Lunit, a leading medical AI provider, announced that it signed a license agreement with Cathay Life Insurance, one of Taiwan’s largest insurance companies. Through this contract, Lunit INSIGHT CXR—Lunit’s AI solution for detecting abnormalities via chest x-ray image—will be officially integrated into Cathay Life’s current underwriting workflow.
“This is the first time Lunit’s AI has been officially integrated into an insurance underwriting process, and this proves that our solutions can also bring value to workflows outside of hospital settings,” said Brandon Suh, CEO of Lunit. “We are excited to expand the reach of our AI-powered software and become the driving force of digital innovation in the global insurance market.”
Cathay’s adoption of the Lunit INSIGHT CXR software comes after a two-year test period between the two companies. In 2020, Lunit and Cathay Life launched a proof-of-concept (POC) to demonstrate the value of Lunit’s AI solution. Through the POC, the insurance provider, which reviews 30,000 chest x-ray images annually, deployed Lunit INSIGHT CXR to evaluate the impact of the software in Cathay’s x-ray review process, especially in terms of profitability, efficiency, and accuracy.
Also Read: Flipside Crypto Launches SDK for Blockchain Analytics Accessible via SaaS NFT
Nearly 200 chest x-ray images were used as samples in a retrospective analysis using Lunit’s AI. The cases were reviewed internally by radiologists working with Cathay Life to confirm the algorithm’s accuracy in identifying abnormalities. The results showed that the AI’s detection rate for positive cases (sensitivity) was 83% and 92% for negative cases (specificity). The overall accuracy rate was 90%. Furthermore, an analysis of the economic benefit of employing Lunit AI estimated up to a 20% increase in reading accuracy, and a decrease of up to 90% in overall reading time.
Cathay Life’s Underwriting Department said, “Lunit’s AI algorithm accurately detects small nodules and pulmonary fibrosis, which are not easy to distinguish on chest radiographs. Furthermore, the AI-aided review process significantly reduces the reading time, compared to the previous manual review process.”