Mosaic Clinical Technologies, the technology and AI services division of Radiology Partners, has acquired Cognita Imaging, a developer of vision language models (VLMs) for radiology.
With the acquisition, Cognita’s VLMs will be integrated with Mosaic’s MosaicOS enterprise-scale operating system, the firms said. (VLMs process visual information as well as text.) Cognita’s models have been incorporated into Mosaic Drafting, an application that analyzes x-rays and head CTs and drafts preliminary results for physicians to review and confirm.
Cognita, which has been collaborating with Radiology Partners and Mosaic for over a year, will continue to operate as an independent business unit with an AI focus.
According to the firms, the Cognita models have demonstrated up to 52% increased detection rates, a fourfold reduction in significant diagnostic errors, up to 76% shorter read times, and high acceptance of the results among radiologists, with 87% to 98% of the AI-generated results requiring one or fewer clinically significant edits.
Additionally, Mosaic said that it is expanding the commercial availability of MosaicOS nationwide, with certain components pending U.S. Food and Drug Administration (FDA) authorization.
Cognita’s models are currently undergoing Institutional Review Board-approved trials as part of the FDA authorization process.













![A normal mammogram confirmed by three-year radiologic follow-up illustrates reader-marked regions of interest (ROIs) during (A) unaided (round 1) and (B) artificial intelligence (AI)–assisted (round 2) reading. Each colored dot represents an ROI for recall by a human reader. Readers could mark more than one ROI per case, represented by multiple dots of the same color. During AI-assisted reading, the AI system displayed three visible prompts: two with suspicion of malignancy scores of 35% (left mediolateral oblique [L MLO] and craniocaudal [L CC]) and one with a suspicion of malignancy score of 10% (right craniocaudal [R CC]), shown as polygonal overlays. Without AI, six of 10 readers (60%) marked a false-positive ROI. With AI assistance, this fell to two of 10 (20%). R MLO = right mediolateral oblique.](https://img.auntminnie.com/mindful/smg/workspaces/default/uploads/2026/07/2026-07-14-radiology-mammogram-ai-auto-bias.H0bYO8QlWs.jpg?auto=format%2Ccompress&fit=crop&h=112&q=70&w=112)





