Surgical Room Understanding

Real-time multimodal pipeline for surgical room understanding at Labib AI

At Labib AI, I engineered a real-time surgical room understanding pipeline that integrates speech recognition (Whisper-3), language understanding (Qwen-2.5), and pose estimation (HRNet) to monitor what is happening in the operating room, with the goal of reducing wrong-site surgeries.

The system was demoed to healthcare professionals through a Django web application.