Latest / Becker’s Healthcare Podcast / R. Kannan Mutharasan, MD, MBA, Medical Director of Health Network Development, Sarah Plaskett, PA, Vice President of Operations, and Kaleigh Powers, MSN, RN, Director of Operations at Northwestern Medicine Bluhm Cardiovascular Institute

Transcript

R. Kannan Mutharasan, MD, MBA, Medical Director of Health Network Development, Sarah Plaskett, PA, Vice President of Operations, and Kaleigh Powers, MSN, RN, Director of Operations at Northwestern Medicine Bluhm Cardiovascular Institute · Becker’s Healthcare Podcast

• 0:02
  - The playbook for running a successful ASC in
• 0:04
  - 2026
• 0:06
  - looks almost nothing like it did a few
• 0:07
  - years ago.
• 0:08
  - And the one for 2029
• 0:10
  - probably won't look like this one either.
• 0:12
  - Case complexity is increasing. Payer models are shifting.
• 0:16
  - Technology is creating new possibilities
• 0:19
  - and new cost pressures at the same time.
• 0:21
  - The spine and orthopedic leaders who are staying
• 0:24
  - ahead aren't just keeping up with the changes.
• 0:27
  - They are in rooms where the changes get
• 0:29
  - talked about honestly. Becker's twenty third annual Spine
• 0:29
  - Orthopedic and Pain honestly. Becker's twenty third annual
• 0:32
  - Spine Orthopedic and Pain Management Driven ASC Conference
• 0:36
  - is such a space.
• 0:37
  - More than 1,000 attendees,
• 0:39
  - 275
• 0:40
  - plus physician and ASC speakers, three days of
• 0:43
  - programming across the clinical,
• 0:45
  - operational, and strategic
• 0:47
  - with CME credits available.
• 0:49
  - Join us June eleventh through the thirteenth. For
• 0:52
  - the agenda and event details, visit Becker's hospital
• 0:55
  - review and click on the events tab in
• 0:57
  - the upper right.
• 1:00
  - Welcome to the Becker's healthcare cardiology and heart
• 1:03
  - surgery podcast. I'm Mariah Taylor, assistant editor, and
• 1:06
  - I'm thrilled to be speaking with two leaders
• 1:08
  - from Northwestern Medicine's Bloom Cardiovascular
• 1:11
  - Institute.
• 1:12
  - First, we have doctor Kennan Mutharison,
• 1:14
  - medical director of Health Network Development,
• 1:16
  - and we have Sarah Plaskett, vice president of
• 1:19
  - operations.
• 1:20
  - Thank you both so much for joining me
• 1:22
  - today. I'm really excited to learn more about
• 1:24
  - how you're thinking of building AI care pathways.
• 1:27
  - But before we dive into my questions, I'd
• 1:29
  - love for you both to introduce yourselves and
• 1:31
  - tell us a little bit about your about
• 1:32
  - your background.
• 1:33
  - Doctor Mathurison?
• 1:35
  - Thanks very much, Mariah, for having us. We're
• 1:38
  - very excited to talk about the implementation of
• 1:40
  - artificial intelligence into clinical practice.
• 1:43
  - I serve as medical director for Health Network
• 1:45
  - Development that looks like serving as diet partner
• 1:48
  - to Sarah Plasquette,
• 1:50
  - who is vice president of operations.
• 1:52
  - I work on strategy, physician recruitment, and building
• 1:55
  - research into
• 1:57
  - clinical arenas as well as artificial intelligence implementation.
• 2:03
  - Great. Thanks, Mariah, as well. I'm, privileged to
• 2:06
  - be here today in partnership with doctor Mutharason
• 2:08
  - to speak about the AI platform specific in
• 2:11
  - Northwestern Medicine. I've had the privilege of serving
• 2:13
  - as the vice president of operations and had
• 2:15
  - a great history here at Northwestern Medicine over
• 2:18
  - twenty five years. And I'll have to say,
• 2:20
  - when we think about AI, it's really that
• 2:22
  - territory of unknown, and we're excited to navigate
• 2:25
  - this pathway and how we can really impact
• 2:28
  - patient, patient care delivery,
• 2:30
  - and and develop an infrastructure that's really gonna
• 2:32
  - catapult us into the future of where we
• 2:34
  - see AI going. So thanks for having us.
• 2:37
  - Of course. Thank you both for being here.
• 2:39
  - I'm so excited to dive into this topic
• 2:41
  - more because
• 2:42
  - like you all, there's a lot of leaders
• 2:44
  - out there trying to figure out how to
• 2:45
  - make AI fit into their cardiovascular
• 2:48
  - programs. So how is AI reshaping your referral
• 2:51
  - relationships and your care coordination
• 2:53
  - between Northwestern and any affiliate or community hospital
• 2:56
  - partners?
• 2:59
  - So so the word you use interesting, which
• 3:02
  - was how do you make AI fit into
• 3:05
  - our cardiovascular programs?
• 3:07
  - And I really think about it as how
• 3:09
  - do we have AI
• 3:11
  - lever as a tool to unlock our cardiovascular
• 3:14
  - programs. What I mean by that is
• 3:17
  - we at Northwestern are really looking at delivering
• 3:20
  - cardiovascular care at scale,
• 3:22
  - and that looks like improving access, improving the
• 3:25
  - ease with which patients can
• 3:28
  - approach our clinical offerings,
• 3:30
  - improving the ease with which our referring doctors
• 3:33
  - can,
• 3:34
  - approach us and
• 3:35
  - making our testing better, more accurate, more predictive
• 3:39
  - of disease so that we can jump on
• 3:41
  - things earlier and help patients
• 3:43
  - navigate to better health sooner. So
• 3:45
  - when I think about artificial
• 3:47
  - intelligence, I think about
• 3:48
  - our current
• 3:50
  - paradigms of doing things,
• 3:52
  - disrupting those paradigms,
• 3:54
  - or at least improving and making more facile
• 3:57
  - those paradigms,
• 3:58
  - where where they're already working pretty well but
• 4:00
  - could go faster. And And so that in
• 4:02
  - term impacts how our referral patterns work. And
• 4:05
  - what we would
• 4:06
  - like to see is that
• 4:08
  - members of the community referring physicians
• 4:11
  - understand that we are highly enabled to take
• 4:14
  - care of patients quickly, rapidly, and get patients
• 4:16
  - from that initial concern
• 4:19
  - to
• 4:20
  - the best achievable health. And
• 4:23
  - to me then, AI is a way to
• 4:25
  - improve our engine for caring for cardiovascular patients.
• 4:30
  - Excellent.
• 4:31
  - Are you guys building any AI tools in
• 4:33
  - house?
• 4:35
  - So we have and we are.
• 4:37
  - For example, a major tool that we have
• 4:40
  - built is an advanced heart failure screener that
• 4:42
  - we have built
• 4:44
  - over the last four, five years. So any
• 4:46
  - patients who come into Northwestern Medicine,
• 4:49
  - the screener automatically evaluates the ejection fraction of
• 4:52
  - patients,
• 4:54
  - the
• 4:55
  - data that's available in clinical notes,
• 4:58
  - creates a score, runs in the cloud,
• 5:00
  - understands whether or not a patient is at
• 5:03
  - risk for
• 5:04
  - advanced heart failure and needs advanced heart failure
• 5:07
  - therapies like an left ventricular
• 5:10
  - assist device or transplantation
• 5:12
  - or perhaps
• 5:14
  - better intensification of guideline directed medical therapy. The
• 5:18
  - beauty of this system is that feeds back
• 5:19
  - right into our electronic medical record, and then
• 5:23
  - a nurse navigator will select whether or not
• 5:26
  - this patient is appropriate for referral to advanced
• 5:28
  - heart failure. So this is one example of
• 5:31
  - a home built
• 5:32
  - AI tool that has allowed us to care
• 5:35
  - for more advanced heart failure patients accurately.
• 5:38
  - And what we're finding now actually is that
• 5:40
  - it's very helpful
• 5:41
  - in identifying
• 5:43
  - advanced heart failure patients who may benefit from
• 5:46
  - palliative care interventions that we're also building at
• 5:48
  - Northwestern. That's just one example of a homegrown
• 5:51
  - tool that we've built.
• 5:52
  - Very cool.
• 5:54
  - Now on the opposite end of things, how
• 5:56
  - is your team approaching vendor evaluations and contract
• 6:00
  - structuring to ensure that you're getting the best
• 6:01
  - tools and ROIs for your hospital?
• 6:05
  - Yeah. Thanks, Mariah. It's it's a great question
• 6:07
  - because there are a lot of great platforms
• 6:09
  - out there I think the health care systems
• 6:10
  - are looking at. And here at Northwestern Medicine,
• 6:12
  - we try to take a very disciplined
• 6:14
  - outcomes based approach.
• 6:16
  - You know, we wanna make sure that we
• 6:18
  - also are providing
• 6:19
  - proof over promise. And so how do we
• 6:21
  - continue prioritizing a pile of a program or
• 6:24
  - maybe starting with a deployment of a limited
• 6:26
  - scope
• 6:27
  - to really validate that impact before rolling out
• 6:30
  - to the rest of the health network or
• 6:32
  - the rest of the system?
• 6:34
  - We clearly want to ensure that there's a
• 6:36
  - return on investment within the framework that we're
• 6:39
  - building, and I think that's important that we
• 6:41
  - have those levers and metrics that can support
• 6:44
  - the real, not only just patient impact, but
• 6:46
  - the financial impact that can help benefit the
• 6:48
  - healthcare system.
• 6:50
  - And, I think, furthermore, you know, how do
• 6:52
  - we utilize Epic within our electronic head of
• 6:55
  - medical records
• 6:57
  - to also integrate within a platform that already
• 7:00
  - exists? And, so, I think system integration focused
• 7:03
  - rather
• 7:05
  - than looking potentially at standalone solutions and interoperability
• 7:08
  - is something that's important for not only just
• 7:10
  - what that Ken and I are thinking about,
• 7:12
  - but also how do other healthcare systems think
• 7:14
  - about what vendors they might want to partner
• 7:16
  - with to continue providing solutions. And
• 7:19
  - And I think that's the most important is
• 7:20
  - we wanna invest in the solutions that demonstrate
• 7:22
  - and improve patient care and obviously have that
• 7:25
  - financial performance at the end of the day.
• 7:29
  - Absolutely.
• 7:31
  - Now for other health systems that are building
• 7:33
  - out their cardiovascular
• 7:34
  - service lines and want to use AI for
• 7:36
  - their throughput or their operations,
• 7:39
  - what areas do you think AI has the
• 7:41
  - highest potential for improvement, and where do you
• 7:43
  - think AI should be avoided?
• 7:47
  - Thanks, Mariah. Great question. I think as we
• 7:50
  - think about our AI journey or any clinical
• 7:53
  - implementation
• 7:55
  - improvement, what we find is always critically important
• 7:58
  - is to have
• 8:00
  - extreme
• 8:01
  - crystal clarity
• 8:03
  - around who we are, what our operations are,
• 8:06
  - what we're trying to do to support the
• 8:08
  - health of patients.
• 8:10
  - So who are we? We're an academic medical
• 8:13
  - center where
• 8:14
  - patients come first. So that is the center
• 8:17
  - of everything that we do. We center on
• 8:19
  - the patient. We want to improve that patient
• 8:21
  - care journey.
• 8:23
  - We want that to be something that we
• 8:24
  - have the ability to deploy at scale so
• 8:27
  - we can impact as many lives as possible
• 8:29
  - so patients can go from being a patient
• 8:31
  - to being a person, picking up their grandkids
• 8:33
  - because they're no longer short of breath,
• 8:36
  - being able to walk with a loved one
• 8:38
  - down the street, or grocery shop. That's what
• 8:41
  - we're here to do. In addition, we're here
• 8:43
  - to create new knowledge
• 8:45
  - and
• 8:46
  - discover. We're here as an academic health system
• 8:49
  - to generate
• 8:51
  - the ability to do new things for patients
• 8:54
  - to support that care journey.
• 8:56
  - So that clarity around who we are
• 8:59
  - and how cardiovascular care is delivered, which broadly
• 9:02
  - we think of as
• 9:04
  - a patient notices something or perhaps we notice
• 9:07
  - something about the patient and get to that
• 9:10
  - root cause and then the treatment as soon
• 9:12
  - as possible.
• 9:14
  - That's the beginning. That's step zero even before
• 9:17
  - you begin to say the word artificial intelligence.
• 9:20
  - Then it becomes very easy to understand how
• 9:22
  - artificial intelligence can help. And
• 9:25
  - anything that artificial intelligence touches
• 9:27
  - that helps you do any of those things,
• 9:30
  - whether it is supporting
• 9:32
  - a patient,
• 9:33
  - supporting a team member in working at top
• 9:36
  - of license,
• 9:38
  - supporting
• 9:39
  - the ability to predict predict disease so that
• 9:42
  - you
• 9:43
  - have that disease handled in a more expeditious
• 9:46
  - manner, whether it allows you to enroll patients
• 9:49
  - into a clinical trial more efficiently.
• 9:52
  - That's what is the benefit of artificial intelligence,
• 9:55
  - and that's exactly what we're focusing on
• 9:57
  - in our newly developed core laboratory
• 10:00
  - within our center for artificial intelligence,
• 10:03
  - what we're terming AI click or artificial intelligence
• 10:06
  - clinical implementation
• 10:07
  - core laboratory. And we're very excited to partner
• 10:10
  - with
• 10:12
  - patients,
• 10:13
  - with outside entities,
• 10:15
  - with each other within the health system on
• 10:16
  - this journey.
• 10:18
  - Yeah. No. I I think think thank you
• 10:20
  - for that overview. And, you know, I think
• 10:21
  - for me too, when it comes to an
• 10:23
  - operational
• 10:24
  - leader lens, it's important that how do we
• 10:26
  - think about some of this AI technology
• 10:29
  - to feed and support some of the back
• 10:31
  - end operations that can truly help identify and
• 10:33
  - create
• 10:35
  - information
• 10:36
  - available to the physician who can then really
• 10:39
  - spend that face to face time with the
• 10:41
  - patient
• 10:42
  - and and not do some of the work
• 10:44
  - that's being done to be able to ensure
• 10:46
  - that they have the right information available at
• 10:48
  - their tool at their fingertips. So I think
• 10:50
  - a lot of what AI when I see
• 10:52
  - opportunities within the the operations itself and, you
• 10:55
  - know, Northwestern Medicine, I think, has done a
• 10:57
  - great job with some of our AI deployment,
• 11:00
  - to date, and we're continuing to further vet.
• 11:03
  - The other opportunity when it comes to how
• 11:05
  - to support a cardiovascular service line with AI
• 11:08
  - is creating intelligence around scheduling opportunity. Access is
• 11:11
  - always so critically important for patients to make
• 11:14
  - sure that we're getting the right access for
• 11:16
  - the right patient at the right time.
• 11:18
  - And I think helping to understand how AI
• 11:20
  - can continue to support patient care delivery and
• 11:23
  - access is another really big focus topic, at
• 11:26
  - least for us here at Northwestern Medicine.
• 11:29
  - Absolutely.
• 11:31
  - Now we here at Becker's recently published a
• 11:33
  - few different stories about how hospitals are ending
• 11:36
  - a lot of their AI programs or
• 11:39
  - sunsetting some of the projects that aren't giving
• 11:41
  - them the ROIs they're looking for,
• 11:43
  - And fewer hospitals are starting new AI tools
• 11:47
  - or investing in that. So I'm curious for
• 11:49
  - you all. How are you evaluating your AI
• 11:52
  - tools to determine if they should be continued
• 11:54
  - or sunset?
• 11:57
  - The critically important question, especially in a space
• 11:59
  - that's so fast moving as AI. Whenever there's
• 12:02
  - a
• 12:03
  - a field that's moving very quickly,
• 12:05
  - traditional metrics around ROI don't always work at
• 12:07
  - first because so much of the return on
• 12:10
  - investment or or
• 12:11
  - interest
• 12:12
  - perhaps really comes from the intellectual engagement and
• 12:16
  - the learnings that the organization unlocks from just
• 12:19
  - being part of the journey.
• 12:21
  - At some point, though, you have to be
• 12:22
  - a little bit hard nosed and understand whether
• 12:25
  - or not there's true ROI because without that
• 12:27
  - margin, there is no mission.
• 12:29
  - So one of the things that we think
• 12:30
  - about is every time we make something or
• 12:32
  - partner with something, we with someone rather,
• 12:35
  - we also have to measure. We have to
• 12:36
  - measure what the impact is. And we understand
• 12:38
  - that at first, there might not be a
• 12:40
  - true hard dollar ROI.
• 12:43
  - But at some point, this has to be
• 12:44
  - something that sustains operations and drives it forward.
• 12:48
  - We're lucky enough that we are early enough
• 12:50
  - in our journey that I'm not aware, at
• 12:51
  - least from the Cardiovascular service line, of of
• 12:55
  - services that we've had to turn off because
• 12:57
  - of this. But there are some conversations we're
• 12:58
  - having
• 12:59
  - around whether or not the juice is really
• 13:01
  - worth the squeeze
• 13:03
  - on particular
• 13:04
  - AI implementations.
• 13:06
  - I'd be curious about Sarah's perspective on this
• 13:08
  - too.
• 13:09
  - Yeah. I I think that's right. You know,
• 13:11
  - making sure that we're very rigorous and continuing
• 13:13
  - an ongoing evaluation framework is is something that
• 13:17
  - is important
• 13:18
  - when we are looking at what really allows
• 13:20
  - us to achieve the desired outcome and goal
• 13:23
  - that we're trying to do that. And sometimes
• 13:24
  - that maybe
• 13:26
  - we may be aware of that early on,
• 13:27
  - and sometimes that may take a few more
• 13:29
  - trials to be able to really understand that
• 13:32
  - that value. And, you know, the financial modeling
• 13:34
  - is extremely important, but just the overall impact
• 13:37
  - of of efficiency within operations when I go
• 13:40
  - back to some of the operational comments made
• 13:42
  - earlier is is also important. And so continuing
• 13:45
  - to develop that framework and the infrastructure
• 13:48
  - and and as well as ensuring we have
• 13:50
  - the right
• 13:51
  - opportunity
• 13:53
  - to sup be supported in the health care
• 13:54
  - system outside of just the cardiovascular service line.
• 13:57
  - So we've really been partnering with our innovation
• 13:59
  - teams and with our analytics teams to be
• 14:02
  - able to have a really multi specialty and
• 14:05
  - multi disciplined approach. And how do we think
• 14:07
  - about AI for cardiovascular service line? And and
• 14:10
  - so these conversations
• 14:11
  - are continuing to be ongoing and and vetted,
• 14:14
  - and I think that's also important when you
• 14:16
  - think about sustainability
• 14:17
  - within the AI framework,
• 14:19
  - within the health care system.
• 14:22
  - Amazing. Well, thank you both for being here
• 14:24
  - on the podcast today. This has been such
• 14:26
  - a great discussion.
• 14:28
  - Great. Thanks for having us. Thank you so
• 14:29
  - much for having us.