MEDICAL IMAGE ANALYSIS Final Project - 3D Breast Ultrasound - - PowerPoint PPT Presentation

▶
medical image analysis
SMART_READER_LITE
LIVE PREVIEW

MEDICAL IMAGE ANALYSIS Final Project - 3D Breast Ultrasound - - PowerPoint PPT Presentation

Mar 12, 2024 257 likes •451 views

VIBOT9 - Universitat de Girona Medical Image - 3D Breast Ultrasound Segmentation May 13th, 2015 MEDICAL IMAGE ANALYSIS Final Project - 3D Breast Ultrasound Segmentation Students: Flvia Dias Casagrande Marcel Sheeny de Moraes

slide-1
SLIDE 1

MEDICAL IMAGE ANALYSIS

Final Project - 3D Breast Ultrasound Segmentation

Students: Flávia Dias Casagrande Marcel Sheeny de Moraes

VIBOT9 - Universitat de Girona Medical Image - 3D Breast Ultrasound Segmentation May 13th, 2015

slide-2
SLIDE 2

Outline

  • Introduction
  • Active Contour
  • Level Set Segmentation
  • Algorithm Implemented
  • Preprocessing
  • Processing
  • Evaluation
  • Results
  • Conclusion

VIBOT9 - Universitat de Girona Medical Image - 3D Breast Ultrasound Segmentation May 13th, 2015

slide-3
SLIDE 3

Introduction

  • Objective: 3D breast cancer ultrasound segmentation
  • 5 images and their ground truths
  • Propose segmentation method

VIBOT9 - Universitat de Girona Medical Image - 3D Breast Ultrasound Segmentation May 13th, 2015

slide-4
SLIDE 4

Active Contour

  • Geometric alternative for “snakes”
  • Geometric flow (Partial Differential Equations)
  • Geodesic active contours

VIBOT9 - Universitat de Girona Medical Image - 3D Breast Ultrasound Segmentation May 13th, 2015

Ref: http://scialert.net/fulltext/?doi=ajaps.2011.101.111&org=12

slide-5
SLIDE 5

Level Set

  • Moving fronts: curve propagation
  • Idea: represent the evolving contour using a signed

function whose zero corresponds to the actual contour.

VIBOT9 - Universitat de Girona Medical Image - 3D Breast Ultrasound Segmentation May 13th, 2015

Original front Level set function

Ref: https://leticiateran.files.wordpress.com/2013/04/63a28-2.png

slide-6
SLIDE 6

Level Set

  • The zero level is spread, reflecting the propagation of the

contour

  • For one direction: fast marching

VIBOT9 - Universitat de Girona Medical Image - 3D Breast Ultrasound Segmentation May 13th, 2015

Evolution of the zero level contour (red curve) Evolution of the level set function (the red curve is the zero level contour) Ref: http://www.imagecomputing.org/~cmli/code/

slide-7
SLIDE 7

Algorithm Implemented

  • ITK: Pipeline of filters

VIBOT9 - Universitat de Girona Medical Image - 3D Breast Ultrasound Segmentation May 13th, 2015

Anisotropic Diffusion Filter Gradient Magnitude Filter Sigmoid Filter Fast Marching Level Set Filter Geodesic Active Contours Filter Binary Threshold Filter

Output image Input image Seed point Preprocessing

slide-8
SLIDE 8

Smoothing

  • itk::CurvatureAnisotropicDiffusionImageFilter
  • Noise reduction
  • Parameters: time step, number of iterations, conductance

parameter

VIBOT9 - Universitat de Girona Medical Image - 3D Breast Ultrasound Segmentation May 13th, 2015

Axial – Sagittal – Coronal Smoothing results for pacient 3

slide-9
SLIDE 9

Gradient Magnitude

  • itk::GradientMagnitudeRecursiveGaussianImageFilter
  • Boundaries enhancement
  • Parameters: standard deviation 𝜏

VIBOT9 - Universitat de Girona Medical Image - 3D Breast Ultrasound Segmentation May 13th, 2015

Axial – Sagittal – Coronal Gradient Magnitude results for pacient 3

slide-10
SLIDE 10

Sigmoid

  • itk::SigmoidImageFilter
  • Contrast enhancement
  • Parameters: 𝛽 and 𝛾

VIBOT9 - Universitat de Girona Medical Image - 3D Breast Ultrasound Segmentation May 13th, 2015

Axial – Sagittal – Coronal Sigmoid results for pacient 3

slide-11
SLIDE 11

Fast Marching

  • itk::FastMarchingImageFilter
  • Estimate an initial rough contour
  • Parameters: seed point, speed constant and initial

distance

VIBOT9 - Universitat de Girona Medical Image - 3D Breast Ultrasound Segmentation May 13th, 2015

Δx

www.na-mic.org/Wiki/.../Insight- Segmentation.ppt

Axial – Sagittal – Coronal Fast marching results for pacient 3

slide-12
SLIDE 12

Geodesic Active Contour

  • itk::GeodesicActiveContourLevelSetImageFilter
  • Refine initial approximation of contour
  • Parameters: propagation scaling (inflation), curvature

(smoothing) and the advection

VIBOT9 - Universitat de Girona Medical Image - 3D Breast Ultrasound Segmentation May 13th, 2015

Axial – Sagittal – Coronal Geodesic active contours results for pacient 3

slide-13
SLIDE 13

Binary Threshold

  • Itk::BinaryThresholdImageFilter
  • Produce a binary mask
  • Parameters: lower and upper thresholds, outside and

inside values

VIBOT9 - Universitat de Girona Medical Image - 3D Breast Ultrasound Segmentation May 13th, 2015

Axial – Sagittal – Coronal Binary thresholds results for pacient 3

slide-14
SLIDE 14

Evaluation

  • itk::LabelOverlapMeasuresImageFilter
  • Dice coefficient
  • Jaccard index
  • Confusion matrix: specificity and sensitivity
  • Sensitivity Average: 0.667247
  • Specificity Average: 0.999846
  • Jaccard Index Average: 0.4131118
  • Dice Coefficient Average: 0.573118
  • Computation time: around 145s (from reading to saving file)

VIBOT9 - Universitat de Girona Medical Image - 3D Breast Ultrasound Segmentation May 13th, 2015

Measurement Image 1 Image 2 Image 3 Image 4 Image 5 Sensitivity 0.670722 0.838789 0.737402 0.215896 0.873426 Specificity 0.999932 0.999647 0.999894 0.999997 0.99976 Jaccard 0.520453 0.335186 0.545407 0.212369 0.452144 Dice 0.684603 0.502082 0.705842 0.350337 0.622726

slide-15
SLIDE 15

Results

Pacient 1 Pacient 2

slide-16
SLIDE 16

Results

Pacient 3 Pacient 4

slide-17
SLIDE 17

Results

Pacient 5

slide-18
SLIDE 18

Conclusion

  • Impressions from the implemented segmentation method
  • Advantages:
  • Accuracy/stability when contour length changes
  • Deal with corners
  • Topology changes are automatically handled
  • Disadvantages:
  • Really sensitive to noise
  • Highly sensitive to the placement of seed points.
  • Impressions of the project

VIBOT9 - Universitat de Girona Medical Image - 3D Breast Ultrasound Segmentation May 13th, 2015

slide-19
SLIDE 19

References

  • Geodesic Active Contours. Vicent Caselles, Ron Kimmel, and Guillermo
  • Sapiro. Int. J. Comput. Vision .1997.
  • Cerebral Artery Segmentation with Level Set Methods. H. Ho, P. Bier, G.

Sands, and P. Hunter. Proceedings of Image and Vision Computing New

  • Zealand. 2007
  • Fast geodesic active contours. Goldenberg, R. Dept. of Comput. Sci.,

Technion-Israel Inst. of Technol., Haifa, Israel. 2001.

  • Image Segmentation Using Active Contour Model and Level Set Method

Applied to Detect Oil Spills. M. Airouche, L. Bentabet and M. Zelmat. Proceedings of the World Congress on Engineering. 2009.

  • Methods based on PDEs and wavelets in medical image segmentation.

<https://leticiateran.wordpress.com/>. 2015.

  • ITK Image Segmentation <www.na-mic.org/Wiki/.../Insight-

Segmentation.ppt>

  • Moving interfaces and boundaries: level set methods and fast marching
  • methods. <https://math.berkeley.edu/~sethian/2006/level_set.html>