Accuracy Analysis and Error Correction for Anatomical Landmarks Based Registration in Image-Guided Neurosurgery

摘要

Accurate patient to image registration is the core for successful and safe image-guided neuro-navigation.Point-matching is the most common technique in practice to achieve this registration.While SkinAdhesive Markers (SMs) are widely used in point-matching registration, a proper implementation ofAnatomical Landmarks (ALs) may overcome the inconvenience brought by the use of SMs.However,the accuracy and applicability of ALs registration in neuro-navigation have remained a controversialissue due to the great variability of the reported results and conclusions.Therefore, we are providing ananalysis for the accuracy when using different sets of ALs as well as the applicability of thecorresponding surgical fields.In addition, one of the significant factors that influence the registrationaccuracy at the target point is the distribution of the fiducial points.The optimal distribution may bedifficult to achieve either due to the limited number of distinct anatomical features on head surface oreven due to the poor planning of skin adhesive markers.Therefore we have also tried to overcome thisproblem and correct the quality of the registration in the areas that normally suffer low accuracy whenusing ALs-based registration.
　　In the accuracy analysis study, we propose a set of three configurations using nine ALs.Theseconfigurations are defined according to the required positioning of the patient's head during surgery andthe resulting distribution of the expected Target Registration Error (TRE).We first evaluated theseconfigurations by simulation experiments using real clinical data of 20 patients from two hospitals, andthen tested the applicability of them in eight real clinical surgeries of neuronavigation.
　　In the correction study, the proposed method is based on an initial point-matching registration using sixdistinct ALs as fiducial points, followed by selection of some surface points on the patient's head atlocations where natural ALs are not available to improve the distribution of the fiducial points.Theprojection of the surface points from patient space into image space reflects the errors introduced duringthe initial registration process.These errors can be identified and inspected in the image space bycalculating the distance between the projected surface point positions transferred with the standardspatial relation and also by calculating their nearest point positions on the head surface.This informationis then used to improve registration accuracy by adopting the calculated nearest points (the moreaccurate points) in the image space, instead of the projected positions, along with their correspondingactual surface points in the patient space, as additional pairs of registration points.
　　The simulation experiment of the accuracy analysis showed that, by incorporating a FiducialRegistration Error (FRE) of 3.5 mm measured in the clinical setting, the expected TRE in the wholeskull was less than 2.5 mm, and the expected TRE in the whole brain was less than 1.75 mm when usingthe configuration with all the nine ALs.A small TRE could also be achieved in the correspondingsurgical field, by using the other three configurations with less ALs.In the clinical experiment, theFiducial Localization Error (FLE) ranges in the image and the patient space were 1.4-3.6 mm and 1.6-5.5 mm, respectively.The measured TRE and FRE were 3.1 ± 0.75 mm and 3.5 ± 0.17 mm, respectively.
　　For the correction trials, experiments with real clinical data showed that when using the surface points tocorrect the initial registration transformation, the TRE decreased in the whole-brain area byapproximately 20％, and this improvement is more dominant in the posterior and superior parts of thebrain.
　　The ALs configurations proposed in the accuracy analysis provide sufficient registration accuracy andcan help to avoid SMs disadvantages if used clinically.The correction technique helps as well toovercome the naturally impaired distribution of the ALS, which is the most significant factor thatprevents their wide use in registration.The method also allows more precise selection of correspondingfiducial points than traditional ALs and without the need for tagging adhesive markers.Results showedan improvement in registration quality in the targeted area in all cases by this kind of correction.

目录

TABLE OF CONTENTS

LIST OF SYMBOLS AND ABREVIATIONS

ABSTRACT

CHAPTER 1 INTRODUCTION

1．1 Introduction

1．2 Design of the Thesis

CHAPTER 2 ANATOMICAL AND CLINICAL BACKGROUNDS

2．1 Anatomical Backgrounds

2．1．1 Surface Anatomy of the Head

2．1．2 Anatomical Landmarks of the Head

2．1．3 Anatomical Landmarks Used for Image Guided Neurosurgery

2．1．4 The Brain Structure

2．2 Clinical Backgrounds

2．2．1 Brain Conditions

2．2．2 Brain Procedures Utilize Image Guided Systems

2．2．3 Patients Positioning in Neurosurgery

CHAPTER 3 IMAGE GUIDED SURGERY SYSTEMS

3．1 Introduction

3．1．1 Motivations and Necessity of Image Guided Systems

3．1．2 History

3．2 Today’s Image Guided Surgery

3．2．1 Components，Phases and Technologies

3．2．2 Pre-operative Imaging

3．2．3 Pre-operative Planning

3．2．4 Display and 3D Visualization

3．2．5 Intra-operative Patient to Image Registration

3．2．6 Tracking Systems

3．2．7 Application Examples

3．2．8 Drawbacks and Challenges

3．2．9 Brain Shift

3．2．10 Accuracy and Errors

CHAPTER 4 THE REGISTRATION ALGORITHM AND TRACKING SYSTEM

4．1 Introduction to Registration Algorithms

4．2 Singular Value Decomposition Approach

4．2．1 The determinant of RΛ

4．2．2 Algorithm Summary

4．3 Conclusions

4．4 The Tracking System，Polaris

CHAPTER 5 LITREATURE REVIEW OF THE REPORTED USE OF ALs IN IGNS

5．1 Introductions

5．2 Literature Review

5．2．1 Reported Inferior Accuracy

5．2．2 Reported Sufficient Accuracy

5．3 Conclusions

CHAPTER 6 APLICABILITY AND ACCURACY ANALYSIS OF ALs POINT MATCHING REGISTRATION IN IMAGE GUIDED NEUROSURGERY

6．1 Background Summary

6．2 Material and Methods

6．2．1 TRE Estimation

6．2．2 The Experiments

6．3 Results

6．3．1 The Distribution of TRE on the Patient Head

6．4 Discussion

6．5 Conclusion

CHAPTER 7 CORRECTION OF ALs POINT MATCHING REGISTRATION IN IMAGE GUIDED NEUROSURGERY

7．1 Background Summary

7．2 Material and Method

7．3 Experiment

7．4 Results

7．4．1 The TRE at Different Positions on the Head

7．4．2 The Distributions under Different Displacements

7．4．3 The Distributions of TRE in Different Patients

7．5 Discussion

7．6 Conclusion

CHAPTER 8 CONCLUSION AND FUTURE PERSPECTIVE

References

LIST OF FIGURES

LIST OF TABLES

PUBLICATION LIST

ACKNOWLEDGEMENTS

声明