Making Sense Of Big Data For Better Health With Tyler Cohen Wood

 

Big data in healthcare is only advantageous if you know how to make sense of it. Tyler Cohen Wood is the founder and CEO of MyConnectedHealth and former US Defense Intelligence Agency (DIA) Cyber Deputy Division Chief. Using the methodologies she’s learned from her background in cybersecurity, Tyler created a healthcare solution that makes diagnosis easier for doctors and patients alike. She joins host Ben Baker to share how her own health scare experience taught her that working with multiple doctors without collaboration or proper documentation only dragged out the process. This became the motivation for MyConnectedHealth where she aims to help not only herself but other people battling the same ineffective system. Tune in to get her insight on how AI and digitalization will contribute to the healthcare system’s future.

—

Listen to the podcast here:

Making Sense Of Big Data For Better Health With Tyler Cohen Wood

I’ve got Tyler Cohen Wood from MyConnectedHealth and we’re going to talk about big data from cybersecurity to healthcare. What have we learned? Tyler, welcome to the show.

Thank you so much for having me. I’m excited to be here especially to talk about these topics.

It’s exciting because you have a past that most people don’t have. I’m going to let you tell your story but I think that there is a lot of congruencies between what we’ve learned about cybersecurity and big data and healthcare and big data and what we can take from one area to the other. Let’s start off by finding a little bit more about you. Where did you come from? Where are you? We’ll talk about where we’re going.

My background is in cybersecurity. I’ve worked for the Department of Defense Cyber Crime Center, doing digital forensic cases and incident response so major crimes and then intrusion cases and then I went to the Defense Intelligence Agency where I was a Deputy Division Chief for a special comms branch. There I developed many cyber capabilities that are weird but it’s just a certain way of thinking that I used in that position.

I was involved with a lot of stuff cyber in the intelligence community, I loved it. I’ve written a couple of books. One of my books Catching the Catfishers. That was a book that is designed to teach parents and kids how to be safe and protect themselves in the online domain. I included easy-to-understand checklists if who you’re talking to is who they say they are. The book is doing great. To make a very long story short, I got sick in 2018 and that changed my trajectory incredibly. It’s an interesting career.

I want to get into all this but I want to go back to cybersecurity. I don’t want you to talk about specific cases or specific things but what are the things that you’ve learned that helped you? What are the tools that you built through your learnings through big data and cybersecurity that allowed you to become a far better analyst or work with analyst teams to disseminate information easily? That’s what it’s all about. It’s not about the information because we’re throwing information a million times a day. It’s how do we make sense of the information.

That’s a difficult thing to do and in each position that I’ve had whether it was working AT&T or DIA, it was a different skillset but all of the different skillsets dealt with piecing together information in a way that it maybe isn’t normally pieced together that way. Also, looking behind the scenes to get the full picture. It’s an investigative mind and it’s something that I’m good at.

I will never forget. I had a case. I had just left one position where I was doing forensics and incident response and I’ve been working on an intrusion case. When you’re working on those, you start to learn IP addresses in your head. You know them and other types of tools or things that you see. We finished that case. I left and moved to another position.

A lot of the times it’s the background things that turn out to be much more important. Share on X

At that position, my first intrusion case, I had literally come off this other case with another company. It’s because I had those IP addresses and things still in my head, I typed in keyword searches to see what would happen. It was completely the same hacker. I was thinking about that and that got me thinking about how there is such an incredible need for collaboration in this space because if there was some collaboration between those two entities, it would have been discovered. It made me think about all the things that we missed because we’re not working together. That was an incredibly important lesson that I learned.

I look at it as solving a puzzle on your own, 5,000 pieces or 10,000-piece puzzles versus having 5, 10, 15, 20 people each taking little sections and putting them all together and having that enabled to be able to have the sections in the cross-sections and say, “This piece goes with this.” All of a sudden, you have a much larger picture because that’s the challenge. Not any one person has all the information.

It’s impossible. Whether we’re dealing with cybersecurity, healthcare, even our own family, we don’t have the full picture. That’s where I see the benefit of big data and analysis is to be able to have a more congruent system by enabling a larger number of people to work together to be able to solve a problem. Is that where the real success was?

Yes because with cybersecurity, when you’re doing these responses and you’re doing forensics or you’re piecing together solutions to keep special forces protected in high-risk areas, you’re working with other people and there are not always easy answers. You may not have all of the evidence. It is a collaborative effort and it’s a way of thinking what you see directly in front of you at first may not at all be what the answer turns out to be.

There’s not only an incredibly collaborative effort but when you’re searching, you’re searching in a much more expansive way than in other verticals when you’re trying to figure out a problem because there aren’t always those easy answers. You have to think in a certain way. I’ll use a medical example. If you have a CT scan of the abdomen and if that CT scan was ordered to look at the stomach, they may notice things that are in the scan.

Maybe there’s a granuloma on a lung or a lesion on a kidney. These things are sometimes annotated along with what the test was ordered for whatever they saw within the stomach. Those other items are pretty much often missed because of what’s happening in the background, a lot of times it’s those background things that turn out to be much more important.

Let’s get the circle rounded a little bit and let’s move from your cybersecurity days, which were important to talk about but get into healthcare and why to make the change. You said that there was a medical issue. Without getting into too much detail about what the medical situation was, obviously, it was enough to change your life. What happened and what were the major challenges that led you to make that a-ha moment and brought you into MyConnectedHealth?

 

I got sick in 2018 and I know this now. I was sick before but we just didn’t know it. I ended up in the hospital and then started the whole process of trying to figure out what it was and what was wrong. I ended up finding myself in what I call specialty siloing and that’s where maybe there’s an endocrine issue too. The GI doctor and the endocrinologist cannot talk to each other, cannot communicate and so you end up being shuffled back and forth between doctors. It’s very difficult too when you’re sick, you have to be the person that takes all of your records, reviews your records and it’s not easy to do.

I kept getting sicker and the puzzle was becoming bigger. It was a difficult thing to go through and we did get a partial diagnosis but we knew there was something else. It was difficult because I kept trying to get help but I wasn’t getting answers. It became apparent to me that I needed to do this a different way and I needed to find a way to help the doctors.

The first thing that I did was I took all of my test results and I put them in a chronology. I set up an easy little script to notate if something was off. It was like providing an executive summary for the doctors because there’s not a human alive that can go through every single medical record, put it in a chronology and annotate what’s important. It would take forever. You wouldn’t see any other patients. That’s where AI and machine learning is going to come in handy. It’s not replacing doctors but doing the dirty work and the heavy lifting so that it makes their jobs a lot more palatable and they have more information because it’s right there at hand.

I have a personal story about this. I remember I was in a bad car accident many years ago. It ended up being a medical-legal thing. I went to go see a specialist one day. It was for a medical-legal assessment. I looked behind them and there are 6 or 7 4-inch binders full of paper. I said, “How many clients are that?” It was an enormous amount of paper and you’re sitting there going, “How many clients are that for?”

He goes, “Those are your records.” You sit there and say, “This is all sitting in a paper-based system.” Maybe the lawyers are able to categorize it but how do they take all that information and be able to disseminate it and be able to take there and say, “We’ve got fifteen different experts that all assess them on different things and be able to bring all that information together. Not only to be able to create a diagnosis but a legal argument that goes along with it.”

The challenge is with that is the more complex the issue, the more complex the problem, the longer it goes on, the bigger the record becomes. It becomes more complex and harder to sit there and say, “It started here and then here’s the chronological scenario of what happened and things get missed.” Where do we go from there because this is a challenge that’s not only you? This happens to hundreds of thousands, if not millions of people around the world every year.

That’s the problem. There is so much information and there are so many records that pile up. There are electronic medical record systems, EMRs but we can talk about that on another call because that’s a topic in and of itself. In that paper form, you’re right. There is just so much information that you end up going down rabbit holes and the patients don’t necessarily know what the doctors need to see so they bring everything.

The doctors are sifting through things but what they need is not something the patient brought. There’s also a translation issue there as well. It’s very difficult. I needed to know what was wrong, I will sit with me and I will say that we were going down the wrong path. We went to some of the best hospitals in the country, we were just getting nowhere and I was getting sicker. Knowing what I know now, we were in a totally different specialty.

Your system is only going to be as good as the data that you have and the algorithms that are written. Share on X

You needed that piece of the puzzle that might’ve fallen underneath the table somewhere or got swept under the rug and you didn’t even know that it was there.

No, I didn’t. I started learning to read medical journals. I learned how to retest and how to interpret the test, especially in reference to other tests. If something is high, you expect something to be low but sometimes it isn’t. I learned to try to interpret that and you have to understand that I was searching high and low for what the problem could be because I knew I was in trouble. I knew I was really sick. Honestly, it was COVID that changed things for me because once we knew that there was a problem, I got scared. That’s when I decided I’m doing something with MyConnectedHealth and I’m going to make the system work because I was afraid of what this was going to mean for the world.

I wanted to show that even just the concept or the theory would work. What I did was I took my cybersecurity background and I used the methodologies that I had learned from my cybersecurity career. I went back to the records and I took the records but I did something different. Instead of focusing in those CT scans, I started focusing on some of the things that were just in the background that happened to be annotated. I put them into a statement with a known condition and then a test from an alternative doctor, an herbalist. She’s a DO but she believes more in herbs. She had done an inflammation panel and I just happened to notice and I was like, “I don’t know what that is but that’s high.” I just popped that in and I ended up with an autoimmune disease that is only 300,000 people in this country are known to have.

Now I’m seeing a rheumatologist and a whole other specialty of doctors but that’s what it took. It was looking at it from a very different lens. I believe that this is the answer. It sucks that this is what it had to come to. My story is not unique. Everyone knows at least someone that has a child or they themselves have some rare condition that no one can figure out. Sometimes these things are debilitating and they’re life-altering. To be able to offer that hope to people and give people and physicians a chance at getting ahead of disease, that’s what MyConnectedHealth is about.

From what I’m understanding, what MyConnectedHealth is, it’s the ability to put a large number of data points, which could seem relevant, may not seem relevant, into a centralized database and be able to have it compared to an even larger set of data around the world and sit there and say, “When you put this and 10,000 things together, here’s a diagnosis.”

Unless you’re able to look at all the small little minute points in concerts, which no one doctor can do by themselves, it’s almost impossible. No one human being can do this. You’re going to be able to come up with a more significant diagnosis and be able to see patterns that the naked eye could never see. Is that what I’m understanding?

Yes. That’s part of it. Remember when you’re working with healthcare and AI, your system’s only going to be as good as the data that you have and the algorithms that are written and what they’re looking for, how they’re piecing it together. That’s where our part of our secret sauce comes in and then the other part is changing the structure of how harder to diagnose cases are done and that’s by utilizing doctor teams. You have communication between the doctors working together around a single case.

 

There is a lot of up-skilling of the system as well because the system will be learning everything that’s worked, that’s been suggested and everything that didn’t. Eventually, the thinking is that it will become predictive and it can be used by the pharmaceutical industry to also help better, very strategically target medications to diseases. It’s like, “Let’s bring it all together.”

The challenge comes into being able to acquire the data in a way that doesn’t breach medical doctor, patient confidentiality, to be able to look at data in aggregate. I’m assuming you’re looking at the data in aggregate. There are no names attached to these things. Maybe there are sequencing numbers that can tie stuff back but those are under a completely different medical record system so they don’t interfere with each other.

You’re looking at it, sit there and say, “All we need is the data. What’s going on? What are the symptoms that people are having? What are the medications that people are using? How is this interacting with this?” I’m assuming the more data you have from around the world, the better the system is.

How do we create a system that can bring in all this medical data from 100 or 1,000 different systems that may or may not talk to each other, to be able to get the aggregation of data, to be able to allow the AI and to do the predictive analytics that it needs to do? To me, that’s the big problem. Aggregating the data in such a way that it can be effectively diagnosed, for lack of a better word, and utilized.

That’s one of the big issues with electronic medical records. There are quite a few companies that do these electronic medical records, the consolidations of these records. If you’re utilizing multiple sources of consolidators then you’re going to have to have the right APIs and some of the data may not be usable. This is a problem that I’ll be interested to see what the resolution is. There are ways that it can be done a lot easier but when you’re working with so much information and multiple consolidators, that’s where things get tricky.

I have this vivid memory. My GP retired several years ago. He was an old school doctor that hand wrote everything, a few pieces of paper that were sitting in your file and he would have to fit there and go back in order to be able to figure stuff out. When he retired, you couldn’t get your medical records. It had to go to a central aggregator and then have the new doctor bring it in. What we were allowed to do, when I went to the central aggregator, we could apply to have a digital copy of the records for ourselves. I did. I said, “We’re going to get a digital copy of the records for myself and my entire family,” from that point.

What I realized was that it was a high-speed scan. You still had the handwritten notes that were illegible. You still had the little scribbled pieces of paper that may or may not have made any sense whatsoever. There didn’t seem to be any OCR technology that was taking this stuff, turning it into documentation forms, creating some type of a search pattern for it so, therefore, it could be used to be able to diagnose what happened to me many years ago.

Doctors have become more technological and a lot of the doctors are online but you still have all this old record-based technology that’s paper-based. Realistically, are we ever going to be able to disseminate that information or is it, “Maybe we can go back few years and everything beyond that, we’re lucky if we get it.” Is that more of a realistic viewpoint?

We’ve seen more innovation in this kind of digital healthcare world since COVID. Share on X

I will say there is a lot of innovation in this space and we’ve seen more innovation in this digital healthcare world. It’s just jumped since COVID. I do believe that there is a company or business that takes those records and makes them electronic. You’re going to ask two questions. Is it possible? Yes, it is possible. Is it useful information? I don’t know because there’s that rabbit hole but it could be useful information if it’s something that’s persisted for that long but it just would have to depend case-by-case. I know in my particular case, having so much information took us down rabbit holes.

I’m sure it does in most people. Is there a change in how doctors are being trained to be able to notate and communicate, to be able to make sure that the information is more disseminated on a more consistent basis? Is there a methodology that people are working on? From this point forward, we’re able to realize that the information is not garbage in garbage out. There is more of a detailed approach or more of an annotated approach that is going to allow for that material to be accessed and made sense of by big data. Is that something that you’re hoping for? Is that something that MyConnectedHealth is relying on? Where are we with that?

Just to make sure I understand the question. Are you asking is there a place where all of this information is being stored.

There are two things. The story of information is one thing because that’s the big thing. We have to have the information stored in such a way that your system can use the algorithm to be able to do what your system does. The other thing is how do you get that information into the system and train the doctors to be able to look at documentation as a vital part of their actual day-to-day job.

That’s challenging for doctors because there’s never enough time in the day. There are never enough hours to do that documentation. To be able to have it in such a way that it’s little things that may be important don’t get ignored or be seen as superfluous and become a side scribble on a piece of paper that could unlock a secret but never see the light of day.

Some of that with telehealth systems gaining popularity and momentum, we’re seeing some of that go away because things are just becoming more electronic in this day and age. Also, you will have some doctors that don’t want to make a move to these telehealth systems or these types of systems to help them.

You’re going to have the majority of doctors who do want to utilize systems like this because it helps them. Frankly, most doctors want to help their patients and when they can’t help their patients, it’s hard. There are a lot more doctors that are interested in jumping on to working with systems like this. There are the standouts that aren’t ready for that yet. You also have to understand too, that the generation of doctors that are coming up, they’re the Millennials who’ve grown up on technology. What’s below the Millennials?

 

The Gen Z. They’re digitally native.

That’s just how things are going to move because I don’t think we’re done with COVID yet. I think this is something we’re going to be dealing with for a while.

Take COVID as an example and the data that’s coming out of COVID. We are part of the largest clinical trial in history and because of that, the amount of data that is being collected citywide, regional wide, statewide, countrywide, WHO. Everybody is collecting data on this and being able to amalgamate that data into a way that you can disseminate it to make sense of it, is just astounding where vaccines should have taken 10 to 20 years to be able to create in eighteen months.

We’re dealing with MSRA that technologies that already existed and it was just using an existing platform to be able to come up with a vaccine that was going to be able to work but it’s pure data and analytics that allowed for that to happen, the worldwide collaboration. Is that where MyConnectedHealth is heading, to be able to enable that level of cooperation?

Yes, it is. I do believe that we are going to see different variants and strains of COVID. I’m not a doctor. I have no medical knowledge whatsoever but I do read things very carefully, I can piece things together and I tend to be in that futuristic realm anyway. Predicting how things are going to turn out, I think we’re going to be dealing with this for a little bit longer than we thought.

What’s concerning is we may start seeing more bio weaponized diseases and viruses coming out. One of the things that concern me is, with the genetic database being much more prolific, I think that needs to be guarded and secured incredibly. The thought of a nation-state actor getting access to those databases, wouldn’t be a very difficult step to go from there to creating bioweapons that are genetically targeting a specific population.

Your clients are government healthcare systems. Is it not designed for you and I and the average person to be able to use this?

It is.

Having too much information takes you down rabbit holes. Share on X

We as average users can buy into this system and be able to import our information and be able to utilize it.

When we’ve launched, yes.

We’re still pre-revenue. We do get that but the thought process behind this is the future of where biotech is heading will enable us to have them both as individuals, as doctors and as societies, a far more congruent idea of what the entire puzzle looks like instead of those 30 pieces that are right in front of us. Let me ask you one last question then I’ll let you go. We’ll make sure that how people can get in touch with you. Is there a website that people can reach?

MyConnectedHealth.net. You can also find me on LinkedIn and Twitter @TylerCohenWood.

The question I ask everybody and this is something I ask, as you leave a meeting, you get in your car and you drive away, what’s the one thing you want people to think about you and MyConnectedHealth when you’re not in the room?

What we’re trying to do is we’re trying to change the way that humans interface with technology to help diagnose, give people with a diagnosis a better chance, to bring the world together and be able to give healthcare to every single person on this planet.

Tyler, thank you for being a wonderful guest. Thanks for the insights, showing us things that I hadn’t even thought about. It’s amazing that this is where our world is going and it gives me faith for the future. Thank you for being such an amazing guest.

Thank you so much for having me.

Important Links:

• https://www.LinkedIn.com/company/myconnectedhealth/
• https://Twitter.com/MyCH_Inc
• https://www.Facebook.com/Myconnectedhealth-inc-113598890647907
• MyConnectedHealth
• Catching the Catfishers
• LinkedIn – Tyler Cohen Wood
• @TylerCohenWood – Twitter

About Tyler Cohen Wood