-->

For a live demonstration, call 1-800-998-4874

The Birth of Cavo Health

The advanced technologies that power Cavo Health were honed in the ultra-demanding electronic discovery industry.

Curious how Cavo Health came to be? This next-generation product for computer-assisted medical coding has a fascinating story behind it. Cavo Health’s origins can be found in the launch of Cavo eD in 2009 for electronic discovery serving the corporate litigation. During the years that followed, Cavo eD evolved to become one of the most complete, feature-rich electronic discovery products on the market. Cavo eD pioneered several advanced search technologies, including a neural-network based Thematic Search and a superior predictive coding solution that uses three tunable similarity measures instead of just a single hardcoded measure. These advanced search technologies are especially important in corporation litigation since just a few emails out of millions can decide a multi-million-dollar case.
In 2016, the CEO of a health insurer asked Cavo eD to use its advanced search expertise to find a way of locating HCCs in medical records. His goal was to improve his ACA Risk Score because his Risk Transfer Payments were hurting his bottom line. With this request, Cavo Health was born.
The engineers at Cavo Health decided early on that the fuzzy logic of NLP machine learning technologies was not a good fit for the precise requirements of medical coding. Confirming the presence of an ICD in medical records is an exact science. So, the engineers invented a Precise Matching engine, that efficiently searches a medical record with “queries” that, when matched, confirm the presence of a particular ICD. The Precise Matching engine automated the exact work that medical coders do. When certified coders review the results of Cavo Health’s auto-coding, they can make decisions quickly, and review multi-hundred page medical records in minutes. More importantly, however, the Precise Matching engine can thoroughly and accurately review a medical record because it can look for hundreds of thousands of queries in a chart in seconds, thus finding HCCs missed with other methods as well as claims data not supported by the medical record.