Active Summarization: The Future of Medical Data Synthesis 🚀

How Machine Learning Solves Impossibly Hard Big Data Insurance Problems

In an in-depth interview with Carrier Management, Tomas Vykruta, co-founder and CEO of EvolutionIQ, explains how we’re bringing machine learning approaches to f …

Tomas Vykruta
March 9, 2022

In an in-depth interview with Carrier Management, Tomas Vykruta, co-founder and CEO of EvolutionIQ, explains how were bringing machine learning approaches to formerly intractable big data problems in claims management. Weve pulled out some highlights below or you can link to the full article here.

Insurers have massive data problems, but theyre tackling these problems by using older-style heuristics and rules-based methods, explains Tom on why he left Googles Applied Machine Learning team to focus on insurtech. My job was to see if I could transition them into a more modern approach by moving them to a modern AI system.

But he quickly learned that some of the hardest problems he had ever encountered in his career were in the insurance claims space. There are 70,000 different medical diagnoses described in the ICD [International Classification of Diseases] dataset, he says. The claims are changing constantly. One update will completely change the outcome of that patients health. Its just been an impossible problem for the examiners to solve in their heads. And theyre juggling 200 claims at a time and given new claims every month.

Enter EvolutionIQs platform, which is essentially a system that is able to guide the entire claims organization to know where to focus their time where to spend their time, where to not spend their time We are able to use AI to do a deep explanation and pick out the bits that really matter. Was there some change recently in a medical prognosis? Is there some combination of comorbidities in the health that indicates this patient will never recover?

At its core, the platform has a specialty focus on bodily injury claims. Such claims, Tom tells the magazine, are really incredibly complicated because youve got these narratives that are open for many years. Some of these claims are 15 or 20 years old. They have as many as 30 to 40 medical diagnoses. Our AI evaluates the full history of every claim up until today and makes a prediction: Is this a claim on which we can take action? Is there going to be an outcome that makes sense? And then from [the insurers] 50,000 claims or whatever they have, we are able to select the 10, 20 or 30 claims on that particular day that, if they take action, will lead to the greatest outcome to the claimant and the carrier and the client.

For example, EvolutionIQs AI generates scores that are then used to help guide examiners to action. Its like a credit score from 0 to 750. And the score indicates how resolution-ready or action-ready that claim is today, he says. We don actually make any decisions. Our system is guiding them and helping them understand where to spend their time.

One surprise in developing the technology was how the institutional knowledge of veteran claims adjusters became even more important, not less. There is this big movement into auto adjudication, reducing staff, Tom explains. Our view is we dont want to reduce the examiner staff Our early carriers are actually growing the team sizes because now they can see that the team is just becoming far more productive.

There are essentially two systems inside the platform, he notes one that makes predictions and one that explains the forecasted outcomes in the language of claims examiners. It gives them a really strong starting point to understand whats happening inside the claim. Before this, they had to read through potentially 150 pages of notes and try to make sense of it all.

Importantly, when it comes to unstructured data, such as the notes written by examiners on their internal systems that contain some of the most useful information, our system reads those notes like a human, and that typically gets us to a very accurate system.

Nobody has solved this before. We are inventing this key methodology that never existed Were not iterating on some solution and making it better. Were inventing this artificial intelligence guidance for carriers that nobody has done before Were not rules-based. We are taking a deep learning approach. We are able to understand every nuance of every claim. We are able to predict the overall outcome.

Read the full interview

Were hiring! Join us.

previous

Related posts

EvolutionIQ
Active Summarization: The Future of Medical Data Synthesis
Blogs
All
EvolutionIQ
5 Questions with: Sarah Schuermann, Vice President of People
Blogs
All