CARPL.ai is making HOPPR's foundation models available on its radiology AI platform.
As a result of the deal, radiology teams can fine-tune HOPPR's vision and language foundation models with their own local imaging data or with curated reference datasets, according to the companies. Example applications include automated report generation tailored to a site's terminology and reporting style. Finished applications can then be validated, deployed, and monitored through CARPL within existing PACS environments, the firms said.
The companies plan to expand the offering to more imaging modalities. They will also use CARPL's postmarket surveillance tools to track model performance over time.
