ARTIFICIAL INTELLIGENCE
Hello, I’m Maxence.
I am a PhD candidate at King’s College London, working with the SIE-CAI4CAI Lab since 2021 under the supervision of Prof. Sebastien Ourselin. My goal is to develop superintelligent and autonomous machines that make the world a better place.
Prior to my doctoral studies, I earned an MSc in Machine Learning, Computer Vision, and Robotics from the University of Surrey, where I conducted research at the CVSSP Lab under the guidance of Prof. Kevin Wells. My work is currently supported by UKRI funding.
Research Interests
- Surgical Video Understanding: Recognising the surgical environment from endoscopic videos to enable general understanding of the anatomical structures and procedural steps [ LoViT ].
- Expert Imitation Learning: Generating expert-like sequences of actions using video recordings from expert surgeons to train a basemodel with supervised fine-tuning [ SuPRA, SWAG ].
- Workflow Discovery and Planning: Generate multiple paths with a basemodel and use reinforcement learning to discover workflows that maximises the desired procedural outcomes for the patient, beyond human imagination [ In progress ].
Featured Projects
View All Projects
APPLICATION
Surgical Assistant App
Claude Language Model for Translating predictions into Recommendations.
COMPUTER VISION
Swarm Aerial Intelligence
Autonomous Drone (in progress)
ROBOTICS - DRONE
FPV Drone
Printing and Assembling a custom FPV eVTOL Drone
AERIAL INTELLIGENCE
Trajectory Planning
European Tech Defense Hackathon - Tactical Anticipation
Short Biography
Maxence's PhD research focuses on surgical video understanding, expert imitation learning, and workflow discovery and planning. He is also interested in the intersection of AI and robotics, particularly in the context of autonomous drones and aerial intelligence. Max has a strong background in machine learning, computer vision, and robotics, with a focus on developing innovative solutions for real-world challenges. He has published several papers in top-tier conferences and journals, including MICCAI and IJCARS.