Research On Tactile Sensor And Palpation-Based Tumor Detection Technology For Robot-Assisted Minimally Invasive Surgery

In traditional open surgery,since the stiffness of tumor tissues changes significantly,surgeons can obtain the location,size,shape and other relevant pathological information of tumor tissues through palpation to determine the scope of resection.In recent years,minimally invasive surgery has been developed rapidly due to its advantages of small surgical trauma,quick recovery and small surgical risk.Then there comes robot-assisted minimally invasive surgery,which has the advantages of accurate motion positioning,small vibration,and short operation time.However,in minimally invasive surgery,traditional palpation is no longer applicable because the surgeons cannot directly contact the surgical tissue and the medical instruments are not able to provide tactile feedback,which lead to the difficulty of performing intra-operative tumor localization.Meanwhile,surgeons may inadvertently cause damage to healthy tissue and even injure vital organs by performing surgeries without tactile feedback,raising the demand for doctors' skills and experiences.By introducing tactile sensors into minimally invasive surgery,establishing tactile feedback for minimally invasive surgery robots and realizing accurate tumor localization with tactile feedback,the problems mentioned above can be effectively improved.In order to meet the needs for tactile feedback and tumor localization in robot-assisted minimally invasive surgery,a tactile sensor for minimally invasive surgery robots and an autonomous robot-assisted palpation-based tumor detection method are designed in this paper.Firstly,according to the functional requirements for tactile sensors in minimally invasive surgery,a stiffness tactile sensor based on piezoelectric vibration is developed in this paper.With a unique design of the double cantilever beam structure,the sensor not only can be used in contact detection and tissue stiffness measurement through resonant frequency shift and changes of electric impedance of piezoelectric ceramics,reconstruct the tactile sensation for minimally invasive surgery,but also has the sensitivity-tunable ability for testing samples in different stiffness range and ensure accuracy of the stiffness measurement results.The contact model between the sensor and tissue samples is simplified to a two-degree-of-freedom vibration system,and the resonant frequency formula of the system is given,theoretically verifies the feasibility of the sensor for measurement tissue stiffness and sensitivity-tunable ability.The design is verified by finite element simulation,and the actual sensor is fabricated and the experimental platform is set up.The sensor is calibrated by the stiffness measurement experiment of the several silica gel samples.Then the ability of the sensor to realize tumor detection is verified by the lump detection experiment of pig liver tissue.Secondly,in order to realize tumor localization and shape detection,this paper studies a fast tumor detection algorithm based on gaussian process regression.The algorithm is able to estimate the location and shape of tumor with local stiffness measurement results of tissue sample,without the dependency of the surgeon's experience and automatedly select the optimal probe points,so as to continuously optimize estimation results and finally get the exact tumor location and shape.By studying and improving the selection strategy of palpation sampling points in the algorithm,the improved algorithm can simultaneously detect multiple tumors in the sample and greatly reduce the total moving distance of the sensor,thus shortening the detection time and improving the detection efficiency.Finally,a robot autonomous tumor palpation detection system with six-axis industrial robot as the target is developed with the tactile sensor and palpation-based tumor detection algorithm,and the hardware components of the system is designed and selected.The procedure of tumor detection and the software system are designed.And then an experimental platform for autonomous tumor palpation detection is built,and the ability of the system to autonomously detect and determine the location and shape of tumor through palpation is verified with several silica gel samples embedded with hard objects.