DBP3:Utah:AutoEndoSeg - Revision history

Cates: /* Larger Atlas Images */

2011-11-30T20:19:14Z

Larger Atlas Images

Cates at 20:18, 30 November 2011

2011-11-30T20:18:47Z

Ggardner at 18:28, 29 November 2011

2011-11-29T18:28:06Z

Ggardner at 08:49, 29 November 2011

2011-11-29T08:49:22Z

Ggardner at 08:30, 29 November 2011

2011-11-29T08:30:31Z

Ggardner at 08:30, 29 November 2011

2011-11-29T08:30:08Z

Cates: /* Notes on Yi's Slicer Module */

2011-11-08T20:51:28Z

Notes on Yi's Slicer Module

Cates: Created page with '= Notes on the Endocardial Segmentation Algorithm = Algorithm is a multi-atlas segmentation. # Register (cropped) heart region of patient DEMRI to all of the n atlas DEMRI images…'

2011-11-08T20:49:26Z

Created page with '= Notes on the Endocardial Segmentation Algorithm = Algorithm is a multi-atlas segmentation. # Register (cropped) heart region of patient DEMRI to all of the n atlas DEMRI images…'

New page

= Notes on the Endocardial Segmentation Algorithm =
Algorithm is a multi-atlas segmentation.
# Register (cropped) heart region of patient DEMRI to all of the n atlas DEMRI images and record n minimum metric values (mutual information)
# Transform all n atlas segmentations to patient DEMRI space
# Segmentation is a weighted average of the transformed atlas segmentation, where weights are determined by the mutual information values from step 1
# Pick a threshold as your segmentation

= Notes on Yi's Slicer Module =
* The exposed parameter is the ratio of pixel samples for mutual information to the total number of pixels in the image.
* We could also expose the threshold parameter in the final step
* We will partner with Yi for module development at the Jan 2012 project week

@@ Line 40: / Line 40: @@
 |}
-= Larger Atlas Images =
+= Larger Atlas Images 11/30/11=
 * A second atlas was generated from cropped images that included a larger region of the initial MRI scans (namely, the appendage and a greater extent of the PVs was included in this second atlas).
 * This atlas showed no marked improvement over the initial atlas

← Older revision		Revision as of 20:18, 30 November 2011
Line 1:		Line 1:
		+	= Current Activities =
		+	* 11/30/11 Greg and Josh met with Yi via Skype
		+	** Algorithm is still very sensitive to image cropping -- registration seems to fail unless cropping is exact
		+	** We can input an atlas image and get a correct result
		+	** Yi will investigate multiscale registration to see if that improves result
		+	** Yi will look for any bugs in code that might be affecting registration
		+	** Greg will send Yi current data for new atlas with larger crop region
	= Notes on the Endocardial Segmentation Algorithm =		= Notes on the Endocardial Segmentation Algorithm =
	Algorithm is a multi-atlas segmentation.		Algorithm is a multi-atlas segmentation.

@@ Line 60: / Line 60: @@
 | [[File:CAMRA_N26_2_LGE.png|thumb|center]]
 | [[File:CARMA_N26_2_Endo.png|thumb|center]]
 |-
+|
@@ Line 66: / Line 70: @@
 |-
 |}

← Older revision		Revision as of 08:49, 29 November 2011
Line 33:		Line 33:
	\|}		\|}

		+	= Larger Atlas Images =
		+	* A second atlas was generated from cropped images that included a larger region of the initial MRI scans (namely, the appendage and a greater extent of the PVs was included in this second atlas).
		+	* This atlas showed no marked improvement over the initial atlas
		+	* Segmentations with the second atlas tended to not capture the entire blood pool, whereas, the first atlas tended to overestimate the blood pool.
		+	* Registration issues?

−	= ~~Larger Atlas Images =~~	+	{\| class="wikitable"
		+	\|-
		+	!
		+	! Mask and LGE Image
		+	! Mask and Expert-defined Segmentation
		+	\|-
		+	\| Example 1
		+	\| [[File:CARMA_N24_LGE.png\|thumb\|center]]
		+	\| [[File:CARMA_N24_Endo.png\|thumb\|center]]
		+	\|-
		+	\| Example 2
		+	\| [[File:CARMA_N25_LGE.png\|thumb\|center]]
		+	\| [[File:CARMA_N25_Endo.png\|thumb\|center]]
		+	\|-
		+	\| Example 3
		+	\| [[File:CAMRA_N26_LGE.png\|thumb\|center]]
		+	\| [[File:CARMA_N26_Endo.png\|thumb\|center]]
		+	\|-
		+	\| Example 4
		+	\| [[File:CAMRA_N26_2_LGE.png\|thumb\|center]]
		+	\| [[File:CARMA_N26_2_Endo.png\|thumb\|center]]
		+	\|-
		+	\|
		+	\| Cropped MRI image overlaid with the algorithm-defined blood pool.
		+	\| Expert manual segmentations (white) overlaid with the algorithm-defined blood pool (green).
		+	\|-
		+	\|}

@@ Line 14: / Line 14: @@
 = Initial Experiences =
-* Comparison of Yi and Utah-derived segmentations
+* Comparison of GA Tech and Utah-derived segmentations
 (click to enlarge)

← Older revision		Revision as of 08:30, 29 November 2011
Line 12:		Line 12:
	* Uses the ITK mutual information code. Does affine registration followed by b-spline		* Uses the ITK mutual information code. Does affine registration followed by b-spline
	* We will partner with Yi for module development at the Jan 2012 project week		* We will partner with Yi for module development at the Jan 2012 project week
		+
		+	= Initial Experiences =
		+	* Comparison of Yi and Utah-derived segmentations
		+	(click to enlarge)
		+
		+	{\| class="wikitable"
		+	\|-
		+	! GA Tech Segmentation
		+	! Utah Segmentation
		+	! Thresholding Difference
		+	\|-
		+	\| [[File:CARMA_GAT_Endo.png\|thumb\|center]]
		+	\| [[File:CARMA_UT_Endo.png\|thumb\|center]]
		+	\| [[File:CARMA_Diff_Endo.png\|thumb\|center]]
		+	\|-
		+	\| Segmentation of an atlas image by GA Tech, using their endocardial auto-segmentation algorithm.
		+	\| Segmentation of the same image with the same algorithm by Utah.
		+	\| User-based variability in thresholding the probability mask output. (Green region - Utah; white region - GA Tech).
		+	\|-
		+	\|}
		+
		+
		+	= Larger Atlas Images =

@@ Line 9: / Line 9: @@
 * The exposed parameter is the ratio of pixel samples for mutual information to the total number of pixels in the image.
 * We could also expose the threshold parameter in the final step
 * We will partner with Yi for module development at the Jan 2012 project week