Intelligent Monitoring of a Surgical Assistant Robot Using Artificial Intelligence.

Funding: Public-private partnership project

The activity is part of the MIRAIA project, funded by MCIN/AEI/10.13039/501100011033 by the European Union "NextGenerationEU"/PRTR "4.


File reference number: CPP2021-008716




Recent years have seen a large advancement in the development and application of artificial intelligence to solving perception problems in robotics.

This enables the application of robotics to new and challenging domains, especially to unstructured environments in fields as diverse as industry and medicine. In this last field, surgery is being supported by new chirurgical assistant robots that support surgeons during the procedure: improving the precision of the procedure, reducing risks to patients and improving the final prognosis. The objective of MIRAIA is to develop new perception technologies that enable a robotic chirurgical assistant to act in a flexible way and react safely in a complex and not-structured environment such as an operating room.


Project goals:

This problem will be studied from three different perspectives: the integration of the perception hardware, the development of artificial intelligence models and the study of the integration of both onto a prototype for a new robotic system capable of running in real-time.

The project will focus on two main research areas:

  • Monitoring the patient (chirurgical access points for the robot) and the chirurgical instrumental in hard real-time, that is, within the control cycle of the robot.
  • The interpretation of the activity and obstacles surrounding the robot and within its workspace, to adapt its normal behaviour within a changing environment.


Ceit's role in the project:

Ceit will lead the research of new perception methods based on artificial intelligence (neural networks) and 3D vision systems like stereo and RGB-D cameras. The AI models will support the monitoring of the pose of key elements in the surgical scenario: the pose of the patient, the position of the robot and medical personnel.