AI Insights: Cardiovascular Care
Cardiovascular diseases remain the leading cause of death worldwide and rising expense, making this area ripe for AI-enabled innovations. This AI Update includes discussions from the 2019 World Medical Innovation Forum, an interview from Licensing Manager Teja Reddi, PhD and an “In the News” article featuring the Apple Heart Study from Calum MacRae MD, PhD.
2019 WMIF | Cardiovascular Care: Reinvented Through AI
Teams are pursuing a range of AI-based tools in cardiovascular medicine: including AI-powered drug discovery and diagnostics to automated cardiac image analyses and AI-guided care delivery pathways. Panelists from the 2019 World Medical Innovation Forum will discuss where AI is having a sizeable impact. The discussion will also include the perspectives of a patient who benefited from AI-enabled cardiovascular care.
Calum MacRae MD, PhD on how AI is breaking down barriers in healthcare
Calum MacRae, MD, PhD, Vice Chair for Scientific Innovation, Department of Medicine, BH shares his thoughts on how artificial intelligence is breaking down barriers in cardiovascular care, and healthcare as a whole.
Sebastian Guth, PhD on the promise of AI
Sebastian Guth, PhD, President, Bayer Pharma Americas Region, Bayer discusses artificial intelligence and its potential ability to tailor patient care and drive efficiencies.
Interview with Teja Reddi, PhD, Licensing Manager, Innovation
- Please describe some of your work with the Cardiovascular research teams at Partners.
- Research geared towards therapeutics and diagnostics remains a large proportion of the work being done here; that said, there is a trend towards integrating algorithmic approaches in the identification and stratification of patients into pathophenotypes (can we define this, such as “customized roadmaps” to personalize patient care.
- For example, cardiopulmonary exercise testing (CPET) is a continuous hemodynamic and functional assessment to distinguish between possible causes of exercise intolerance, e.g. heart failure. There are over 30 variables collected as part of this test, and clinicians typically use peak rate of oxygen consumption (pVO2) as the primary variable to inform a diagnosis. A team in BWH cardiology identified a subnetwork of 10 variables that allowed patients to be stratified across four different patient clusters – each with a distinctly different all-cause hospitalization rate, which is a clear clinical outcome and a prognostic that can be understood and applied across the data collected from different patients who undergo this testing.
- Another example is one around a rare cardiovascular disorder – a debilitating disease that is phenotypically hard to distinguish from heart failure with preserved ejection fraction (HFpEF). In fact, a patient typically sees a few different clinical specialties before being referred to the appropriate cardiologist to diagnose the patient appropriately. A foundation within the Brigham and Women’s Hospital is considering a project to integrate machine learning-based approaches on echocardiographs and EKGs to distinguish between these patients to provide appropriate clinical care in a more timely fashion.
- I would like to highlight a project that is taking on the daunting task of characterizing the “exposome”, to bring to fruition the vision of network medicine. The exposome is an –omics style approach to characterizing the environmental factors that humans are exposed to, e.g. nutrition, the chemicals in the environment, etc. and correlating it with better-studied drivers of disease, such as the genome and epigenome. We are talking about variables upwards of ~27,000 on one axis, and all the potential permutations of probably just as many variables on other axes to identify important causal nodes.
- How is this technology transformative?
- The project on the rare cardiovascular disorder is an example of the potential for this technology to identify and stratify patients to provide real-time, personalized clinical care using existing information in the EHR – it can be applied across multiple indications that pheno-copy as HFpEF, once the data sets are trained appropriately. This will eventually require a shift in the paradigm from the stage-gated approach used in treating patients today. The CPET project resulted in a handy risk algorithm calculator where physicians can type in the measurements for the different variables collected over the course of CPET to risk-stratify patients, increasing the accessibility and functional use of this test to provider organizations outside of the Partners healthcare system, and possibly even outside of tertiary care centers.
- The network medicine project offers a new standard on how to reach the goal of practical precision medicine and has the potential to identify biomarkers, prognostics and molecules that contribute to homeostasis in a way that would not have been feasible up till the sheer computing power and technological advances available to us today.
- What’s most exciting about your role in helping the team move through the phases of commercializing their research?
- In this role, I get to apply my background in consulting and the training I have received as a licensing manager in Partners Innovation. I have the opportunity to think through the different facets of new product planning for algorithmic-based approaches in clinical care – from learning about the different cardiovascular disorders, the current therapies and corresponding clinical trial data from those in development, current “gold” standards of technology, to thinking about the barriers to market entry.
- For the rare cardiovascular disorder project, there is the potential to address tactical problems such as the accessibility of the EHR, and strategically also think about how to structure the intellectual property being developed. I also have the opportunity to implement a path towards commercialization in some cases. In the CPET project, I identified a champion within a commercial entity (from LinkedIn of all places!), and we have since successfully moved onto an agreement that allows the commercial entity to assess if the risk algorithm calculator can be integrated into the devices used in CPET today. For the network medicine project, I have an opening to help structure the potential funding for this project by working through the different paths that are available in the Partners healthcare system.
- What do you like best about your role?
- It is a position where I am privy to a cross-section of insights from all the stakeholders involved in shaping a thought bubble into a tangible product that will change patient care. Not only do I get the opportunity to learn first-hand about deep and complex mechanisms of disease from world-class experts and key opinion leaders, I also get the chance to execute on developing a commercialization strategy. For example, the commercial path for the CPET project was only feasible with continual engagement by the inventor with the commercial entity to implement the “evaluation phase” of the project. I also get a detailed perspective in how industry perceives the innovation that originate from the healthcare system, which empowers me to provide tactical feedback to the project teams with the goal of developing the best possible product offering in mind. With sufficient time (and patience), I hope to be able to see the efforts put in these project materialize into tangible clinical impact for the patients our hospitals see.
In the News: Apple Heart Study
Apple has partnered with Brigham and Women’s Hospital and the American Heart Association on a study of how heart rate and mobility signals relate to hospitalizations, falls, heart health and quality of life to promote healthy movement and improved cardiovascular health.
“We are excited to be working with all the study participants and with Apple…to empower individuals to maximize their own health,” said Calum MacRae, the vice chair of Scientific Innovation for the Department of Medicine at Brigham and Women’s Hospital and Associate Professor of Medicine at Harvard Medical School.
To read the press release visit: https://www.apple.com/newsroom/2019/09/apple-announces-three-groundbreaking-health-studies/
Printed from https://innovationblog.partners.org/ai-insights-cardiovascular-care · Published 10 Oct 2019
© PARTNERS HEALTHCARE. ALL RIGHTS RESERVED