During a panel discussion at HPN San Diego, leaders from Mayo Clinic and Hoag Health shared how they are transforming care through digital innovation while still maintaining clinical excellence and financial sustainability.

This conversation highlighted how health systems can leverage AI, data, and integrated care models to enhance access, improve outcomes, and develop scalable, patient-centered solutions.

Here are the key takeaways…

Discussion on innovation in healthcare

Kicking off the discussion, Sameer Keole, MD, Medical Director Provider Market at Mayo Clinic Platform’ shed light on the current healthcare system, comparing the growing volume of medical records to the explosion of streaming subscriptions.

‘I remember when I had one cable TV subscription, and now I’ve got 10 subscriptions that are on my credit card bill each month, from Netflix to Hulu to Disney Plus, and stuff that I haven’t even heard of that somehow just magically appears. I think for clinicians, especially for the practicing clinicians in the room… it’s this whole arc of how a patient enters your system. It’s this funnel of medical records, which just grows by the day, and sometimes it’s as many as a couple hundred, even 2,000 pages.’

This overwhelming volume of patient data can make it difficult for clinicians to balance patient interaction with administrative tasks like reviewing records, ordering tests, and navigating prior authorizations.

Sameer acknowledged the significant impact ambient listening has made when it comes to workflow efficiency, commenting, ‘I do think ambient listening has been a significant addition to the clinical workflow in the last few years. I know it’s not everywhere, but I think most places that are using it recognize it’s here to stay.’

Kathy Azeez Narain added her perspective, focusing on the broader use of AI in healthcare, and how to navigate a fragmented system.

She explained how her team at Hoag is currently evaluating whether to build or buy solutions.

‘One of the things we are taking a stronger look at is that we should be building some things. We see the value where AI can take us, but the more fragmented we are in point solutions, the harder it becomes to really get to that full journey for the patient… that’s been a big strategic area and focus.’

She also spoke about women’s health being a major area of focus, highlighting how Hoag are supporting women through life’s stages – not just through illness.

‘We built a product and platform called Nona, and it’s really about taking the relationship from that transaction of having a baby and moving it into a more life journey. For your health and for moms and women especially, being head of household, it’s busy. So, we’ve seen that product and platform continue to get advanced and really be attractive for our women’s market.’

This platform reflects a shift toward preventive and continuous care, especially for women who are often the primary decision-makers for health and care in their households.

Data management and AI in healthcare

When it comes to implementing AI – data is key. As Kathy explained, ‘Data is just one of those things – you can’t really do any innovation without it.’

Speaking about her experience in other industries, Kathy noted that many organizations have spent years grooming their data to make AI truly effective. In contrast, healthcare remains heavily fragmented. So, what can healthcare systems do to make sure they’re getting the most out of these new tools and technologies?

For Hoag, having a strong data layer is key. As Kathy explained:

‘Within our ecosystem, we are actively looking at that data layer as core. We’re not going to make a decision unless we know that we can unify some of that data. But it is a long and hard journey because you’re dealing with so much fragmented data. So, from a strategic perspective, it’s critical to start doing that within our ecosystem.’

Although large amounts of data can come with challenges, it also presents a number of possibilities. Sameer talked about how applying AI to real-world data has enabled the brain tumor group at Mayo to uncover insights they otherwise wouldn’t have found.

’Because we have the large data set and because now the AI tools are much more powerful, we found a very interesting observation that has a low p-value. So, it could be due to chance, but it’s probably pretty unlikely. One particular anti-seizure medication called Lamotrigine was associated with a higher survival.

So, in general, seizures are associated with higher survival and GBMs, and this may be paradoxical, but it’s because it leads to an earlier diagnosis. Then if it’s something deep seated, it can grow… So, seizures in general are associated with a better outcome in GBM because it allows you to diagnose them earlier and get going with treatment.

But even amongst all those patients, there was one particular class of drugs that seemed to indicate a much higher rate of survival. I would call this hypothesis generating… there’s a few theories on why this could be, but it’s going to lead us down a path, and this could be theoretically practice changing, because if we find out that one particular seizure medication is more effective than others in terms of the end outcome, obviously that would change things.’

Sameer praised AI’s ability to uncover these types of patterns, stating, ‘We wouldn’t have been able to do this type of research five years ago, and certainly not 10 or 20 years ago.’

Future of AI and medical education

AI has made great strides in recent years, so why are there still clinicians who are hesitant to embrace it? As Sameer explained, ‘I think there’s this fear that we’re going to replace physicians with AI. I do believe in that adage; physicians aren’t going to be replaced by AI; they will potentially be replaced by a physician who uses AI.’

While AI has the ability to increase efficiency and uncover insights – there are many things it can’t do. As Sameer put it, ‘For a patient with a head and neck cancer, there’s never going to be anything that substitutes putting a camera down the throat and looking at the vocal cords, right? There’s no AI that’s going to do that.’

To future proof their careers, clinicians need embrace these technological advancements so they’re able to re-focus their attention on the parts of the job that AI can’t reproduce.

Sameer also highlighted the impact AI is having on medical education, commenting:

‘I think the overall trajectory of medical school is certainly going to get tweaked. Most medical schools are already changing their curriculum to reflect that the medical students have to be better versed in data and how to understand data and where to use it and I do think it’s going to impact how we distribute physicians in the future.’

Join our future discussions

A big thank you to our speakers Sameer and Kathy for sharing their insights. 

If you want to take part in future discussions, register your interest to join us at HPN USA 2026!