面向复杂病变的血管介入手术机器人控制和安全性研究

CASIA OpenIR > 毕业生 > 硕士学位论文

	面向复杂病变的血管介入手术机器人控制和安全性研究
	赵含霖
	2022-06
页数	100
学位类型	硕士
中文摘要	血管介入手术已经成为治疗冠心病的主要方式，但手术过程中的 X 射线辐射导致医生患白内障和癌症的风险增加。血管介入手术机器人能使医生远离手术中的 X 射线辐射，从而改善医生的工作环境。然而现有的血管介入手术机器人缺乏多器械协同配合治疗复杂病变的能力，限制了机器人在临床的应用。本文在国家重点研发计划项目“面向复杂病变的多器械协同递送血管介入手术机器人关键技术及应用研究”（2019YFB1311700）、国家自然科学基金项目“面向复杂病变的血管介入手术机器人自主操控关键技术研究”（62073325）和国家自然科学基金重点项目“经皮股动脉介入术后体征监测与穿刺点即刻止血机器人关键技术研究”（U1913210）支持下，围绕血管介入手术机器人系统中的多器械控制和安全性等方面展开研究，本文的主要内容和创新点如下：（1）设计了多器械协同递送血管介入手术机器人，该机器人包含从端递送装置、主端操作台和定位机械臂。从端递送装置采用仿生结构设计，可以实现导丝的推送和旋转。单个主动轮和两个从动轮之间的协同工作实现对双导丝的交替控制，使得机器人能够完成复杂病变所需的高难度手术术式。主端操作台保留医生操作习惯，主从采用 TCP/IP 网络通信协议，结合可视化界面使医生实时获取手术信息从而进行远程手术。医生可调节定位机械臂使从端递送装置处于合适位置从而进行手术。（2）提出了一种导丝近端阻力的计算方法，可通过主动轮驱动电机的电流计算导丝递送时的近端阻力，并通过实验验证该方法的可行性。利用线性回归方法建立递送力与夹紧电机电流间的定量关系，从而施加合适的电流避免从动轮打滑。此外，为判断导丝递送状态是否异常，设计了基于神经网络的导丝递送状态判别算法。实验结果表明，本文提出的导丝递送状态判别模型识别准确率高、可解释强，能在导丝处于异常状态时及时提醒医生从而调整手术策略，提高介入手术的安全性。（3）通过实验评估了血管介入手术机器人在不同递送速度下的递送误差，在不同速度下，机器人最大递送误差小于 0.4 mm。通过人机对比试验，记录徒手和机器人在血管模型中递送导丝到达目标位置的时间和导丝尖端的移动距离，结果表明机器人手术性能可以达到徒手递送水平。在此基础上，利用本文所设计的机器人进行了远程动物试验，成功在活体动物的左回旋支和左前降支递送了双导丝和双球囊，验证了机器人系统在临床手术中的可行性。
英文摘要	Vascular interventional surgery has become the standard treatment strategy for coronary heart disease, but X-ray radiation during surgery puts doctors at an increased risk of cataracts and cancer. The vascular interventional surgery robot can keep doctors away from X-ray radiation during surgery and improve the working environment of doctors. Existing vascular interventional surgery robots cannot coordinate multiple instruments to treat complex lesions, limiting the clinical application of robots. Supported by the National Key R & D Program of China (Grant 2019YFB1311700) ，the National Natural Science Foundation of China (Grant 62073325) and the National Natural Science Foundation of China (Grant U1913210), this dissertation focuses on the multi-instrument control and safety of the vascular interventional surgery robot system. The main contents and innovations of this paper are as follows： (1) A multi-instrument co-delivery vascular interventional surgery robot is designed, including a slave delivery device, a master console and a robotic arm. The delivery device from the end adopts a bionic structure design, which can realize the pushing and rotation of the guidewire. Alternate control of double guide wires is achieved through the cooperative work between a single driving wheel and two driven wheels, enabling the robot to complete complex surgical procedures required for complex lesions. The master console retains the doctor’s operating habits and the communication between the master console and the slave delivery device is based on the TCP/IP protocol. The doctor can obtain the surgical information in real-time and perform remote surgery with the visual interface. (2) This paper proposes a technique for estimating the proximal resistance of the guidewire through the current of the driving motor, and experiments verify the feasibility of the method. A quantitative relationship between the delivery force and the clamping motor current is established by linear regression so that the appropriate current is applied to avoid slippage of the driven wheel. In order to judge whether the delivery state of the guidewire is abnormal, a guidewire state detection algorithm based on a neural network is designed. The experiment shows that the proposed model has high recognition accuracy and strong interpretability. The method can timely remind the doctor to adjust the operation when the guidewire is abnormal to improve the safety of interventional procedures. (3) The delivery error of the vascular interventional surgery robot under different delivery speeds is tested, and the experimental results show that the maximum delivery error is less than 0.4 mm. The human-machine comparison experiment records the time when the guidewire reaches the target position and the moving trajectory of the guidewire tip in the blood vessel model by hand and robot. The results show that the robot can reach the level of free-hand delivery. The designed robot carries out a remote animal experiment. Double guidewires and double balloons are successfully delivered to the left circumflex and left anterior descending branches of living animals, which verifies the feasibility of the robotic system in clinical surgery.
关键词	手术机器人血管介入力检测神经网络导丝状态判别
语种	中文
文献类型	学位论文
条目标识符	http://ir.ia.ac.cn/handle/173211/48499
专题	毕业生_硕士学位论文
推荐引用方式 GB/T 7714	赵含霖. 面向复杂病变的血管介入手术机器人控制和安全性研究[D]. 中国科学院自动化研究所. 中国科学院自动化研究所,2022.

条目包含的文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
赵含霖_面向复杂病变的血管介入手术机器人（13867KB）	学位论文		限制开放	CC BY-NC-SA