An Intravascular Catheter Bending Recognition Method for Interventional Surgical Robots

Robot-assisted interventional surgery can greatly reduce the radiation received by surgeons during the operation, but the lack of force detection and force feedback is still a risk in the operation which may harm the patient. In those robotic surgeries, the traditional force detection methods may have measurement losses and errors caused by mechanical transmission and cannot identify the direction of the force. In this paper, an interventional surgery robot system with a force detection device is designed and a new force detection method based on strain gauges is proposed to detect the force and infer the bending direction of the catheter in the vessel by using BP neural network. In addition, genetic algorithm is used to optimize the BP neural network, and the error between the calculated results and the actual results is reduced by 37%, which improves the accuracy of catheter bending recognition. Combining this new method with traditional force sensors not only reduces the error caused by the traditional mechanical transmission, but also can detect the bending direction of the catheter in the blood vessel, which greatly improves the safety of the operation