グーグルのAI「ジェミニ」、試験中に他社へサイバー攻撃 米報道

Korea's Yonsei University has received funding from the Ministry of Science and ICT and the Korean Research Foundation to develop an AI-controlled cerebral vascular intervention robot over the next three years. The research team, led by Professor Hyo-seok Hong from the Software Integration Department, will operate the robot laboratory and collaborate with professors from the Mechanical Engineering Department at Kyunghee University, Software Integration Department at Kyunghee University, and the University of Toronto.

The goal is to create a robot that can autonomously diagnose and perform cerebral vascular interventions, replacing the need for doctors to manually manipulate instruments.

Currently, cerebral vascular interventions rely heavily on the doctor's experience and hand-eye coordination, which can lead to inconsistencies in treatment outcomes. This research aims to address these limitations by developing a flexible and rigid hybrid robot that can dynamically adjust its stiffness and flexibility based on the patient's vascular condition. Additionally, the team plans to use medical twins to virtually reconstruct patients' vascular systems and AI for pathfinding and treatment guidance.

The team will also focus on overcoming data scarcity issues, using a single MRI scan to create a 3D vascular map and a generative AI to augment lesion data. By integrating machine learning with human expertise, the research team aims to create an autonomous robotic system for cerebral vascular interventions. This system could potentially reduce treatment time by 20%, increase successful pathfinding rates to 95% or higher, and improve safety by 30%.

The ultimate goal is to improve the quality of care in regional hospitals and enhance emergency response capabilities.