WellAI CEO: Why Is Epic So Wrong and So Right?

By Sergei Polevikov

This month we welcome guest-blogger Sergei Polevikov, CEO of WellAI.

Do you know who controls your medical records?

If you are not familiar with Epic, you are probably an outsider to the healthcare industry.  Epic is a leader (some may even say – a monopoly) in the electronic medical records (EMR) business.  It’s not just medical records.  It’s a slew of services for hospitals and doctor offices, from appointment scheduling to patient data analytics.

A recent Forbes article The Billionaire Who Controls Your Medical Records explains how Epic has been and continued to be so successful, even in the age of AI and other recent technological advancements.

Another recent article How a Largely Untested AI Algorithm Crept into Hundreds of Hospitals tells a story of how Epic introduced an algorithm called Deterioration Index during the COVID pandemic that allows triage patients to determine priority of medical care.

I’d like to use this article to highlight key differences between a health technology company that has existed for decades, like Epic, and a newer health technology company, like WellAI.

1)  Epic’s Deterioration Index is factor (or decision tree) based.  On the other hand, WellAI is machine learning that utilizes neural nets algorithms.  In our backtests, we demonstrate that in medicine, smart machine learning algorithms have much higher prediction accuracy than the decision trees (or rules based algorithms).

2)  Model transparency.  “The deployment of a largely untested proprietary algorithm into clinical practice—with minimal understanding of the potential unintended consequences for patients or clinicians—raises a host of issues.”  This should have been a benefit of Epic, not a detriment.  The decision tree approach should be easier to attribute compared to WellAI’s neural nets “black box”.  However, for whatever reason, Epic doesn’t seem to provide this attribution.  For WellAI, transparency is imperative.  We are offering a subsample of medical articles to doctors and patients which highlights the same symptoms the patient is experiencing and point to the place in the article where the diagnosis is coming from.

3)  Traditional AI biases – a point made in recent popular press with respect to minorities, among other biases.  “Studies have shown that, among other injustices, physicians underestimate the pain of minority patients and are less likely to refer women to total knee replacement surgery when it is warranted.”  This perfectly describes another problem with Epic-like rules based algorithms: Most medical researchers analyze databases of doctor-patient interactions – not because it’s necessarily right, but because it’s easier.  Medical studies do have some of the same biases, but to a lesser extent – to publish an article, one has to develop a statistically unbiased study.  Often medical articles themselves are addressing these biases via various statistical techniques.

4)  A different kind of bias: doctors and nurses are affected by the solution an AI model is suggesting.  “[P]hysicians in training would be learning medicine on Epic’s terms.”  Here I’d like to disagree with the authors.  Yes, sometimes such biases happen.  However, I believe a more common scenario is as follows: doctors are augmenting their experience with suggestions from WellAI’s scientific ML algorithm.  I am not sure how transparent Epic is when introducing their products to their clients. At WellAI, before launching our platform to patients, we are implementing thorough education and training with nurses, PAs and doctors. 

5)  Epic keeps things proprietary, and doctors don’t know the strengths and weaknesses of Epic’s algorithms.  “Companies like Epic should also engage more deliberately and openly with the physicians who use their algorithms; they should share information about the populations on which the algorithms were trained, the questions the algorithms are best equipped to answer, and the flaws the algorithms may carry. Caveats and warnings should be communicated clearly and quickly to all clinicians who use the indices.”  As we mention above, at WellAI, product education is critical.  It is important that doctors and nurses know ML model’s strengths and weaknesses, and feel comfortable overriding WellAI’s suggestions when they feel, for example, that they may know something about a particular patient or a particular health condition that a ML algorithm may be missing.

6)  The article doesn’t talk about safety and security of medical data.  That’s where WellAI’s expertise in blockchain comes in.  We think HIPAA standards and other regulations have been outdated.  We believe blockchain is the next frontier of safe and at the same time transparent medical data storage.  We want to be industry leaders when this transformation finally happens.

So how could Epic be so wrong and so right at the same time?  Some great technology companies have tried to demonstrate that they can provide the same service Epic provides, but better and faster.  They failed for three reasons:

  • Big technology companies seem to be very retail consumer centric. Doctors and hospitals don’t necessarily need the shiniest and the brightest technology retail consumers are demanding.  Whether the reasons are justifiable, it’s irrelevant.  They just don’t…  They go with Epic because Epic has been providing the same reliable service for four decades.
  • Most technology companies don’t provide the universal service that Epic provides. It is usually a piecemeal software solution.  Outside of WellAI, there are not many companies providing the true “full workflow” automated solution for a doctor’s office or a hospital.  In that sense, Epic is probably the name that comes to mind right away. 
  • Security of patient data. While big technology companies are supposed to be the best in providing safety and security of patients’ data, they have also created a reputation of the ‘big brother’ – they collect a lot of information about their customers.  But no one seems to know what exactly the big technology companies do with that information.

Epic may not be providing the best technology.  Their customers may not like that.  But it’s what you have in your hands that matters.  Not what you are dreaming about.  The fact that Epic has been in business for over four decades and is the biggest player in this industry means it’s doing something right.

WellAI is changing this status quo for doctors and hospitals.  We believe that our one-of-a-kind technology would help doctors, hospitals and patients achieve better care.  Our platform is a combination of ML algorithms that have read 30+ million (and counting) medical studies at a human doctor level and blockchain for patients’ security and transaction transparency.

We believe that’s the future of healthcare…

At WellAI, we use artificial intelligence for what we believe artificial intelligence is supposed to be used for: quick, efficient, cheap and accurate information access to the knowledge of 30+ million medical studies.

If you want to learn more about this break-through technology that is meant to provide unique access to scientific health information to everyone and to dramatically improve efficiency of your health spending please schedule a 1-on-1 demo:

  • If you are a doctor, an employer, a health benefit advisor, or an insurance broker and would like a cool benefit that makes your employees healthier and happier, and saves your business time and money, please schedule a virtual meeting here.
  • If you are a potential investor who would like to be part of a unique once-in-a-lifetime investment opportunity that will forever revolutionize healthcare, please schedule a virtual meeting here.
  • If you are a potential customer, a patient or you are just curious about any of the WellAI high tech products – ‘Health Alexa’, next gen telehealth, or the scientific chronic disease management program – please schedule a virtual meeting here.

Get your WellAI!  Ask questions!  Stay Healthy!

Posted in

Leave a Comment