In a significant breakthrough in healthcare AI, GenHealth.ai has published a study that details an advanced generative AI model, which decisively outperforms established industry benchmarks by over 14% in healthcare cost and risk prediction. This model sets a new standard for precision in healthcare prediction, challenging traditional approaches by leading firms like Milliman, Cotiviti, and Johns Hopkins.
The study, titled “Introducing the Large Medical Model: State of the Art Healthcare Cost and Clinical Risk Prediction,” highlights how GenHealth.ai has leveraged a healthcare-native approach to generative AI to achieve superior results. The Large Medical Model (LMM) combines a unique vocabulary tailored specifically for healthcare, the technology of neural network transformers, and data from trillions of healthcare events across 140 million patients. The resulting model achieves a performance leap in the healthcare domain akin to the improvement seen in large language models over traditional natural language processing approaches.
“We’ve created the Large Medical Model to address a broad range of healthcare applications where LLMs and traditional analytics fall short,” said Ricky Sahu, CEO of GenHealth.ai. “Our healthcare-specific AI is built from the ground up to support everything from population health analytics to automating prior authorizations and detecting fraud, waste, and abuse. This isn’t just about streamlining administrative tasks—though our model excels there. This is the first step towards truly personalized patient care. It’s a shift from how healthcare organizations typically use data to find general patterns to a model where, in seconds, each patient and encounter can have better-than-human insight that is actionable. It will be equivalent to the impact large language models had on text interaction.”
Key highlights of the study include a new tokenization scheme that uses generative AI on healthcare-specific data to predict patient futures holistically, state-of-the-art performance in cost prediction, and accuracy in predicting a wide variety of chronic conditions. These features help insights transition from research into practice, reducing wasteful spending and improving patient management.
“Our team brings years of hands-on experience with healthcare claims and clinical data,” added Eric Marriott, CTO of GenHealth.ai. “This deep knowledge has given us unique insights crucial for achieving high performance and versatility in our AI model. We’ve been able to fine-tune it for a wide range of applications because we’re used to dealing with the nuance and complexity present in healthcare data.”
The GenHealth.ai team is now building applications on top of the Large Medical Model, including a co-pilot for population health analytics and prior authorization automation software, to make the LMM more accessible. “I’m so excited about what we’re building and how it can help make life easier for both providers and health plans,” commented Ethan Siegel, COO at GenHealth.ai. “We can use our AI systems to help reduce the amount of time people have to spend on manual processes so they can spend more time ensuring patients get the right treatment at the right time.”
