Image Guided Surgery in Prostate Brachytherapy

1
Image Guided Surgery in Prostate Brachytherapy

• Rohit Saboo

2
Prostate Cancer

• A growing problem in US and world over with
  increasing longevity
• Methods of treatment
• Surgery
• Irradiation
• Many problems of existing methods

3
Where is it?
4
Brachytherapy procedure

• Localized and prolonged dose
• Brachytherapy overview

5
Brachytherapy - old way

• Pre-planning CT
• Outline prostate
• Develop plan for needle movement
• Guide needle at time of surgery with help from
  Ultra sound images

Dynamic Brachytherapy of the prostate under
active image guidance, Gang Cheng et. al, MICCAI
2001
6
Brachytherapy old way

• Problems
• Prostate movement
• Prostate size/shape variance
• Due to anesthesia effects
• Time and hormonal therapy
• Drawbacks
• Lots of error due to prostate movement
• More than necessary needle insertions

7
Brachytherapy new way

• General outline of prostate from pre-planning CT
• Outline prostate in real-time during surgery
• Provide guidance for the needle with real time
  prostate outlining.
• Track needle errors in real time

8
Steps in automation

• Acquire volumetric ultra-sound images on the fly
• Automatically recognize the prostate/rectum and
  other structures (segmentation)
• Analyze dose distribution

9
Segmentation

• The process of outlining the prostate (or any
  organ) is called segmentation
• Two chief ways to deal with it
• Model-based
• Image based

10
Segmentation problem
11
Ultrasound problems

• Noise!
• Speed of sound is not uniform
• Image distance incorrect in one axis

12
Approaches to segmentation

• Model based
• Shape model
• Probable shapes
• Probable intensity/texture variations
• Examples ASM, AAM, M-reps
• Image based
• Outline drawn by expert on one image (atlas)
• Image intensity/feature based registration
• Outline carried over

13
Feature Model Ruo et. al

• Set of boundary points
• i - Sample object
• Xi Tuple representing ith object
• Each object has m points on the boundary

14
Feature Model

• Xi
• (Li, ri1, ri2, rim )T

Rim
Li
15
Shape variation

• Mean shape
• Covariance matrix

16
Feature Model

• Eigenvalue decomposition of covariance matrix
• pi Principal components (eigenvectors)
• Eigen-values
• Sort the eigenvalues, choose the largest t

17
Eigenvalues
18
New plausible models
19
Optimization - GA
20
Image match

• Fitness function
• outi and ini average of intensities along a
  profile (15 pixels long)

21
Image Match

• Simplify and speed it up
• vi unit normal

22
GA parameters
23
GA parameters

• 90 crossover rate
• 1 mutation rate
• population size 200
• 2000 generations
• repeated 15 times

24
Experiment

• 40 images
• training from 27, 3 poor quality
• test on remaining 10
• Expert segmentation by two different experts
• human-human disagreement vs automated-human
  disagreement

25
Results
26
Results
27
Results
28
Results
29
Summary

• Very good analysis
• Point based boundary models are poor
• Parameter tuning
• No reasoning for fitness function

30
Methods

• Model based methods
• Image based methods
• Fully automatic, registration
• Deformable registration Wei, 2004
• Snakes

31
Image based

• Overview
• MRI/MRSI and US data
• prostate carefully outlined on MRI data
• US image is acquired during operation time
• The two images are brought into alignment during
  operation time.
• They do for biopsy, but same techniques work for
  brachytherapy

32
Global Alignment

• US data is poor
• Model is pre-segmented in MRI
• Surface to volume alignment methods are used
• Gradient of image is computed in US and model
  information is used to roughly find the
  correspoding boundary in US

33
Global alignment

• GA based optimization
• fitness function

34
Registration

• Deformation is elastic
• Two orthogonal directions
• Therefore 2-parameter model for deformation

35
Registration

• Obtain Curvature images
• Obtain mapping g for a few points
• Use these points to drive a TPS deformation.

36
Validation

• Phantom
• Gelatin made prostate phantom
• 15 fiducial markers implanted inside the prostate
• Soft container filled with water placed on top to
  simulate pubic bone.

37
Validation
38
Results on phantom
39
Patient study
40
Patient Study
41
Summary

• Two step registration technique
• Phantom Studies
• Only tested over one patient

42
Methods

• Model based methods
• Image based methods
• Fully automatic, registration
• Deformable registration Wei, 2004
• Snakes
• semi-automatic
• In between both

43
Snakes

• Give an initial approximate contour
• Two forces act on a snake
• Internal force
• Based on curvature
• External force
• Based on image gradients
• Let the model evolve using ordinary force
  equations till equilibrium

44
Snake based methods

• Approach used by Zhouping Wei, 2005

45
Snake based methods

• 1.19 /- 0.14 mm on average

46
Questions?
47
References

• Dynamic Brachytherapy of the prostate under
  active image guidance, Gang Cheng et. al,
  MICCAI, 2001
• Automatic Prostate Boundary Recognition in
  Sonographic Images Using Feature Model and
  Genetic Algorithm, Ruo Yun et. al, Journal of
  Ultrasound in Medicine, Vol 19, Issue 11, 2000
• Deformable Registration Between MRI/MRSI and
  Ultrasound Images for Targeted Robotic Prostate
  Biopsy, Wei Shao et. al, Proceedings of the 2004
  IEEE Conference on Cybernetics and Intelligent
  Systems
• A Discrete Dynamic Contour Model, Steven
  Lobregt et. al, IEEE Transactions on Medical
  Imaging, vol 14, no 1, March 1995
• Dynamic Intraoperative Prostate Brachytherapy
  Using 3D TRUS Guidance with Robot Assistance,
  Zhouping Wei et. al, Proceedings of the 2004 IEEE
  Engineering in Medicine and Biology, 2005