97.1% Medical Terminology Recognition Accuracy… GC MediAI's 'Doctor’s Companion AI'

"Internal Data Processing Infrastructure, an Advantage for Security"

"The essential factor in artificial intelligence (AI) that transcribes conversations from the consultation room into electronic charts ultimately comes down to 'how accurately the AI can understand what the doctor says.' The proprietary speech recognition model implemented in 'Doctor’s Companion AI' boasts a 97.1% accuracy rate in recognizing medical terminology. This figure surpasses the mid-90% range of general-purpose cloud speech-to-text (STT) models and the roughly 90% level of open-source models."

On August 26, at the 'GC MediAI Doctor’s Companion AI Media Day' held at Conrad Hotel in Yeongdeungpo-gu, Seoul, Jin-Tae Kim, CEO of GC MediAI, cited the "excellent proficiency in recognizing medical terminology" as the core strength of GC MediAI’s latest release, Doctor’s Companion AI.

Doctor’s Companion, the flagship electronic medical record (EMR) solution from GC MediAI, is the number one electronic charting system in the industry, used by over 16,000 medical institutions nationwide. Earlier this year, the company unveiled Doctor’s Companion AI, a new product equipped with AI functionalities. Doctor’s Companion AI is an AI clinical solution that instantly transcribes consultation conversations in real time directly into the EMR chart. After transcribing conversations between clinicians and patients through speech recognition, the solution structures information according to the SOAP format (Subjective, Objective, Assessment, Plan), seamlessly connecting patient summaries before consultations, follow-up management after visits, and even clinic management index checks in a single workflow.

CEO Kim stated, "Doctor’s Companion AI combines decades of accumulated expertise in running electronic charts with medical AI technology, creating a product that transforms work processes in the consultation room. By reducing the time spent on charting and code searching, it sets a new standard for the 'next-generation consultation room,' allowing clinicians to concentrate on patient care and meaningful interactions."

Doctor’s Companion AI consists of five main functions: ▲'STT·SOAP Charting,' which uses medical-terminology-specialized speech recognition to transcribe and organize consultation conversations in real time into the SOAP format; ▲'Historical Consultation Summarization,' which extracts only the key points from accumulated consultation records and test results; ▲'AI Disease·Prescription Search,' which analyzes consultation content and suggests appropriate disease codes and prescription candidates; ▲'AI Assistant,' which responds to queries based on past medical records; and ▲'Doctor’s Companion Insight,' which visualizes clinic management indexes, including patient volume trends. The AI not only drafts charts but also helps select billing codes.

For data security—a matter of highest sensitivity among private practitioners—the company highlighted its use of internal infrastructure. Medical data such as consultation conversations and records are not sent to external AI services, but are processed entirely within the company’s own infrastructure. Unlike approaches that call application programming interfaces (APIs) of general large language models (LLMs), all steps from speech recognition to summarization and structuring are handled by in-house models, which the company emphasizes as an advantage for security.

CEO Kim stated, "With the adoption of Doctor’s Companion AI, we will expand the role of EMR from a 'system for documentation and billing' to a platform supporting clinical practice and overall clinic operations." The company aims to evolve toward a 'Medical OS' that connects to external services via cloud-based EMR by March 2027.

This content was produced with the assistance of AI translation services.