR set on, but I don't know, I never know where the glasses are or we don't know where tomorrow really is.

It's probably here already. Yeah, absolutely. Engagement leads to better experiences. Yes, definitely.

So do you mind sharing with us some of your favorite AI tools? Is that OK? Because just from a practical, like your go-to, so I'll share, like I use Perplexity for research and might use Claude for writing and, you know, Chad for maybe custom GPTs. And I've started to use Lovable for vibe coding.

Like I'm experimenting in different ways. I guess your question is, so what's your thought about like people maybe, you know, like in your health system trying to figure out what they can use day to day or what's your favorite tools? I know I put you on the spot. It might sound like a very easy question, but it's one of the most toughest questions to answer.

And I don't want to be promoting one or the other. I would say that, you know, whatever works best for whoever the end user is. Yeah. At the end of the day, that matters the most.

Agree. Right. You can be in the healthcare industry, you can be in a different sector or a different industry that requires for you to use a different tool. That's what you should embrace.

Yeah, I agree. I agree. I might add in, you know, without naming anything, I think experimenting and trying different things and not landing on one and sticking with one would be my advice. I think AI tools are doing their job right.

But the key is educating everyone from prompt engineering perspective. Absolutely. Well, that's exactly right. They can't treat it just like another search tool and how powerful you're thinking about the prompts.

That's when really you can change. You can see the value of it. Yes, absolutely. I agree.

Dr. Quadri, thank you so much for this discussion. It was really just an exercise for me in learning. And I know that everyone that's listening gains so much value.

So thank you for being on the surgical journey and with us. Hopefully we can do this again. Thank you. Thank you so much.

And, you know, looking forward to staying in touch with, you know, smart people like yourself. I learned this note. Someone, one of my mentors told me that if you are surrounded by smart people, it only makes you look smarter. Yes.

Thank you. Yes. We appreciate you. Thank you, Dr.

Quadri. Thank you. Okay. I'm going to pause this for a minute and, oh, no, I don't want to leave the meeting.

It's not letting me do anything. Hold on. Give me one second.