Difference between revisions of "Projects:DBP2:Harvard:Registration Documentation"

From NAMIC Wiki
Jump to: navigation, search
Line 1: Line 1:
= Registration =
+
= Same subject Registration =
 
Slicer 3 contains a number of different ways to register two images.  In this document we will examine the results of these modules on three different cases. The convention used will be (moving image) -> (fixed image) = (resulting moved image).  The cases are:
 
Slicer 3 contains a number of different ways to register two images.  In this document we will examine the results of these modules on three different cases. The convention used will be (moving image) -> (fixed image) = (resulting moved image).  The cases are:
 
* Registering a t2 weighted image to an arbitrary t1, then registering this transformed image back to the original t2.  Ideally this will produce a identical image.
 
* Registering a t2 weighted image to an arbitrary t1, then registering this transformed image back to the original t2.  Ideally this will produce a identical image.
Line 5: Line 5:
 
* Registering a realigned t2 weighted image to another t2.
 
* Registering a realigned t2 weighted image to another t2.
  
==Tasks completed==
+
==Slicer modules==
  
 
===Affine===
 
===Affine===
Line 15: Line 15:
  
 
*DTI-baseline -> 53-t2 = aff3
 
*DTI-baseline -> 53-t2 = aff3
 
+
<nowiki>[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]</nowiki>
 
Image shows the 53-t2 target image and the registered image side by side. DTI distortion remains but aligned well to ventricles in center.
 
Image shows the 53-t2 target image and the registered image side by side. DTI distortion remains but aligned well to ventricles in center.
  
 
*53-t2 -> DTI-baseline = aff4
 
*53-t2 -> DTI-baseline = aff4
 
+
<nowiki>[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]</nowiki>
 
The registered image does not stretch to the full extent of the DTI at the midline, though the match is quite close more laterally.
 
The registered image does not stretch to the full extent of the DTI at the midline, though the match is quite close more laterally.
  
Line 30: Line 30:
  
 
*DTI-baseline -> 53-t2 = lin3
 
*DTI-baseline -> 53-t2 = lin3
 
+
<nowiki>[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]</nowiki>
 
Middle is clearly matched, coronal slices look okay, sagittal shows DTI distortions remain.
 
Middle is clearly matched, coronal slices look okay, sagittal shows DTI distortions remain.
  
 
*53-t2 -> DTI-baseline = lin4
 
*53-t2 -> DTI-baseline = lin4
 
+
<nowiki>[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]</nowiki>
 
Brainstem and cerebellum matched well, normal DTI distortions still present.
 
Brainstem and cerebellum matched well, normal DTI distortions still present.
  
Line 45: Line 45:
  
 
*DTI-baseline -> 53-t2 = rig3
 
*DTI-baseline -> 53-t2 = rig3
 
+
<nowiki>[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]</nowiki>
 
Brainstem aligned, normal DTI distortions present.
 
Brainstem aligned, normal DTI distortions present.
  
 
*53-t2 -> DTI-baseline =rig4
 
*53-t2 -> DTI-baseline =rig4
 
+
<nowiki>[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]</nowiki>
 
Similar.
 
Similar.
  
 
===Deformable B-spline===
 
===Deformable B-spline===
[[File:53-t2 Bspline.png|600px|thumb|left|BSpline case 1: Serious distortions.|]]
+
[[File:53-t2 Bspline.png|600px|thumb|left|BSpline case 1: Frontal enlargement, rear upward shift, distortions throughout.]]
 
*53-t2 -> 31-t1 = Bspline1
 
*53-t2 -> 31-t1 = Bspline1
 
*Bspline1 -> 53-t2 = Bspline2
 
*Bspline1 -> 53-t2 = Bspline2
Line 61: Line 61:
  
 
*DTI-baseline -> 53-t2 = Bspline3
 
*DTI-baseline -> 53-t2 = Bspline3
[[File:Baseline-bsp-t2-composite.png|600px|thumb|left|BSpline case 2: Matches cortex to skull.|]]
+
[[File:Baseline-bsp-t2-composite.png|600px|thumb|left|BSpline case 2: Terrible distortion. Matches cortex to skull.]]
 
Fits the cortex on the baseline to the skull on the t2.  Retry with a skull
 
Fits the cortex on the baseline to the skull on the t2.  Retry with a skull
 
stripped t2.
 
stripped t2.
  
*DTI-baseline-masked -> 53-t2-masked = Bspline4
+
*53-t2 -> DTI-baseline = Bspline4
 +
<nowiki>[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]</nowiki>
  
 +
*DTI-baseline-masked -> 53-t2-masked = Bspline3
 +
<nowiki>[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]</nowiki>
 
Bad registration, numerous deformations, midline shift in frontal lobe, right
 
Bad registration, numerous deformations, midline shift in frontal lobe, right
 
side deformations, cerebellum doesn't fit any more.
 
side deformations, cerebellum doesn't fit any more.
  
 
===Diffeomorphic Demons===
 
===Diffeomorphic Demons===
 +
[[File:53-t2 demons.png|600px|thumb|left|Demons case 1: Many minor deformations, clipping of brain at top, roughly correct placement overall.]]
 
*53-t2 -> 31-t1 = Demons1
 
*53-t2 -> 31-t1 = Demons1
 
*Demons1 -> 52-t2 = Demons2
 
*Demons1 -> 52-t2 = Demons2
Line 77: Line 81:
 
expansions, contractions, shifts.  Roughly in the same place overall.
 
expansions, contractions, shifts.  Roughly in the same place overall.
  
===Register Volumes Module===
+
*DTI-baseline -> 53-t2 = Demons3
 +
[[File:Basline-demons-t2.png|600px|thumb|left|Demons case 2: Terrible distortions everywhere, unclear what it was trying to match.]]
 +
 
 +
*53-t2 ->
 +
 
 +
==Register Images Module==
 +
===Initial===
 +
[[File:53-t2 RV initial.png|600px|thumb|left|Initial case 1: Minor shift down and anterior]]
 +
[[File:53-t2 RV init histo.png|600px|thumb|left|Initial case 1: histograms of the two images.]]
 
*53-t2 -> 31-t1 = init1
 
*53-t2 -> 31-t1 = init1
 
*init1 -> 53-t2 = init2
 
*init1 -> 53-t2 = init2
====Initial====
 
 
Huge shift in count values, rigid translation down ~4 pixels, left ~1 pixel.
 
Huge shift in count values, rigid translation down ~4 pixels, left ~1 pixel.
  
====Rigid====
+
*DTI-baseline -> t2 = init3
Huge shift in grey values. Few other problems.
+
<nowiki>[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]</nowiki>
  
====Affine====
+
*t2 -> DTI-baseline = init4
Again, shift in grey values, mostly okay otherwise.  Slight rightward shift.
+
<nowiki>[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]</nowiki>
  
====B-Spline====
+
===Rigid===
 +
[[File:53-t2 RV rigid.png|600px|thumb|left|Rigid case 1: Shift in grey values.]]
 +
*53-t2 -> 31-t1 = rigid1
 +
*rigid1 -> 53-t2 = rigid2
 +
Shift in grey values. Few other problems.
 +
 
 +
*DTI-baseline -> t2 = rigid3
 +
<nowiki>[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]</nowiki>
 +
 
 +
*t2 -> DTI-baseline = rigid4
 +
<nowiki>[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]</nowiki>
 +
 
 +
===Affine===
 +
[[File:53-t2 RV affine.png|600px|thumb|left|Affine case 1: Slight posterior shift and grey value shift.]]
 +
*53-t2 -> 31-t1 = affine1
 +
*affine1 -> 53-t2 = affine2
 +
Again, shift in grey values, mostly okay otherwise.  Slight posterior shift.
 +
 
 +
*DTI-baseline -> t2 = affine3
 +
<nowiki>[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]</nowiki>
 +
 
 +
*t2 -> DTI-baseline = affine4
 +
<nowiki>[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]</nowiki>
 +
 
 +
===B-Spline===
 +
[[File:53-t2 RV bspline.png|600px|thumb|left|BSpline case 1: Minor distortions, grey value shifts.]]
 +
*53-t2 -> 31-t1 = Bspline1
 +
*Bspline1 -> 53-t2 = Bspline2
 
Again grey values shift.  Minor distortion in frontal lobe and cerebellum.
 
Again grey values shift.  Minor distortion in frontal lobe and cerebellum.
  
 
*DTI-baseline -> t2
 
*DTI-baseline -> t2
TAKES FOREVER. Inferior frontal, temporal, stretched downwards. Cerebellum is
+
<nowiki>[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]</nowiki>
 +
Slow process. Inferior frontal, temporal, stretched downwards. Cerebellum is
 
terrible. Very top of the brain is stretched. Pretty bad overall.
 
terrible. Very top of the brain is stretched. Pretty bad overall.
 +
 +
*t2 -> DTI-baseline
 +
<nowiki>[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]</nowiki>
 +
 +
 +
= Different subject registration =

Revision as of 20:39, 22 April 2009

Home < Projects:DBP2:Harvard:Registration Documentation

Same subject Registration

Slicer 3 contains a number of different ways to register two images. In this document we will examine the results of these modules on three different cases. The convention used will be (moving image) -> (fixed image) = (resulting moved image). The cases are:

  • Registering a t2 weighted image to an arbitrary t1, then registering this transformed image back to the original t2. Ideally this will produce a identical image.
  • Registering a DTI baseline image to the corresponding realigned t2 weighted image, and vice-versa. This is a realistic test of one of our (the PNL) everyday uses of registration.
  • Registering a realigned t2 weighted image to another t2.

Slicer modules

Affine

Affine case 1: Minor loss of detail.
  • 53-t2 -> 31-t1 = aff1
  • aff1 -> 53-t2 = aff2

The image shows these side by side. The are nearly the same except for the loss of fine detail.

  • DTI-baseline -> 53-t2 = aff3

[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]] Image shows the 53-t2 target image and the registered image side by side. DTI distortion remains but aligned well to ventricles in center.

  • 53-t2 -> DTI-baseline = aff4

[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]] The registered image does not stretch to the full extent of the DTI at the midline, though the match is quite close more laterally.

Linear

Linear case 1: Minor shift at front, loss of detail.
  • 53-t2 -> 31-t1 = lin1
  • lin1 -> 53-t2 = lin2

The image shows lin2 and 53-t2 side by side. The image shows there appears to be some change beyond loss of detail, possibly a L-R shift in the front.

  • DTI-baseline -> 53-t2 = lin3

[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]] Middle is clearly matched, coronal slices look okay, sagittal shows DTI distortions remain.

  • 53-t2 -> DTI-baseline = lin4

[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]] Brainstem and cerebellum matched well, normal DTI distortions still present.

Rigid

Rigid case 1: Significant loss of detail.
  • 53-t2 -> 31-t1 = rig1
  • rig1 -> 53-t2 = rig2

Significant blurring, but every feature is in it's proper place.

  • DTI-baseline -> 53-t2 = rig3

[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]] Brainstem aligned, normal DTI distortions present.

  • 53-t2 -> DTI-baseline =rig4

[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]] Similar.

Deformable B-spline

BSpline case 1: Frontal enlargement, rear upward shift, distortions throughout.
  • 53-t2 -> 31-t1 = Bspline1
  • Bspline1 -> 53-t2 = Bspline2

Superior frontal enlargement, odd upward shifting in rear of brain, distortions everywhere.

  • DTI-baseline -> 53-t2 = Bspline3
BSpline case 2: Terrible distortion. Matches cortex to skull.

Fits the cortex on the baseline to the skull on the t2. Retry with a skull stripped t2.

  • 53-t2 -> DTI-baseline = Bspline4

[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]

  • DTI-baseline-masked -> 53-t2-masked = Bspline3

[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]] Bad registration, numerous deformations, midline shift in frontal lobe, right side deformations, cerebellum doesn't fit any more.

Diffeomorphic Demons

Demons case 1: Many minor deformations, clipping of brain at top, roughly correct placement overall.
  • 53-t2 -> 31-t1 = Demons1
  • Demons1 -> 52-t2 = Demons2

Multiple deformations. Much larger ventricles. Small errors everywhere, expansions, contractions, shifts. Roughly in the same place overall.

  • DTI-baseline -> 53-t2 = Demons3
Demons case 2: Terrible distortions everywhere, unclear what it was trying to match.
  • 53-t2 ->

Register Images Module

Initial

Initial case 1: Minor shift down and anterior
Initial case 1: histograms of the two images.
  • 53-t2 -> 31-t1 = init1
  • init1 -> 53-t2 = init2

Huge shift in count values, rigid translation down ~4 pixels, left ~1 pixel.

  • DTI-baseline -> t2 = init3

[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]

  • t2 -> DTI-baseline = init4

[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]

Rigid

Rigid case 1: Shift in grey values.
  • 53-t2 -> 31-t1 = rigid1
  • rigid1 -> 53-t2 = rigid2

Shift in grey values. Few other problems.

  • DTI-baseline -> t2 = rigid3

[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]

  • t2 -> DTI-baseline = rigid4

[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]

Affine

Affine case 1: Slight posterior shift and grey value shift.
  • 53-t2 -> 31-t1 = affine1
  • affine1 -> 53-t2 = affine2

Again, shift in grey values, mostly okay otherwise. Slight posterior shift.

  • DTI-baseline -> t2 = affine3

[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]

  • t2 -> DTI-baseline = affine4

[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]

B-Spline

BSpline case 1: Minor distortions, grey value shifts.
  • 53-t2 -> 31-t1 = Bspline1
  • Bspline1 -> 53-t2 = Bspline2

Again grey values shift. Minor distortion in frontal lobe and cerebellum.

  • DTI-baseline -> t2

[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]] Slow process. Inferior frontal, temporal, stretched downwards. Cerebellum is terrible. Very top of the brain is stretched. Pretty bad overall.

  • t2 -> DTI-baseline

[[FILE:FILENAME|600px|thumb|left|TYPE case X: explanation]]


Different subject registration