A real-time, recursive, fact-first architecture reduces AI-generated ‘note bloat’ by 65% and cuts down on post-visit edits.

Corti has launched FactsR, a real-time agentic reasoning system designed to enhance ambient documentation during clinical consultations by reducing extraneous content and minimizing the need for post-visit edits.

The system integrates with ambient AI tools and uses a recursive, fact-first architecture to identify, validate, and structure clinical information—such as symptoms, vitals, and medications—as conversations unfold. According to Corti, early tests show that FactsR reduces AI-generated note bloat by over 65%, helping clinicians produce more concise and relevant documentation.

Unlike traditional large language models retrofitted for healthcare, FactsR is powered by Corti’s recursive fact-first reasoning loop, a purpose-built engine designed to surface, validate, and structure clinical knowledge in real time as conversations unfold. The tool is delivered as a modular API, allowing developers to embed it directly into healthcare applications.

Why a Recursive Approach Matters

Many traditional ambient documentation tools process raw transcripts through general-purpose language models after a consultation ends, often producing lengthy summaries that clinicians spend up to three hours a week correcting.

FactsR takes a different approach, from passive summarization to active reasoning. Its workflow consists of four key stages:

  1. Listen and Extract in Real Time: As the consultation unfolds, FactsR continuously identifies and surfaces structured clinical “facts”—such as symptoms, vitals, medications, and social history.
  2. Vet and Refine with Specialized AI: Each fact is automatically reviewed and improved through an AI-driven feedback loop. If something is unclear, the system refines it until it is accurate, consistent, and ready to use.
  3. Clinician-in-the-Loop: Clinicians can review, accept, or adjust facts as they go. Early adopters report far fewer post-visit edits and rarely need to add missing information after the consultation, according to Corti. 
  4. Generate EHR-Ready Notes: Once the facts are finalized, the system assembles a concise summary that’s free from long, verbose summaries or irrelevant content.

Reported benefits include:

  1. Less Screen Time: Early trials show that users spend minutes, not hours, on corrections
  2. Better Patient Focus: Real‑time reasoning means decisions stay in the consultation, not in hindsight.
  3. Audit‑Ready Transparency: Every fact carries a timestamp, confidence score, and link back to the conversation.
  4. Healthcare AI that Delivers: Reduces general-purpose AI-driven “note bloat” by 65%.

The innovation behind FactsR has been published together with evaluation results on the public benchmark Primock57 dataset. The evaluation shows that FactsR increases clinical completeness by 13%, capturing significantly more of the relevant medical information compared to traditional ambient scribes, while reducing note bloat by over 65% with a clinician in the loop.

FactsR is offered today through a consumption‑based API with enterprise‑grade HIPAA and GDPR compliance. 

ID 80884410 | Doctor Patient © Imtmphoto | Dreamstime.com

FDA Clears Voice-Assisted Interventional X-Ray System