Robotic surgery is a field that has been progressing at full steam in recent years. After its AASTR system (Autonomous System for Tumor Resection), Johns Hopkins University has struck again. One of their teams developed an imitation learning system allowing a robot learn surgical skills by watching operation videos carried out by experienced practitioners.
The results of their work were presented to the Conference on Robot Learning (CoRL) which was held in Munich from November 6 to 9.
Artificial intelligence is coming to the operating room
The device is based on the da Vinci surgical system; Often referred to as a robot, it is rather a set of instruments, normally controlled remotely by a teleoperator. Robotic arms, miniaturized surgical instruments, high-definition 3D visualization system: a true jewel of technology. This is already widely deployed in hospitals around the world.
John Hopkins researchers have integrated into this robot a new approach to learning that shares its architecture with that of ChatGPT. However, where ChatGPT excels in manipulating words and text, this model “ robot speaks » through kinematics, precisely converting surgical movements into complex mathematical formulas. A bit as if he were giving a equation » to the robot to explain to him how to move.
To form it, scientists exploited a gold mine of surgical data. The team collected and analyzed hundreds of video recordings captured by the miniature cameras positioned on the articulated arms of the da Vinci robots during real interventions.
These sequences, initially archived for post-operative analysis, served as a learning corpus for this da Vinci 3.0. The richness of this database is explained by the omnipresence of the da Vinci system: nearly 7,000 units are currently in service across the globehandled daily by a community of more than 50,000 qualified surgeons.
The robot was therefore able to assimilate all the subtleties of surgical procedures by simple observation, as an intern would do alongside an experienced doctor. The recordings capture not only technical movements, but also variations and adaptations to different situations encountered in operation.
From observation to practice: impressive results.
The tests focused on three fundamental actions: needle manipulation, tissue lifting and suturing. For each of these tasks, the robot demonstrated a skill comparable to that of human surgeons. “ The model is fascinating: we only provide it with images and it predicts the robotic movements necessary for the intervention. For us, this is a breakthrough that opens the way to a new era in medical robotics » explains Axel Krieger, assistant professor in the department of mechanical engineering at Johns Hopkins.
The major innovation lies in learning relative movements (movements defined in relation to the current position of the tool) rather than absolute (movements defined in relation to a fixed point in space), making it possible to compensate for inaccuracies inherent to the da Vinci system. Thus, by learning relative movements, the robot can adapt in real time to small variations and inaccuracies of the systemadapt more easily and gain precision.
The researchers noted that the robot develops even abilities not programmed in advance. “ If the needle falls, it automatically picks it up and continues the operation. This behavior was never taught to him », Specifies Professor Krieger. Capabilities that will completely transform the approach to surgical robotics. Instead of manually coding each step – work that could take a decade for a single procedure – learning a new procedure now takes just a few days.
The team is currently continuing its work to extend this learning method to complete surgical procedures. If this system becomes democratized one day, the question of ethics arises. If an error occurs during a robot intervention, who is to blame? The supervising surgeon, the robot maker, the programmer, or the robot itself? To what extent should we allow robots to act autonomously? Develop solid ethical and legal frameworks will be imperative to guarantee the safety of such technology.
- A surgical robot learned to operate by observing videos using an imitation learning system inspired by ChatGPT.
- Capable of performing complex surgical procedures, its precision is comparable to that of human surgeons.
- This advance is a small revolution for surgery, but is accompanied by ethical and legal issues that should not be underestimated.
