Dissertation > Excellent graduate degree dissertation topics show

High Precision Surgical Navigation and Its Application

Author: ShenZuo
Tutor: LiuYunCai
School: Shanghai Jiaotong University
Course: Pattern Recognition and Intelligent Systems
Keywords: Surgery navigation high precision positioning computer vision predicting segmentation 3-D reconstruction of surgery instruments predicting weighted filter
CLC: TP391.41
Type: Master's thesis
Year: 2012
Downloads: 136
Quote: 1
Read: Download Dissertation

Abstract


With the rapid development of medical science, surgery put on more and more attention on minimally invasive and refinement. Surgical Navigation System(SNS) matches the patient preoperative image data and intraoperative patient anatomy structure accurately, and plays real-time navigation, which will attribute to higher operation accuracy, smaller operation trauma and less recovery time. According to the needs of SNS, this paper design a new generation of infrared surgery navigation clinical prototype system.This paper studies key technologies of SNS, including binocular camera calibration, stereo calibration, 3-D reconstruction of medical images. In addition, this article does some researches to increase the running speed and the positioning accuracy which are as follows:1. Propose the prediction segmentation algorithm. The surgery instruments’recognition typically uses matching algorithm, which is slowly. Therefore, the prediction segmentation algorithm is proposed, which regarded the trajectory of marks as quadratic curve. It uses the quadratic curve fitted by previous frame to figure out the current prediction coordinate, and then the nearest neighbor algorithm is used to classify the current marks. The algorithm can effectively improve the recognition speed, and can gain high accuracy with template matching algorithm used in combination.2. Propose the high precision reconstruction algorithm of surgery instruments. Traditionally, we recognize the surgery instruments by the marks on themselves, whose coordinates are used to calculate the posture and work point of the instruments. This paper introduces the process of 3-D reconstruction of surgery instruments based on the traditional method. It doesn’t reconstruct the marks separately, but use the structure information as constraints of the projection functions to reconstruct the marks of one surgery instruments. The method can not only reduce the jitter greatly but also improve the accuracy effectively.3. Propose the prediction weighted filtering algorithm. It supposes the trajectory of marks and work point are quadratic curve. The coordinates in previous frame are used to fitting quadratic curve, with which predicted coordinates of current frame can be figured out. The final result is the weighted average of predicted and measured values. This algorithm, compared with common filtering algorithm, has a nice filtering effect while not affecting the real-time tracking. Based on the key technologies above, this paper develops the clinical prototype system of surgery navigation. The hardware includes binocular camera, instruments, focus objects and computer. The software is constituted by digital image acquisition, reconstruction of surgery instruments, reconstruction of medical image, as well as surgery navigation. The prototype system conducted two cases of clinical operation in the Shanghai Ninth People’s Hospital, which indicated that its performance had reached the international advanced level.

Related Dissertations

  1. Research on Nondestructive Detection Technology for External Qualities of Papayas Based-on Vision,S667.9
  2. Study on Detection and Grading of ’Jiro’ Persimmon’s External Quality Based on Computer Vision,S665.2
  3. Research on Inspection Technology of Dehydrated Garlic Slice Based on Computer Vision,TP391.41
  4. 3D Reconstruction Technology Research and Development Based on 2D Image,TP391.41
  5. Research and Implementation of 3D Reconstruction Based on Computer Vision,TP391.41
  6. Research on 3D Facial Texture Synthesis Based on Two Half-frontal Photos,TP391.41
  7. Three-dimensional Cloud-based Particle System Simulation Method and Study,TP391.41
  8. Research on Classification Method of the Wooden Board Color Based on Computer Vision,TP274
  9. Research on Learning-Based Low-Level Vision Problem,TP391.41
  10. Study on Vision Tracking Method of Moving Object Under Urban Traffic Scenes,TP391.41
  11. Research on the Leavers Information of Rice Based on Computer Vision,TP391.41
  12. Computer vision - based fabric defect automatic detection technology,TP391.41
  13. Design and Implementation of the Panoramic Virtual Tour System,TP391.41
  14. Research on Lane Detection Technology for Expressway,TP391.41
  15. Research on Inspection System for Product’s Surface Quality Based on Image Recognition,TP274
  16. Multi-touch Technology Research and System Implementation Based on Computer Vision,TP391.41
  17. Radiometric Compensation on Geometrically Complex Surfaces in Projection System,TP391.41
  18. IC package oriented positioning system computer vision research,TP391.41
  19. Rectangularized CSS Images and a Matching Algorithm,TP391.41
  20. Study on Rapid Detection and Accumulation Characteristic Model and Fertilizer Response of Lycopene on Processing Tomato,S641.2
  21. The Resarch and Development of Grinding Robot System,TP242

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device
© 2012 www.DissertationTopic.Net  Mobile