Harnessing AI and Big Data to Transform Diagnostics with Dr. Christopher Garcia of Mayo Clinic Laboratories

Summary

---

Harnessing AI and Big Data to Transform Diagnostics – Dr. Christopher Garcia, Mayo Clinic Laboratories

Podcast: Becker’s Healthcare Podcast
Episode Date: September 2, 2025
Guest: Dr. Christopher Garcia, Chief Digital Innovation Officer, Mayo Clinic Laboratories
Host: Brian Zimmerman

---

Episode Overview

In this episode, host Brian Zimmerman sits down with Dr. Christopher Garcia to discuss how artificial intelligence (AI) and big data are revolutionizing diagnostic medicine. Dr. Garcia explores the evolving landscape of data-driven diagnostics, recent breakthroughs in AI applications, the promise of democratized expertise, and the balance between innovation and patient safety. He also shares his hopes for the field’s future, emphasizing equity, rigor, and real-world impact.

---

Key Discussion Points & Insights

1. Defining Data-Driven Diagnostics

[00:24–01:56]

• Complexity Management: Dr. Garcia defines data-driven diagnostics as the use of computational tools to tackle complexity previously impossible to manage.

• Three Major Transformations:

  • Reactive to Predictive: Leveraging patterns in large datasets to identify patient risk before symptoms appear.
  • Siloed to Integrated: Combining laboratory results with imaging, clinical notes, genomics, and social determinants to achieve more holistic patient insights.
  • Population Averages to Precision: Shifting diagnostics from broad population benchmarks to individualized baselines and health trajectories.

“To me, data-driven diagnostics is fundamentally about using computational tools to manage complexity that was previously impossible to handle, unlocking new analytic capabilities that bring genuinely new answers for patients.”
– Dr. Garcia [00:42]

---

2. Recent Transformations in Diagnostics

[01:56–04:22]

• AI’s Rapid Advancement: The rise of foundational models and multimodal AI has enabled robust tools for a range of problems, with platforms like ChatGPT gaining mainstream adoption.
• Accessibility: Improvements in cloud computing and algorithms have democratized AI benefits, increasing availability for all practitioners.
• Regulatory Maturity: Authorities like the FDA have begun crafting frameworks for AI-driven diagnostics, promoting safe integration into clinical practice.

“The big move from like small narrow AI models to providing solutions to lots of different problems with the same tools is pretty amazing.”
– Dr. Garcia [02:54]

---

3. What’s Most Exciting in the Field

[04:48–06:18]

• Expertise Democratization: AI is bridging gaps by delivering subspecialist-level insights to community hospitals and underserved areas.
• Continuous Learning Systems: Diagnostics that improve with each use open the door for smarter healthcare tools.
• Human-AI Collaboration: Rather than replace clinicians, AI augments and partners with healthcare professionals, enhancing care.

“Using AI, data, smart communication tools to bring subspecialist level expertise to community hospitals to underserved areas, really making those expert diagnostic insights universally accessible is something that I'm really passionate about.”
– Dr. Garcia [05:03]

---

4. Addressing Equity and the Clinician Shortage

[06:18–06:55]

• Expanding Access: AI tools are seen as essential for addressing healthcare disparities and shortages of clinicians.
• Goal: Universal access to advanced diagnostics, leveling the quality of care across regions and populations.

“There is definitely a need and there is a disparity in the equity of health care throughout this country and around the world. It would be really lovely to have tools to help level set that.”
– Dr. Garcia [06:41]

---

5. Balancing Innovation and Patient Safety

[06:55–08:38]

• Rigor and Validation: Despite rapid advances, diagnostics must maintain high standards: rigorous validation, clinical evidence, and demonstrated impact on outcomes are non-negotiable.
• Value Demonstration: Tools must provide concrete benefits for adoption and reimbursement.

“Even though cutting edge is very exciting, it doesn't exempt any of these tools or these methods from needing to prove their value. ... The safety, the validation, the rigor, all of those things need to be integrated in the life cycle of the development of these tools.”
– Dr. Garcia [07:47]

---

6. The Next Five Years: Predictions and Hopes

[08:55–10:56]

• AI as Necessity: Dr. Garcia hopes AI will transition from novelty to critical infrastructure, proven to deliver essential value in patient care.
• Equity in Access: Diagnostic excellence should be available regardless of geography or resources.
• Trust and Teamwork: A vision for deep integration between human clinicians and AI, grounded in trust and clear understanding of strengths and limitations.

“I really hope that people look at AI not as a curiosity or novelty, but as a necessity. … These not only provide genuine value, but are really necessary for providing the best patient care possible.”
– Dr. Garcia [09:07]

“How do we integrate it into a team so that we all make better decisions while still having a clear understanding of capabilities and limitations?”
– Dr. Garcia [10:18]

---

Notable Quotes & Memorable Moments

• On AI’s shift from hype to practicality:
  “AI is no longer seen as maybe a passing fad or a curiosity, but it really is starting to be considered as a practical tool, depending on your job.” [03:24]

• On continuous learning diagnostics:
  “Diagnostics that improve with every test performed … that is really powerful.” [05:39]

• On balancing innovation and safety:
  “Patient safety still comes first. It's paramount.” [07:23]

---

Timestamps for Key Segments

• Data-driven diagnostics defined: 00:24–01:56
• Recent changes & AI’s evolution: 01:56–04:22
• Exciting new frontiers: 04:48–06:18
• Addressing equity and shortages: 06:18–06:55
• Balancing innovation and rigor: 06:55–08:38
• Looking to the next five years: 08:55–10:56

---

Conclusion

Dr. Garcia provides a sharp, forward-looking perspective on the integration of AI and big data in diagnostic healthcare. He champions the democratization of expertise, insists on rigor and validation despite the allure of breakthrough tech, and calls for a future where human and AI collaboration leads to better, more equitable patient outcomes. This episode offers valuable insight for anyone curious about the real-world trajectory of AI in healthcare diagnostics.

---

Transcript

[A]

Hi, everyone.

[B]

This is Brian Zimmerman with Becker's Healthcare. Thank you so much for tuning into the Becker's Healthcare podcast. Today we're going to talk about harnessing AI and big data to shape the future of diagnostics. Joining me for this conversation is Dr. Christopher Garcia, Chief Digital Innovation Officer with Mayo Clinic Laboratories. Dr. Garcia, thank you so much for being here.

[A]

Thank you so much, Brian, for having me. Excited to be here.

[B]

Yeah, Excited to get into it. Let's, I guess begin with sort of a laying a foundation type of question, sort of setting the stage here. I'm curious to hear how you would define data driven diagnostics. I think folks might have some varying definitions of that. Curious to hear how you think about it.

[A]

Great question. To me, data driven diagnostics is fundamentally about using computational tools to manage complexity that was previously impossible to handle, unlocking new analytic capabilities that bring genuinely new answers for patients. So it'd be kind of in three big categories. First would be using data and computation to move from being reactive to predictive, using patterns in large data sets to identify and stratify risk before symptoms appear. Or the second one would be moving from being siloed to integrated. So combining lab results with other modalities like imaging and clinical notes and genomics, social determinants to provide better insights for patients where they're at now. And the last big transformation for me when it comes to data driven diagnostics is moving from population averages to precision. So really getting that personalized baseline and trajectory for each patient as opposed to how we normally practice still today, which is a very large population. Benchmarks.

[B]

Excellent. I appreciate you sort of laying a foundation there for this conversation as we get a little bit deeper here and it won't surprise you, I think my next question here is to ask you about what's changed the most in the last two to three years. And I imagine a big component of that is AI, right? We talk a lot about AI today, but this seems like particularly ripe ground for AI to make a difference. Am I reading that correctly? And also just. Just what other transformations do you want to shout out that have really occurred over the last two to three years?

[A]

No, Brian, I think I am like you there in thinking about AI. It really has made such a large leap in the last two to three years. The development and growth of foundation models and multimodal AI has been huge. I think all of us recognize companies like ChatGPT now, where three to four years ago most of us had never heard of that before. And there are those of us who are using these kinds of tools every day. So that, that's huge. And I mean, I work a lot in AI in my career and have trained and worked with folks in developing solutions in the lab around AI. And the big move from like small narrow AI models to providing solutions to lots of different problems with the same tools is pretty amazing. And yeah, I think I'm really with you thinking about AI, not only are there new tools, but cloud computing and improved algorithms mean that it's more available for everyone than it was beforehand, which is huge. And I think also around AI, there's a big shift in public perception. AI is no longer seen as maybe a passing fad or a curiosity, but it really is starting to be considered as a practical tool, depending on your job. I think you're feeling it more now than you did beforehand. So, yeah, there's just a ton going on around AI. I would also say around data and AI, there's been a lot of focus and maturity. On the regulation side, the FDA and other groups have really been looking at how do they tackle these new devices and tests that integrate these methodologies. So there's just so much going on, which is exciting.

[B]

It certainly is exciting and there is a lot going on. And it's sort of profound to think about how much transformation has occurred over the last two to three years, to your point. But it's also, it seems we're still in the early innings here. To your point too, about trying to figure out data and regulation, like, there's a lot of work to be done, a lot of advancements to come, and you mentioned excitement. I guess my question for you would then be what excites you the most about this is heading?

[A]

Well, absolutely wanted to tackle that, but also wanted to agree with you about it being early days. There still is a huge gap between what's possible in research and what's being used every day. But looking at that gap and saying what is most exciting, especially practically, I really think it's that democratization of the expertise. So using AI, data, smart communication tools to bring subspecialist level expertise to community hospitals to underserved areas, really making those expert diagnostic insights universally accessible is something that I'm really passionate about. And I know a lot of folks that I work with are as well. Another thing that's really exciting is the idea of continuous learning systems. So these diagnostics that improve with every test performed. And like you said, there's a lot to be done on the regulation side in how we handle those, but that is really powerful. And I think that's super exciting. And the other thing that I wake up a ton and work with my teams on is really that human AI partnership. So not replacing pathologists or laboratorians, but really augmenting us. Collaborating with them is really an exciting movement. And knowing how to do that correctly and to provide as much value is something I'm really excited about.

[B]

And to your point, Dr. Garcia, there's a really deep need for that. Expanding access to that kind of specialty care, that kind of expertise, especially looking at the clinician and physician shortage. Am I right about that? Is that one of the reasons why you're most excited about this?

[A]

Absolutely. I mean, I think as new technologies can really and treatments are out there, we want to make sure that we all have access to that. So there is definitely a need and there is a disparity in the equity of health care throughout this country and around the world. It would be really lovely to have tools to help level set that.

[B]

And of course, Dr. Garcia, all that's very exciting. But there's also a balance that needs to happen here between innovation and diagnostics. A lot of exciting potential, but ultimately clinical adoption matters here. Real world practical utility really needs to come into play here. Thinking about patient safety, of course. So what are your thoughts there on striking that balance?

[A]

It's a great question. I have always thought that even though that this is AI and very exciting and very buzzworthy, gravity still exists. Just as you're saying, you know, the patient safety still comes first. It's paramount. And not only is it important to take care of the patient, that like you said, we have to sustain the infrastructure, the talent, the expertise, everything that it takes to be able to create these tools. And to do that, we have to provide value and also get reimbursed for those. And so I have always, or I'm learning as we work in this space and work with others, that even though cutting edge is very exciting, it doesn't exempt any of these tools or these methods from needing to prove their value. And we have to think about these solutions just as we think about other diagnostic tests. They require rigorous validation, clinical evidence generation and demonstration of real world impact on patient outcomes. So I think that's one of the major things that I've taken away from balancing is still continue to move forward, but make sure that at the same time you're taking care of these other needs. The safety, the validation, the rigor, all of those things need to be integrated in the life cycle of the development of these tools.

[B]

All right, so Dr. Garcia, I'm going to ask you a forecasting question. I know no one has a true idea of what the future is going to bring, but I'll put it to you this way. Looking ahead five years from now, what do you hope will be true about how diagnostics work, feel and influence patient care?

[A]

It's a great, thought provoking question. I personally have a few hopes. The first one I said that we're heading this way, but I really hope that people look at AI not as a curiosity or novelty, but as a necessity. So really that it's moved along far enough in five years that we've proven the value and these not only provide genuine value, but are really necessary for providing the best patient care possible. That's one thing I would really hope that these new tools and methods really ingrain themselves in being critical for patient care. Another one that I'd really hope for is that as we mentioned, we're able to democratize that expertise and drive equity. So looking for those use cases that really bring that expert level diagnostic capability, regardless of geography or resources, to be able to significantly improve health care equity, that would be really my hope. If I could choose something, I would love to see that and be a part of that. And I think the other part that I would really I'm looking forward to see is how do we build that trusted partnership with these kinds of tools? AI is exciting, but it also needs a manager if you're going to be in there. So how do you treat these tools and these capabilities in a way that is trusted but still needs to be verified? And how do we integrate it into a team so that we all make better decisions while still having a clear understanding of capabilities and limitations. In five years, I'm really hoping that these nascent areas that we're talking about get more fleshed out and are understood at a wider level than just those of us who wear the nerd hat of Chief Digital Innovation Officer or medical director of AI?

[B]

Dr. Garcia, thank you so much for your time. I truly appreciate you coming on the podcast.

[A]

Thank you. Thanks for having me. It was very nice spending this time with you.

[B]

We also want to thank our podcast sponsor, Mayo Clinic Laboratories. You can tune into more podcasts from Becker's Healthcare by visiting our podcast [e-mail].