Artificial Intelligence & How Doctors Think: An Interview with Thomas Jefferson’s Stephen Klasko

As I walk into the building, the sheer grandiosity of the room is one to withhold — it’s as if I’m walking into Grand Central station. There’s a small army of people, all busy at their desks, working to carry out the next wave of innovations helping more than a million lives within the Greater Philadelphia region. However, I’m not here to catch a train or enjoy the sights. I’m at the office of the President and CEO of Thomas Jefferson University, Dr. Stephen Klasko, currently at the helm of one of the largest healthcare systems in the U.S.

Let me backup a little.

The theme of nearly every conversation about the future of technology now revolves around Artificial Intelligence (AI). Much weight is placed on the potential capacity of AI to disrupt industries and change them to the very core. This pressure has been felt to a large extent within nearly every aspect of healthcare where AI has been projected to improve patient care delivery while saving billions of dollars.

Unfortunately, most discussions exploring the implications of AI only superficially look at either the product or the algorithm that powers these products. The short-sightedness of this approach is not an easy one to fix. Yes, clinical studies validating AI backed products are vital but AI cannot be viewed just like any other drug or a medical device. There’s much more to be considered when we examine the broader role of this technology, because this technology can shape the entire healthcare system. To place the impact of a far reaching technology, you need an even longer sighted vision. It’s a rare breed of people that have experienced the tumultuous history of change within medicine but can still call upon the lessons learned to execute innovations and bring meaningful results.

There are few people with such vision and foresight. One of these people is Dr Stephen K Klasko.

For the leader of an organization with revenues of nearly $5 billion, Dr. Klasko isn’t your typical CEO. Having grown up in South Philly, he brings a certain flair of grit – of taking risks, elbows out, unafraid of getting things done – that only people from a city like Philly can understand. This has taken him through a fascinating path from getting fired as a disc jockey (on his first night) to training as an Ob/Gyn at a community hospital in suburban Philadelphia (which traditional wisdom dictates could have limited anyone else’s academic career) to becoming the leader of one of the largest academic enterprises within the country.

He will unabashedly state he is a risk-taker. While easy to proclaim, his track record has demonstrated the courage to follow through on taking these risks. Most of Jefferson’s competitor healthcare institutions are investing hundreds of millions of dollars into building over-sized hospitals or constantly working to adapt to changes in the financial reimbursement system. Dr Klasko is taking a decidedly more unconventional route by being bullish on newer forms of care delivery. He’s building a large telehealth presence, hiring multiple developers of patient care apps, building a separate organization which focuses on Digital Innovations in care delivery, among many other initiatives. Perhaps due to his clinical background, he is among the few leaders to actively embrace technology and see its possible role in making a difference in the care of patients. Healthcare and health technology have been through numerous changes within the last several years and he’s adapted, often successfully, through a lot of them. It’s easier to say that an organization will incorporate an entrepreneurial DNA; it’s much harder to invest money and human-power in doing so. Dr Klasko is doing the latter.

Perhaps one of his most recent forward thinking collaborations was a collaboration with IBM Watson. This places Jefferson among a select few healthcare systems in the country to bet big on the impact of AI in medicine while directly examining its possibly large scale impact within patient care.

As a radiology resident physician with a deep entrepreneurial background, I have personally gone through the pains of launching digital health platforms directly and indirectly within patient care. It’s a formidable challenge. There are many different moving variables in medicine and this makes it particularly hard to implement innovations in healthcare.

Change also seldom comes from the top. Yet, Dr Klasko is at the forefront of leading the organization that is actively implementing this huge initiative. He brings a very unique perspective on the future role of AI in healthcare. I had to hear his thoughts. He was kind enough to oblige.

We had a great discussion with many exciting ideas; I’ll summarize a few overarching ones:

Solving patient suffering will always be the main goal

Using Artificial Intelligence based algorithms, financial industries have redefined the way they work. As I explain this to Dr. Klasko, he quickly dismisses this reductive nature in comparison to what we do in healthcare. Where he sees AI making an impact in modern medicine is in addressing the little details of helping the patient.

‘Imagine a Vietnamese only speaking patient, admitted to the hospital for terminal cancer’, he says. ‘Even for a small concern that she may have, like feeling cold in her hospital bed, she has to page her nurse, who has to finish what he’s doing and then come to her, get a Vietnamese-American translator on the phone, make sure she has no immediate medical emergency and then find the thermostat to increase the temperature. An AI enabled smart patient room – utilizing the Internet of Things – can speak to the patient directly, in their native language of Vietnamese, translate their request and change the temperature. Going forward, this AI system will remember these details for the next time the patient is admitted and even document these within the patient’s chart (which could perhaps the biggest boon of all to nurses)’

Dr. Klasko works closely with IBM Watson and is spearheading a partnership with IBM Watson Internet of Things to build ‘smart’ patient rooms at Thomas Jefferson University Hospital. Putting this technology centrally within the patient’s room, AI based systems are being utilized to address an important problem in healthcare – lack of efficient communication. In a healthcare environment where clinicians are addressing multiple needs of an ever more diverse patient population, helping bridge the gaps in care delivery is of paramount importance. There are multiple small but incredibly critical problems like these in healthcare. AI can solve these problems, and many more, but we need to make sure finding ways to improve patient care, even if it is by a small amount, is always the end goal.

The ability of AI to perform a certain task better than a physician is inevitable, and, it will happen. However, it is our duty, as healthcare providers, to find those little problems – with big consequences – and use this technology in solving these problems for the betterment of our patients.

Physicians need to be a part of the AI – and health technology – conversation

People with a purely technical and business background, although well meaning, often fall short in understanding the problems patients and their care providers go through on a daily basis. It’s not an incomprehensible conjecture – just ask any young doctor who’s gone from the books of medical school to clinical work as a resident and young attending physician). This problem is not unique in the context of AI, he asserts. Several years ago, a woman requested to meet with several doctors to seek advice in building an electronic medical record. Nearly all the physicians she contacted universally refused to help. This woman – Judy Faulkner – went on to build Epic systems, the platform that has been the inspiration of multiple other EMR platforms. Physicians didn’t actively participate in that conversation and subsequently, the design of EMR and the excessive time spent using them remains a constant complaint. To avoid this fate, physicians need to become an active force in the development and implementation of AI in healthcare. As Dr Klasko emphasized “to not help physicians be better robots than a robot, (but) for the physician to be (more) human.”

Furthermore, Dr. Klasko believes we can help apply these technologies to shape the future talent pool of physicians. Instead of prioritizing on grades alone, especially that of organic chemistry (a class I think he has a personal vendetta against, not unlike many other physicians) but instead weigh selection criteria on the non-traditional factors of healthcare. As AI based systems get more proficient at diagnosis, management and prevention of medical problems, the future talent of physicians can be selected not just on scores or the ability to regurgitate differentials but on the ability to be better care providers, of being more empathetic and of finding innovative ways to solve the problems that patients face.

The next talent pool of physicians has the power to be more communicative, creative and empathetic precisely because, as a collective physician cohort, we relinquish what the learning machine will be better at – pattern recognition – and instead focus more on ‘being human.’ 

Empathy will remain the cornerstone of healthcare 

As I tell Dr. Klasko about the research that I’m doing involving Artificial Intelligence in medical imaging, he stops and asks me point blank ‘Do you believe that AI will replace Radiologists, Pathologists and Surgeons?’ ‘No,’ I say, in a determined voice, trying to hide an undertone of uncertainty. He emphatically agrees ‘No!’ As someone who has delivered more than 2,000 babies, Dr. Klasko says he knows the challenge that it takes to go and tell a mother of potential complications her newborn might have. Even though there may be an intelligent robot to deliver babies in the future, perhaps better than any other obstetrician in the past, the communication and relationship a physician and their patient have, can never be replaced.

As a radiology resident physician, this brings a level of clarity to my concerns about the future of professions like mine. Whether we translate emergent findings of acute traumatic pathology or we parse through complex oncological imaging findings during multidisciplinary tumor board, our role in patient care, while more indirect, is quite important. Being the medical imaging expert in the room, our role as the ‘doctor’s doctor’ helps quarterback pivotal decisions within patient care. Delivering that human touch, even if it is being the colleague that a fellow medical professional needs to help guide therapy, is an utility that even non-patient facing physicians provide.

Being that expert that people rely on, isn’t a role that AI can replace from physicians. Yes, AI can augment human skill sets – make physicians better, faster, smarter – but working together as a team, looking at multiple different factors in patient care and putting it in the context of what the patient is going through, is a skill that is uniquely human.

With a whirlwind of ideas, and what felt only like moments, our meeting came to a close. I ask him about the Steve Jobs book he has on the table, “one innovator reading about another,” I remark. He chuckles and says, “Steve had the uncanny ability to think ten years into the future and in thinking where healthcare will be ten years from now, we can help shape its future.”

Dr. Klasko remains committed to the implementation of AI in healthcare, building strategic partnerships, launching innovative programs and improving patient care, with a keen eye for the future. In doing so he has demonstrated within himself the same concept of human ingenuity, creativity and grit, that he hopes to bring out from others, and this is something AI can never replace.

Human Spirit, with grit – 1

Artificial Intelligence – 0

Ajay Kohli is a physician with an entrepreneurial mindset. He is currently completing his radiology residency. He is actively working on the application of AI in healthcare. He has been named a ’40 under 40′ Healthcare Innovator by MedTech Boston Magazine. He loves working with people in advancing medicine and healthcare delivery. Reach out to him here http://www.ajaykohlimd.com

1 reply »

  1. Hi Ajay, thanks for the piece about your conversation with Dr. Klasko.

    One thing I’ll add is that doctors also need to be involved in a conversation with lawmakers, regulators, lawyers and programmers so that all parties understand how the underlying technology works, how it is programmed and to what extent it can/should be relied on as a diagnostic tool or regarded as “member of the team”.

    Given that IBM tends to overhype Watson’s capabilities, lawsuits stemming from Watson reliance/error are inevitable and there needs to be sufficient legal education among stakeholders to understand the extent of Watson’s capabilities and responsibilities.

    A colleague and I at NYU wrote a bit about this here: