AiThority Interview with Charles Taylor, Founder, Chief Technology Officer at HeartFlow, Inc

Know My Company

How have you interacted with smart technologies like AI and Cloud-based analytics platforms?

Our team at HeartFlow has combined expertise in Scientific Computing, Bioengineering, Cloud Computing, Computer Vision and AI from leading institutions. We’ve put our combined knowledge from the fields of engineering and medicine to create the HeartFlow Analysis, in order to revolutionize how physicians diagnose and treat Coronary Artery Disease (CAD), the #1 cause of death.

When patients experience symptoms of CAD, they undergo a coronary CT angiography scan. At a high level, our HeartFlow Analysis works by utilizing data from these scans via Deep Learning algorithms to create a personalized 3D model of the patient’s heart. This patient-specific computer model is then used by physicians to identify where blockages in the coronary arteries may be occurring and impacting blood flow to the heart.

Read More: AiThority Interview Series with Daniel Clark, CEO at Brain.fm

How did you start in this space? What galvanized you to start HeartFlow?

Starting HeartFlow was actually the result of a chance encounter. In late 1993, while I was a PhD student in engineering at Stanford, I went to a talk led by the university’s new chief of vascular surgery, Dr. Christopher Zarins (HeartFlow’s co-founder). As I heard him speak about blood flow and cardiovascular health, I realized that computer modeling techniques I had been studying for the last several years could be used to quantify blood flow in patients’ arteries. I completed my doctoral research at Stanford with Chris and Tom Hughes, PhD, a professor of Mechanical Engineering at Stanford and a leading expert in computational fluid dynamics, on the topic of computer modeling of blood flow in arteries and did the first simulation of blood flow in arteries from medical imaging data. In 1997 I became a professor in the school of medicine and engineering at Stanford and continued developing computer simulation technology for another 10 years before Chris and I founded HeartFlow.

What is HeartFlow and how does it leverage Data science and visualization techniques in its operations?

HeartFlow is a medical technology aimed at changing the way cardiovascular disease – the leading cause of death for both men and women worldwide – is treated and diagnosed. We provide a new approach to non-invasive testing for coronary artery disease (CAD), the most common form of heart disease which affects nearly 17 million Americans today.

The problem with CAD is that it can be difficult to diagnose because its symptoms can range from feelings of indigestion to breathlessness – or worse, sometimes patients don’t show symptoms at all. On top of this, while other noninvasive tests are available, they are not effective in helping physicians diagnose CAD and often result in additional tests, such as an invasive cardiac angiogram. And, more than 55 percent of patients sent for an invasive procedure is found not to have significant heart disease, making the invasive procedure unnecessary in hindsight.

Our HeartFlow Analysis aids physicians in identifying the right treatment pathway for each patient whether it’s ultimately through minimally invasive procedures like stenting, or alternative ways of managing their disease (like medication). It utilizes a software-as-a-service model in which computed tomography (CT) images of a patient’s heart are securely sent from a hospital through the cloud to HeartFlow. By using AI and certified human analysts, HeartFlow produces a digital 3D model of a patient’s arteries and solves the equations of blood flow using supercomputing techniques. The end result is easy to interpret, the color-coded model that gives physicians the ability to determine if the patient has any blockages in the arteries, and whether or not the blockages are impeding blood flow and requiring treatment.

Read More: AiThority Interview Series with Kobi Marenko, CEO and Co-Founder at Arbe

What is the state of AI for CT scan technology in 2019? How much has it evolved since the time you first started here?

AI is one of the most important technologies in the field of Medicine, particularly medical image processing. It’s being applied in a myriad of ways from image reconstruction to image segmentation applications. In the last few years, Deep Learning methods for image analysis are yielding exciting results. Although not obvious at first, it is clear to me, that the power of these methods can be maximally leveraged by supervised learning whereby automated algorithms are inspected and corrected by human analysts and the annotated data is then fed back into the Deep Learning algorithms to yield more accurate results. At HeartFlow, we have used this approach to train our coronary artery image analysis methods on data from nearly 10,000 patients and are seeing quite remarkable results. This has enabled us to analyze data from patients with very severe disease that is, quite frankly, not easily interpreted by even the most expert of radiologists.

Why do you think an AI/human collaboration has the potential to change Medicine?

Combining the insights gained from AI and Deep Learning, along with the medical expertise of physicians, more accurate diagnoses are possible, and physicians can determine the best treatment and management plans for individual patients, resulting in improved patient care and health outcomes, and ultimately saving hospitals and patients money. I am a big believer in augmented intelligence to combine the best of the machine and human intelligence.

Read More: AiThority Interview Series With Scot Marcotte, Chief Technology Officer at Buck

What is your vision about the growth of AI-driven healthcare and diagnostic technologies?

For cardiology, in particular, I believe AI’s capabilities will help physicians collate data, as more of it’s gathered and analyzed with Deep Learning technology. AI has the potential to become a powerful partner for physicians and go beyond simply aiding in the accuracy of diagnoses but help them achieve a deeper understanding of the severity of a condition and better explain to patients their symptoms and provide them with catered treatment plans. At HeartFlow, we’re also using AI to learn which factors are most predictive of plaque rupture and heart attacks. This is really exciting, and could one day remove heart attack from its position as the #1 killer of men and women.

How could healthcare businesses leverage Artificial Intelligence technology to strategically price their products?

Leveraging AI to optimize the treatment of individual patients, whether through drugs or devices, could enable healthcare businesses to provide information to physicians and hospital systems that would be priced for performance. For example, patients that might be most likely to respond favorably to a given medication or device might be identified and then this information could be provided to their physician.

How do you see the trend of including AI and Machine Learning in a modern healthcare budget?

There is no doubt in my mind that AI will be one of the most important technologies in precision medicine initiatives and will be prioritized in modern healthcare budgets. The possibility of improved diagnosis of disease and targeted therapies will drive this change for the foreseeable future. For example, today a given therapy might be recommended for all patients based on clinical data showing statistically significant benefit in a slight majority of patients.

Read More: AiThority Interview Series with Dean Abbott, Co-Founder And Chief Data Scientist at SmarterHQ

Who are your main competitors, within this landscape?

HeartFlow’s main competitors are Legacy Diagnostic tests that have dominated cardiac care for decades. On the non-invasive side, these tests include exercise treadmill tests, echocardiograms, and nuclear stress tests, but do not provide physicians a complete picture of what is happening with their patients’ hearts. However, despite clinical data in hundreds of thousands of patients demonstrating that these tests have low diagnostic yield, they’re still used tens of millions of times per year worldwide.

How should young technology professionals train themselves to work better with automation and AI-based tools?

In my view, and based on more than 25 years of experience teaching at Stanford and working at HeartFlow, young technology professionals should learn the foundations of mathematics and computer science and also go deep in a particular problem domain. There will be plenty of opportunities in this field for professionals with expertise in AI, data science, but I think those that are most successful combine this knowledge with something else. For me, it was cardiovascular physiology and scientific computing. For some of our brightest scientists, it is AI and medical image analysis or AI and computational fluid dynamics. Others interested in applying this to medicine may become “double threats” in AI and genetics for example.

What is the biggest challenge to Digital Transformation in Healthcare and Diagnostics in 2019? How does HeartFlow contribute to a successful digital transformation?

I think the biggest challenge is improving healthcare provider awareness and overcoming the healthcare industry’s capacity to adopt digital technology from a budget, bandwidth and expertise perspective. The vast amount of technology innovations that have flooded the industry can be overwhelming and confusing for physicians and providers, so this, paired with the healthcare industry’s slow-moving pace can present itself as a major challenge.

HeartFlow aims to address these challenges and ease the physician burden by changing the way that cardiovascular disease is diagnosed and treated, starting with the clearest picture of a heart’s health, representing a paradigm shift in patient care by improving the overall experience and quality of life.

Healthcare can be a financial drain for both patients and hospitals, but with HeartFlow, physicians can become confident that they are selecting the best and most efficient care pathway for their patients. For example, the HeartFlow Analysis has been shown to reduce the number of unnecessary tests and procedures as well as reduce the overall cost of care by more than $4,000 per patient after one year. To date, over 30,000 patients worldwide have received the HeartFlow Analysis.

In addition, HeartFlow is also working within a physician’s existing workflow and giving them a tool that becomes crucial to their practice. In the last year, we released a mobile app that helps doctors easily access their patients’ HeartFlow Analysis with one swipe. The app provides an interactive and accessible way for physicians to zoom in, orbit around the model, and identify the patient’s FFRct values (the number associated with blood flow pressure drops that are used by physicians in diagnostic decision making) which is not only valuable for the physician but also helps them visually explain the diagnosis to their patients. The app provides a summary of the test results and additional information needed to determine the best course of treatment on the go for cardiologists who may often be working between multiple hospitals.

Read More: AiThority Interview Series with Stuart Brock, Director at Seal Software

What could be the Good, Bad and Ugly about AI that you have heard or predicted?

 The good of AI in healthcare is upon us. We are seeing the transformation of healthcare and I expect it will be much faster than people expect. In a matter of a few years, we will have a number of robust technologies targeting specific diseases and dramatically improving care. The bad is also that AI in healthcare risks being over-hyped as web companies were in the late 1990s. There is a flood of investments in AI in all industries, and I expect it will be challenging at times to identify the authentic and valuable uses of this technology from the myriad of purported applications. Personally, I’m uncertain as to the commitment our society will have to educate individuals with the necessary skills to work in this new age of technology. It’s clear to me that there will be entirely new jobs created that do not exist today. For example, at HeartFlow we have created a job for analysts that are highly trained using our software to read CT imaging data and inspect and correct the results of AI computer algorithms. Other companies are emerging offering data annotation services.

What is your opinion on “Weaponization of AI/Machine Learning”? How do you promote your ideas? 

The weaponization of AI/Machine Learning is indeed a scary prospect. Any era of technological change comes with risk, whether that is with AI, synthetic biology and bio-warfare or nuclear threats. It is a responsibility of those creating technology to educate policymakers and the public as to the risks as well as the benefits of new technology. I promote my ideas principally through online forums and also speaking engagements.

Read More: AiThority Interview Series with Jeff Epstein, VP of Product at Comm100

The Crystal Gaze

What Data-driven healthcare start-ups and labs are you keenly following?

I follow the work in the labs of some of my former doctoral students that are on faculty at Stanford, U.C. Berkeley and the University of Michigan to name a few. We collaborate with the computer science department at Imperial College in London as they have a world-class program in the application of AI to medical imaging. I’m quite interested in the work going on at Paige.ai especially since Leo Grady joined them as their new CEO after working for HeartFlow as our Sr. VP of Engineering. I also follow the AI related work at GE, Siemens, Philips and Canon as they are the leaders in medical imaging.   

What technologies within AI/NLP and Cloud Analytics are you interested in? 

I am most interested in supervised and unsupervised Deep Learning methods these days.

As a tech leader, what industries do you think would be fastest to adopting healthcare technologies with smooth efficiency? What are the new emerging markets for these technology markets?

The medical imaging market within healthcare is one of the fastest adopters of AI technologies. China and India are two emerging markets that are extremely exciting for healthcare AI these days. In both cases, much greater automation and more efficient healthcare delivery systems are needed to support the needs of these billion-plus patient markets.

Read More: AiThority Interview Series with Dr. Massimiliano Versace, CEO and Co-Founder at Neurala

What’s your smartest work related shortcut or productivity hack?

I wake up around 4 am each morning and, after a cup of coffee, spend 30 minutes planning what I am going to do that day. I have three checklists that I maintain and use. One for urgent items that have to be completed that day, a second for the medium term (week-long tasks) and a third for longer-term projects. I then clear my email inbox from the night before so when I get into the office, I can fully engage with our team. 

Tag the one person in the industry whose answers to these questions you would love to read:

Yann LeCunn

Thank you, Charles! That was fun and hope to see you back on AiThority soon.