Sean Shoffstall, Head of AI, Innovation and Data at PaceMate – Interview Series

Sean Shoffstall, Head of AI, Innovation and Data at PaceMate, is a technology and product executive with 20+ years of experience building innovative SaaS solutions and integrating AI technologies that deliver measurable business outcomes and align with strategic business objectives.

Sean specializes in normalizing artificial intelligence, healthcare technology, and data-driven platforms for teams and customers, enhancing productivity and efficiency in clinical workflows. His previous work pioneering AI integration in healthcare management systems delivered significant efficiency improvements and enhanced patient outcomes through intuitive design and data-driven insights.

As a thought leader and public speaker on AI for executives, Sean bridges the gap between technical capabilities and strategic business implementation, focusing on product management and development that serves real clinical needs. He builds teams that deliver innovative solutions addressing genuine healthcare challenges rather than just pursuing technological trends, ensuring alignment with operational excellence and improved patient outcomes.

PaceMate is a healthcare technology company that offers PaceMateLIVE, a cloud-based platform for remote cardiac monitoring and data management. The system integrates data from implantable cardiac devices, ambulatory monitors, and consumer ECGs, using automated prioritization to help clinicians focus on the most critical alerts. It supports interoperability with major electronic health record systems such as Epic, Cerner, and athenahealth, streamlining clinical workflows, improving operational efficiency, and enhancing patient care continuity in cardiac practices.

You’ve built AI-driven healthcare solutions for over 20 years, including the first HIPAA-compliant AI audiogram tool at Auditdata and now the cardiac intelligence platform at PaceMate. What inspired your transition into using AI to transform clinical data into actionable insights?

I’ve always been driven by data. In my early career in digital marketing, my agency’s tagline was “Quantifiable Creativity” – the idea that you can drive emotion and creativity through what you learn from data. When I made the transition into healthcare, I saw a whole new way to use data. Instead of simply reporting what had already happened to a patient, I began to wonder: could we use data to see trends? Could we predict what might happen next? 

Then AI came along and expanded those possibilities. I really believe that if we can use AI in a smart, Human-in-the-Middle way, we can transform healthcare. It can give physicians more time with patients while the mundane tasks around data are handled by AI. We aren’t quite there yet, but the foundation is being built to get us there.

PaceMate manages one of the largest and most robust cardiovascular datasets in the world. How is this data being transformed into predictive algorithms that improve both patient outcomes and clinical efficiency?

Data is the key. To get the most power out of AI and Machine Learning, it all comes down to the training set you have. PaceMate is an industry leader not just because of our data, but because our team has decades of clinical expertise to interpret and validate it. When you combine the training data with that expertise, you start to see how saving a minute here and two minutes there can add up to huge gains in efficiency.

Then, when you layer in the ability to identify trends in patient data over a device’s lifetime, you’re empowering physicians with the insights they need to make the best decisions for their patients—backed by the most comprehensive data possible.

Regulatory frameworks like HIPAA and FDA approvals often slow AI adoption in healthcare. What practical steps can organizations take to innovate responsibly within these constraints?

At PaceMate, we strongly believe in the Human-in-the-Middle philosophy around AI. We don’t offload decisions to AI, which is exactly why the FDA requires serious oversight. Instead, organizations can use AI to mine, organize, and present data while still relying on clinical expertise to determine what’s right for each individual patient.

As for HIPAA, patient privacy is a serious concern in healthcare, and we should always put it first. That’s what’s great about our human-centric approach—AI doesn’t need to know anything about our patients’ personally identifiable information (PII). We can de-identify data and analyze from that. But even with the best philosophy, using the right HIPAA-compliant tools—which have come on the market in the past few years—is essential as well.

Data privacy is always a multilayered approach, and patients’ privacy should always be put first, even ahead of innovation.

Data sensitivity is a major concern in cardiac care. How does PaceMate ensure that AI development maintains the highest levels of patient privacy and trust?

Patient privacy has always been a cornerstone of PaceMate, and the rise of AI only reinforces why that commitment matters. We approach AI development with a “privacy-by-design” philosophy, meaning that data protection isn’t an afterthought, it’s built into every stage.

All patient data used in AI training is rigorously de-identified and encrypted, following protocols that exceed HIPAA requirements. We also follow strict data minimization principles, only collecting and processing what’s absolutely necessary for clinical value.

In cardiac care, we’re entrusted with some of the most intimate health data imaginable. That’s why we conduct regular privacy impact assessments and third-party security audits, because earning and maintaining trust isn’t a one-time effort—it’s a daily responsibility. 

Automation in healthcare can be a double-edged sword. How is AI at PaceMate designed to complement—rather than replace—the expertise of clinicians?

Our Human-in-the-Middle philosophy means that AI is used as a complementary tool, never replacing expertise. Our commitment to using in-house clinical expertise and working with clinicians in hospitals to guide our product development directly shapes our AI practice.

We ask questions like, “What would make you more efficient?”, “What data would help you make better decisions?” and “What’s missing from your current workflow?” We then use those insights to guide how we implement automation and AI without getting in the way.

Many health systems struggle with fragmented data across devices, EHRs, and monitoring tools. How does your team approach unifying these sources to deliver real-time insights that truly matter at the point of care?

Over the last 10 years, we’ve become the hub for data in remote monitoring, and we understand the patient data flow inside and out. You can think of it as an identity graph where data is gathered and served in the increments needed at that point in time. We’ve built the infrastructure to pull from multiple device manufacturers, their remote monitoring platforms and EHR systems, then normalize and contextualize that data so it’s actually useful at the point of care.

The key is understanding not just how to collect data, but when and how to present it. We’ve also become experts at aggregating broad, de-identified data to validate trends and ensure quality when we bring in new data sources. This dual capability—delivering personalized, real-time insights while maintaining the big picture view—is what allows us to turn fragmented data into actionable intelligence.

With your experience across multiple health tech startups, what are some of the most overlooked challenges when integrating AI into clinical workflows at scale?

Privacy and security are intimidating for many smaller healthcare technology companies. But there are great best practices and tools available, and all of the major cloud providers—AWS, Azure and Google Cloud—have advisors with guidebooks and checklists to help startups and enterprises tackle these challenges.

Once that is handled, data at scale becomes the next hurdle. Where you are today with your dataset is going to be completely different in six months. Understanding how to leverage structured and unstructured data with a strong identity graph can be a good foundation to start with, and documenting your approach along the way will help prevent some grey hairs down the road.

Ethical and transparent AI deployment is becoming a defining factor in healthcare innovation. How do you build accountability and explainability into AI systems used for medical decision support?

This goes back to our Human-in-the-Middle philosophy. We design our AI to present insights and patterns, but the clinician always makes the final decision. That creates a natural accountability layer where there’s always a licensed professional reviewing and validating what the AI suggests.

We also focus on showing the “why” behind AI recommendations. Our systems highlight which data points drove a particular insight, so physicians can evaluate whether it makes clinical sense for their specific patient. We’re not asking clinicians to trust a black box—we’re giving them transparency into the logic.

Since learning goes both ways, we’ve also built in continuous feedback loops. When a clinician accepts or overrides an AI suggestion, that informs our model improvement. This creates accountability in both directions and, over time, the AI learns from clinical expertise while clinicians can see how their input directly shapes the system.

Ultimately, explainable AI in healthcare isn’t just about technical transparency, but also about respecting clinical judgment and creating tools that augment rather than obscure the decision-making process.

As someone who regularly advises executives on AI strategy, what mindset shifts are most critical for healthcare leaders looking to move from experimentation to meaningful adoption?

When I talk to business leaders about AI, I try to drive them to a “What if?” or an “I wish…” mindset. One of the most powerful tools in AI is the ability to tap into the world’s view of a problem instead of just the people around you or your own biases. So whenever you think something is difficult or impossible, use your favorite AI to say “I wish I could…” and describe the things in your way. That’s super powerful. I also guide them to have their AI LLM be inquisitive. Tell it your problem, but also tell it to ask you detailed questions. Sometimes, that alone will drive you to some really unique solutions.

AI is also a great tool for planning. Executives need to set a strategy while also creating a plan of action, determining how to measure success, and identifying pitfalls before they happen. AI is great at helping to put together plans to get started.

Lastly, we’re always reading about how AI is taking jobs and companies are laying people off in favor of AI. I think that’s a bad way to think about AI. A company that’s built on products or services for people needs people to understand what they’re offering. Instead of using AI for cost reduction, think about how AI can take the mundane tasks from your people and become an amplifier. If you can reduce costs by 15% or you can increase productivity by 200%, which is the better business decision?

Looking five years ahead, how do you envision AI reshaping the landscape of cardiac monitoring and preventive medicine—and what milestones do you hope to achieve at PaceMate during that time?

AI is going to be more and more part of our work and daily lives over the next five years. As our users become more comfortable and trusting of the output, it opens up endless possibilities and partnerships.

I would love to first provide tools that help clinicians prioritize care for patients most in need—those who may be at highest risk of an adverse event. From there, we can start to use AI and trend data to show different possible outcomes for individual patients, giving physicians a clearer picture of what’s ahead.

Lastly, when we can deliver insights that everyone trusts, finding great partnerships with research hospitals to help them leverage our models would be a lofty goal. That’s where we could really drive innovation across the industry faster and make a meaningful impact on cardiac care as a whole.

Thank you for the great interview, readers who wish to learn more should visit PaceMate.