Follow Datanami:
July 27, 2020

New ‘Deep Claim’ Algorithm Could Save Patients, Hospitals Major Money in the Insurance Claims Process

The for-profit healthcare system in the United States often seems inscrutable, with patients bracing for unpredictable (and often overwhelming) costs that may not be covered by their insurance provider. Now, Silicon Valley-based automation firm Alpha Health is announcing the development of a new algorithm aimed at helping patients navigate the uncertainties of corporate healthcare by predicting healthcare billing claim denials.

The model, called Deep Claim, predicts both when and how much an insurance company will pay for a given claim in advance of any payment they make. It was trained with three million de-identified claims including parameters like demographic information, diagnoses, treatments, and billed amounts. Using this information, Deep Claim can not only predict the date and amount of payments with reasonable certainty, but also the likeliest reasonings for any claim denial in play.

“Deep Claim is an innovative neural network-based framework. It focuses on a part of the healthcare system that has received very little attention thus far,” said Varun Ganapathi, co-author of the paper and Co-Founder and Chief Technology Officer at Alpha Health. “While much attention has focused on the potential of artificial intelligence and machine learning in diagnostics and drug discovery, this paper demonstrates the opportunity to apply these same approaches at scale to the back office of healthcare which could save the U.S. billions annually in wasted healthcare spending.”

According to Alpha Health’s internal research, Deep Claim performs about 22% better than “the best baseline system” serving the same function. The researchers are hopeful that using the ability to predict denials, they can save patients and hospitals untold amounts of money by automating the process and designing claims to achieve the maximum chance of approval before they have a chance to be submitted and rejected for technicalities.

“I am deeply honored to have my work and the work of the team at Alpha Health featured in the Spotlight Session alongside five other papers from prestigious academic research centers, including University of Cambridge, Johns Hopkins University and NASA Frontier Development Labs, among others,” said Byung-Hak Kim, lead author of the paper and AI Technical Lead at Alpha Health. “The fact that our model was trained on real-world claims data and that development included real deployment scenarios will enable us to integrate our research directly into our solution more quickly than a conceptual or theoretical research approach would otherwise allow. This helps us ensure that our research will directly benefit our health system customers as quickly as possible.”