Kiran Mysore, Chief Data and Analytics Officer at Sutter Health, speaks with Vasant Eswarachari, Partner at Guidehouse, about data integration, agentic architecture, and the role of empathy in AI, responsible AI integration, addressing physician burnout, and blending empathy with innovation.

Data integration, agentic architecture, and the role of empathy in AI

Speaking of data integration, Mysore highlights a fundamental shift in how healthcare systems are beginning to harness and synthesize diverse forms of data. This integration is not just a technological feat, as it has direct clinical implications.

According to him, this unified approach to data enables a longitudinal view of the patient, which supports:

• Elevating the level of personalized medicine
• Assessing patient risk more effectively
• Designing proactive care plans
• Reducing the burden on both patients and providers
• Ultimately improving quality of care and patient well-being

The promise of agentic architecture

Another promising frontier Mysore emphasizes is agentic architecture, a concept still in its early stages but one he sees as vital to the future of healthcare efficiency.

“It’s still very early, but I see tremendous opportunities to improve things like operational efficiency through an agentic approach.” He states that realizing this potential will require foundational work:

• Establishing effective data systems
• Ensuring systems can interoperate
• Building consistent, predictable workflows
• Maintaining a human-in-the-loop approach

While there’s excitement about AI’s integration into healthcare, Mysore also acknowledges existing concerns. “As AI starts to get integrated into the healthcare system, there is a lot of openness, obviously, to start using this. But there’s also a little bit of skepticism.” He stresses the importance of deploying AI with the right application and governance frameworks and building and maintaining trust.

This, Mysore believes, is key to achieving positive outcomes for both patients and providers.

Addressing physician burnout with ambient tools, GenAI, and empathy

Highlighting physician burnout, he says, “Nearly half of U.S. physicians are saying they’re burnt out, and this affects the quality of care.” To combat this, Mysore mentions the use of ambient documentation and in-basket messaging automation. These tools not only reduce workload but also ensure more accurate capture of patient information, which in turn enhances doctor-patient communication, addresses patient concerns, and reduces the cost of care.

Further, Mysore cites generative AI’s (GenAI’s) role in supporting radiologists by transcribing interpretations of X-rays and other medical images.

“Radiologists are another category of healthcare professionals who are in high demand and short supply. And so if we can help with their cognitive load, we can lower the cost of care.”

But beyond efficiency, Mysore doubles down on empathy as he says, “Healthcare is all about empathy.” He elaborates on efforts at Sutter Health to ensure AI-generated messaging is both clinically sound and emotionally resonant. This involved training models not just with physicians but also with patient advocates and even customizing empathy levels at the physician level.

“We did a lot of work with our in-basket messaging work to understand how to make sure the messages that are suggested have the right level of empathy.” Mysore sees the future of healthcare as one where AI and data are tightly woven into the fabric of care, yet always with a human touch.

Building a connected, patient-centered future

Delving deeper into the future of healthcare, Mysore reiterates the need to harmonize two powerful forces — technology and human empathy. It is critical to scale successful pilots and translate them into a broader, real-world impact. He therefore points to the need for predictive and preventative care at scale, supported by comprehensive data integration across systems.

“I would love to see us continue to harness data from all of our systems and all of the patients to find treatment that uniquely works for one patient,” says Mysore. He highlights ongoing efforts at Sutter Health to strengthen research capabilities by learning from both internal data and external systems, merging human insight with machine intelligence.

In continuation, Mysore also calls for greater collaboration across the healthcare continuum, with a particular focus on enabling interoperable, patient-controlled longitudinal records.

“We want to build a connected ecosystem where the journey of the patient is fully understood from the beginning and where we are able to offer the best quality of care at the time and place that the patient needs it, through technology, the use of AI, and the use of available digital technologies and tools.”

Laying the groundwork for scalable, responsible AI in healthcare

Further, Mysore notes that AI success hinges on much more than technology alone. “AI has tremendous potential for us in healthcare. But it’s not just the technology; it’s not just the algorithms. It is a combination of many things that include the right application of AI in the workflow.”

He outlines a comprehensive framework for AI adoption that includes

• Consistent processes
• High-quality, trusted data
• Robust governance structures
• Ongoing monitoring to prevent model drift or bias

Furthermore, Mysore states that success must be measured both quantitatively and qualitatively. He adds, “Success of AI ultimately is also measured by input and output metrics.”

In conclusion, he underscores the need for effective scaling, taking pilot successes, and expanding them across the diverse and complex settings of healthcare delivery.

CDO Magazine appreciates Kiran Mysore for sharing his insights with our global community.