AiThority Interview with Tatyana Kanzaveli, CEO and Co-founder at Open Health Network

Hi Tatyana, please tell us about your journey in the healthcare industry and how you arrived at the Open Health Network?

I started Open Health Network after I was diagnosed with cancer. I had a healthy lifestyle and had gone to the doctor for a routine check-up. My diagnosis was sudden and unexpected. No one could give me a good explanation on what caused it. I drew a diagram that displayed all the data sources that might have some impact on people’s health and decided to start a company that uses artificial intelligence to provide insights on causations, treatments and impact outcomes of complex diseases.

As a woman in technology, how do see the recent paradigm shift in data science and AI industries employing more female analysts and women who code to lead innovative product development projects?

I personally know amazing women who are top experts in artificial intelligence. I also see an overall trend of companies hiring diverse candidates for tech roles. This creates an amazing opportunity for women in tech to get great positions in the AI space with top-notch organizations.

Recommended: AiThority Interview with Nelson Petracek, CTO at TIBCO Software

Any key learning from these domains that you would like to share with our readers: 

If I understood this question correctly, you are interested in my insights on what I have learned in the healthcare industry?

Throughout my career, I’ve worked with C-level executives across many industries: high tech, manufacturing, hospitality, automotive and others. Healthcare is by far, the most complex industry for tech companies to work with. It took us a while to get a good understanding of what can and can’t work in this industry. We pivoted many times and our current product is very different from the one we started our company with.

There are very little incentives to change in healthcare. It is critical to learn early on who your customer is: who will pay for the product and why. The good thing is that we’re slowly moving to a value-based care model. That will force healthcare to change and start adopting more technologies. Quite a few innovations have also been triggered by the pandemic – time will tell how many of them will stay in their current state, evolve or disappear, going-forward.

Tell us a little bit about your recent association in The MILBox Project, and what was your partnership outcome with AWS?

I am very excited about the project we are doing with the Miller School of Medicine in partnership with AWS. The forward-thinking goal is to create a “first-in-its class,” data-driven, Digital Twin for a specific set of health issues. We created a sophisticated data lake and are getting lots of data from a variety of sensors, wearable devices, patient reported data, EHR, biometrical data, genomics and more! I expect us to be able to have a good first iteration of the Digital Twin model from this project in a few months. Once we have the validation of the model, it could be used for all sorts of predictive modeling (clinical, research, clinical trials and more).

Recommended: AiThority Interview with Sam Yoshida, EVP & GM of Canon USA

What are the major challenges in the delivery of personalized wellness solutions to patients? How do you bring data to the center of healthcare personalization projects?

As we all know, our healthcare system has been built on a notion of one-size-fits-all. As a result, we have not seen major improvements in overall health metrics, despite increased spending. We also try to over simplify things. For example, some think if we remind people to do their 10,000 steps a day –  they will do it. I wish it was that simple… We’re integrating advanced, motivational interviewing techniques and behavioral interventions into care plans to increase adherence. Personalization is a key piece of the puzzle. For example, we can analyze patients’ data for a previous week/month and use machine learning algorithms [on data collected from this and other patients] to suggest very personalized adaptations to interventions that will most likely work for this specific person. 

What kind of compliance framework should healthcare leaders have in place to ensure data teams don’t breach PII data governance policies? How are you doing at MILBox project?

Great question! The solution must be two-fold — technology and people. From a technology standpoint, healthcare organizations need to safeguard the privacy of patient-related data.It must be securely managed to ensure regulatory compliance across the board, but it also must be readily accessible to healthcare providers. 

Ultimately, the people involved can make the biggest impact. The only way to ensure the guidelines are faithfully followed – even in the face of everyday inconvenience – is to make sure everyone in the organization understands the mission, requirements and reasons for the specific protections in place.

With MILBox, we’re following the guidelines set by the Human Subjects Research IRB, and working closely with University of Miami to establish the procedures and safeguards.

Recommended: AiThority Interview with Patricia Nagle, President, Americas at TeamViewer

Tell us how is Open Health Network expanding on cloud-based technologies for AI, Blockchain, and Big data capabilities.

The new frontier in healthcare is in the development of Digital Twins. A Digital Twin is a simulation model that encapsulates a multi-faceted view of the patient, and allows doctors and researchers to test the efficacy and safety of proposed interventions in-silico. A mature Digital Twin technology will improve outcomes across the spectrum of treatment areas and disease states, and decrease the amount of complications and adverse events.

In order to make Digital Twin models a reality, a veritable mountain of data on the individual patients is needed. The first step in obtaining this data is breaking-down the silos between data stored in EHR, home devices and sensors, claims data on the payer level, as well as environmental and socioeconomic data.

OHN is working to implement a Digital Twin model of sleep and its effect on cardio-vascular diseases in underserved communities, incorporating such diverse data sources as environmental sensors (air quality, light and ambient noise), wearable devices and EHR data. The data is harmonized and stored in an OHN data lake. Our data lake implementation is based on FAIR principles (Findable, Accessible, Interoperable and Reusable).

Your plans for 2022: Where is PatientSphere2.0 heading to in the coming months?

We’re very pleased with the amount of interest PatientSphere, PatientSphere for Cardiology and the PatientSphere based Digital Twin in Healthcare received. Our plans for PatientSphere include: disease-specific offerings similar to the one we did for cardiology; adding support for more wearable devices and sensors; and releasing Digital Twin as a service to be used by researchers, pharma companies, clinicians and other players in the healthcare industry.

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