JMIR Publications

ABSTRACT

The administrative burden in anesthesiology has reached a critical tipping point, as the digitization of healthcare via Electronic Health Records (EHRs) often forces clinicians to spend more time interacting with screens than with patients. In the high-stakes, high-velocity environment of the Operating Room (OR), this "documentation tax" competes directly with cognitive vigilance. While "ambient AI scribes" that listen to and transcribe patient encounters are revolutionizing outpatient care, they remain largely ineffective in the perioperative setting, where care is a complex choreography of physical actions, physiological monitoring, and silent vigilance rather than mere conversation. This Viewpoint argues that the next generation of AI documentation in anesthesiology must evolve from unimodal "listening" to multimodal "sensing." We propose the concept of the "Visual Scribe," an ambient intelligence system integrating Computer Vision (CV) with audio and telemetry data to automatically document the physical reality of surgical care. Synthesizing current research on AI-enabled perioperative workflow analysis, we explore how CV algorithms—such as temporal action localization and pose estimation—can segment surgical cases into granular phases with superhuman precision. Contrasting the retrospective imprecision of manual documentation with the real-time capabilities of multimodal AI, we highlight how emerging architectures can accurately detect and timestamp critical "silent" events like patient transport, intubation, and incision. Automating these data points can drastically reduce clinician burnout, reveal hidden provider-level workflow variability, and enhance patient safety through real-time "sterile cockpit" monitoring. To address the ethical, medicolegal, and ergonomic implications of deploying "always-on" visual sensors, we emphasize the need for a paradigm shift in privacy engineering, utilizing edge-based skeletonization to mitigate surveillance concerns. Ultimately, by equipping the EHR with "eyes" as well as "ears," we can create a self-documenting operating room, transforming the EHR from a distractor into a silent, autonomous partner that restores the anesthesiologist’s unwavering focus to the patient.