JCHO Hokkaido Hospital Launches AI Smartphone EMR Pilot for Clinical Note Drafting

Details

JCHO Hokkaido Hospital initiated a pilot combining standard smartphones, on-premise generative AI, and its EMR platform to automatically generate draft clinical notes. This closed-loop workflow captures doctor-patient conversations, transcribes them using an LLM, and creates a draft SOAP note directly within the EMR. A key feature is that all data processing occurs within the hospital's secure firewall, addressing privacy concerns. The system uses NTT Docomo Business smartphones to record natural conversation. The audio stream is transcribed by a high-accuracy AI speech recognition engine. Subsequently, an on-premise generative-AI server extracts key clinical elements and composes the draft SOAP note. This note is then transferred via the SMART on FHIR standard to the hospital’s EMR (MI·RA·Is V) for physician review and finalization. This initiative addresses documentation overload, a major cause of long working hours and burnout. Early testing with 50 repeat-visit patients showed a reduction of over 20% in the interval between patient entries, saving an average of 10 minutes 43 seconds per encounter. By automating drafting, physicians can focus more time on direct patient care. The use of an on-premise architecture mitigates privacy concerns that have limited cloud AI adoption in medicine. Furthermore, using ordinary smartphones makes the solution scalable to smaller clinics and rural hospitals with limited IT budgets. The pilot is currently running in the General Medicine department, tracking metrics like documentation time, physician satisfaction, and patient wait times, setting a benchmark for secure digital health innovation across Japan.