Voice-Controlled Fluoroscopy C-Arm Simulator with LLM Natural Language Control

Webpage created 5/14/26; verified for accessibility 5/14/26; last updated 5/14/26

This online interactive simulation demonstrates voice control of a mobile fluoroscopy C-arm in a browser-based training environment. The simulator includes a virtual C-arm, patient/table geometry, lumbar spine anatomy, real-time simulated fluoroscopic imaging, and voice-command interaction for radiation-free practice.

In the simulator natural-language commands are converted from speech to text and interpreted by a large language model (LLM) to determine the intended C-arm action and movement magnitude.

Before execution, the system can preview the proposed C-arm movement. Safe movements are shown in green. If a collision with the patient or table is predicted, the simulator notifies the user and can suggest an alternative action. Optional human-in-the-loop confirmation allows the user to verbally confirm or reject the interpreted command before execution.

This simulator is intended for radiation-free education, workflow exploration, and research involving C-arm positioning, radiology technologist communication, natural-language control, and collision-aware device movement.

We recommend using the Google Chrome web browser.

Launch the voice-controlled fluoroscopy C-arm simulator at this link.

3D view of Fluoroscopy C-Arm with patient on a table, simulated fluoroscopy image, and voice-command interface