CliniSense AI Automates Clinical Skills Assessment with Real-Time Feedback

What happened

Researchers developed CliniSense AI, an end-to-end pipeline that transforms raw clinical audio into OSCE-aligned performance assessments using automatic speech recognition and speaker diarization. Tested on 270 simulated consultations, the system achieved Cohen’s κ = 0.824 and sub-1.5-s latency. It enables faculty to remotely monitor trainee feedback across multiple rooms.

Why it matters

Current manual OSCE evaluations suffer from low inter-rater reliability (κ = 0.35–0.58) and slow feedback. CliniSense AI provides real-time, competency-aligned assessment from raw audio, surpassing human annotator agreement (κ = 0.721) and revealing hidden diagnostic reasoning patterns.