Advances in the Use of Neurophysiologycally-based
Download
1 / 23

M. Aguilar, J. R. New and E. Hasanbelliu Knowledge Systems Laboratory MCIS Department - PowerPoint PPT Presentation


  • 70 Views
  • Uploaded on

Advances in the Use of Neurophysiologycally-based Fusion for Visualization and Pattern Recognition of Medical Imagery. M. Aguilar, J. R. New and E. Hasanbelliu Knowledge Systems Laboratory MCIS Department Jacksonville State University Jacksonville, AL 36265. Outline. Introduce Med-LIFE.

loader
I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.
capcha
Download Presentation

PowerPoint Slideshow about ' M. Aguilar, J. R. New and E. Hasanbelliu Knowledge Systems Laboratory MCIS Department' - apu


An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript

Advances in the Use of Neurophysiologycally-based Fusion for Visualization and Pattern Recognition of Medical Imagery

M. Aguilar, J. R. New and E. Hasanbelliu

Knowledge Systems Laboratory

MCIS Department

Jacksonville State University

Jacksonville, AL 36265


Outline
Outline

  • Introduce Med-LIFE.

  • Revisit 3D image fusion architecture.

  • Compare 2D and 3D fusion results.

  • Fusion for segmentation and pattern recognition.

  • Contextual zoom tool.

  • Segmentation results.



3d shunt equation
3D Shunt Equation

Shunting Neural Network Equation:

Grossberg (1968),

Elias & Grossberg (1972)

Where:

A – decay rate

B – maximum activation level (set to 1)

D – minimum activation level (set to 1)

IC – excitatory input

IS – lateral inhibitory input

C, Gc and Gs are as follows:

3D Shunt Operator Symbol




2d vs 3d fusion results

2D Fusion

3D Fusion

2D vs. 3D Fusion Results

MRI-T1

MRI-T2

SPECT

MRI-PD


4 band hybrid fusion architecture

T2

Images

T1

Images

Color

Remap

.

.

SPECT

Images

.

.

.

.

Color Fuse Result

PD

Image

+

_

4-Band Hybrid Fusion Architecture

Q

I

Y

Noise cleaning &

registration if needed

Contrast

Enhancement

Between-band Fusion

and Decorrelation


Hybrid fusion results
Hybrid Fusion Results

2D Fusion

3D Fusion



Contextual zoom visualization
Contextual Zoom Visualization

  • Zoom in place supports:

  • focused attention

  • improved screen real-estate usage

  • Zoom in place:

  • occludes information

  • reduces efficiency by forcing user to maintain context



Contextual zoom visualization2
Contextual Zoom Visualization

  • Developed based on COTS software developed by Idelix

  • Supports visualization of fused imagery at multiple details levels

  • Supports detailed analysis and selection for user-driven pattern learning…


User driven pattern learning
User-Driven Pattern Learning

  • Supervised learning where training data is selected by user/expert (Waxman et al).

  • Results assessed and corrected by user.

  • Fuzzy ARTMAP neural network for fast and stable learning.

  • Address order sensitivity by introducing N voters trained with alternate ordering of the training data.



Heterogeneous voting
Heterogeneous Voting

  • Train 3 Fuzzy ARTMAP systems with parameters as before (different data orderings)

  • Train remaining 2 systems with all parameters as in the 3rd system except for Vigilance (which is a threshold measure that controls the sensitivity of the system).


Homogeneous vs heterogeneous voters
Homogeneous vs. Heterogeneous Voters

5 Homogeneous Voters

5 Heterogeneous Voters


2d vs 3d fusion segmentation results

2D Fusion-based Segmentation

3D Fusion-based Segmentation

2D vs. 3D Fusion Segmentation Results


Generalization
Generalization

Training

Results

Testing

Results

Slice 10

Slice 11


Conclusions
Conclusions

  • Modified fusion approach combines benefits of 2D and 3D fusion.

  • Preliminary learning segmentation results indicate robustness across slices and cases.

  • Demonstrated superior performance of 3D fusion for both visualization and pattern recognition.

  • Heterogeneous voting scheme improves learning performance.



2d vs 3d generalization
2D vs. 3D Generalization

2D Fusion

3D Fusion

Slice 10

Testing

Results



ad