Intraoperative Guidance Method Based On Fusion Of Endoscopic And Ultrasound Images For MIS

With the development of medical technology,image-guided computer-assisted interventional surgery is increasingly used in orthopedics,ENT,and neurosurgery.Although endoscopic image guidance is the mainstream surgical method today,it has the following limitations: it can only obtain image surface information but cannot perceive depth information,it cannot provide sufficient illumination to the surrounding tissues in the cavity,and the information source is narrow due to the lack of information Wait.Ultrasound images can detect lesions inside the patient’s tissue.Compared with endoscopic imaging,it can obtain depth information of the tissue.At the same time,ultrasound imaging has the characteristics of real-time and small damage to patients,which has irreplaceable advantages.Therefore,based on the two imaging methods of endoscope and ultrasound image,this paper researches the intraoperative guidance method based on multi-dimensional information fusion.First,for laparoscopic surgery,a multimodal image guidance method for laparoscopic surgical instruments was proposed.In the optical positioning system,because the relationship between the ultrasound probe and the ultrasound image needs to be determined,a platform and design process are established in this paper,and a set of model-based ultrasound calibration methods is established.The use of ultrasound image guidance system in a three-dimensional virtual environment accurately reflects the relative position relationship in the three-dimensional space between the laparoscopic surgical instruments and the ultrasound images scanned by the ultrasound equipment,combined with the laparoscopic surgical instrument position information and intraoperative images The information is used for surgical guidance,which helps to visualize the relative positional relationship between laparoscopic surgical instruments and lesions in a virtual environment,and realize the intraoperative multimodal information fusion of ultrasound images and endoscopic images.Secondly,for gastrointestinal surgery,a method of fusion of ultrasound and endoscopic images is proposed.This paper uses a hybrid label that can be applied to a cylindrical model,uses feature detection,pattern recognition,pose estimation,and other methods to estimate the fused plane pose and eliminates uncertain solutions in pose problems.This method meets the requirements of real-time performance,and has high recognition rate and robustness.Finally,the research of image fusion error compensation for digestive tract surgical instruments is proposed.Aiming at the shortcomings of existing gastrointestinal surgical instrument image fusion processes,which are greatly affected by the position of the marker,have insufficient accuracy,and are difficult to quantify their fusion results,an improved S-OPELM algorithm is used to perform fusion errors.make up.Experiments show that the errors are compensated in two aspects of position information and posture information,which can accurately reflect the posture relationship between the endoscopic image and the ultrasound image,and meet the accuracy and safety requirements of surgical guidance and operation.