Research Of Automatic Needle Puncture Mechanism Based On Video Image Recognition

Technology of needle biopsy and targeted therapy are minimally invasive medical treatments which have terrific prospects in application, whereas the accuracy of the puncture track is one of the influential reasons to the result. Puncture accuracy heavily depends on the operator’s skill. In the complicated puncture process, even a very experienced doctor still cannot guarantee the success. Thus, here we propose a resolution, automatic needle insertion path planning algorithm, based on the analysis in the characteristics of the shape of the focus. It is an exploration to the possibility of automatic planning of the targeting acupuncture robot’s needle insertion path based on the intelligent image processing.For accurately locating and recognizing the focus region, we first use an improved snake model to segment the medical ultrasound image of the focus region and gain the contour of the focus region. Afterwards, by analyzing the concavity and convexity of the contour we gained, we describe the shape features of the focus region, divide it into several sub-regions and make the centers of the sub-regions to be candidate points of puncture target points. Consequently, we can ensure that the effect of puncture, increase choices of puncture target points and provide the basis for the determination of targeted puncture target. Eventually, since it is possible to have obstacles in the process of puncture, we explore the algorithm of obstacle avoidance and path planning which meet the requirements of puncture. With the concerns of vascular, nerve and other significant tissues, we request a distant research with the obstacle in the avoidance process. Besides, the big angle adjustment harms tissues more with more pains and dangers, which limits the adjustment angle in a small region. Thus, traditional obstacle avoidance in the robot navigation is not suitable for needle puncture path planning. So, we propose an S-type double-arc path planning algorithm for automatic needle puncturing. Through the evaluation of the relative position among the needle, the obstacle and the target, we explore the solution of the puncturing track which is suit for needle’s characteristics and plan the path of needles automatically. It will evaluate the applicability of automatic needle puncture, and transform the planned path into adjustment signal sequence for automatic needle insertion of targeted puncture robot and make path planning that is suitable for needle puncture possible.This method is definitely a beneficial exploration to achieve automatic needle insertion path planning of targeted puncture intelligent robot. By analyzing and guiding puncture needle in real time, it helps to avoid obstacles in the safest path, reasonably plans for puncture needle’s insertion, guaranteeing the best path and the most effectively stabbing the parts of focus region. Meanwhile, it ensures the probabilities of success in the puncture and lowers dangers and harms of needle insertion in a valid way.