Projects:NonparametricSegmentation - Revision history

Wachinge: /* Conclusion */

2012-11-28T20:19:22Z

Conclusion

Wachinge: /* Key Investigators */

2012-11-28T20:16:43Z

Key Investigators

Wachinge: /* Experiments */

2012-11-28T20:12:11Z

Experiments

Wachinge: /* Experiments */

2012-11-28T20:11:42Z

Experiments

Wachinge: /* Experiments */

2012-11-28T20:11:05Z

Experiments

Wachinge: /* Experiments */

2012-11-28T20:10:50Z

Experiments

Wachinge: /* Experiments */

2012-11-28T20:10:17Z

Experiments

Wachinge: /* Experiments */

2012-11-28T20:09:08Z

Experiments

Wachinge: /* Experiments */

2012-11-28T20:04:26Z

Experiments

Wachinge: /* Spectral Label Fusion */

2012-11-28T19:57:16Z

Spectral Label Fusion

@@ Line 60: / Line 60: @@
 == Conclusion ==
 In this work, we investigate a generative model that leads to label fusion style image segmentation methods. Within the proposed framework, we derived various algorithms that combine transferred training labels into a single segmentation estimate. With a dataset of 39 brain MRI scans and corresponding label maps obtained from an expert, we experimentally compared these segmentation algorithms with Freesurfer’s widely-used atlas-based segmentation tool [4]. Our results suggested that the proposed framework yields accurate and robust segmentation tools that can be employed on large multi-subject datasets. In a second experiment, we employed one of the developed segmentation algorithms to compute hippocampal volumes in MRI scans of 304 subjects. A comparison of these measurements across clinical and age groups indicate that the proposed algorithms are sufficiently sensitive to detect hippocampal atrophy that precedes probable onset of Alzheimer’s Disease.
 == Literature ==

@@ Line 77: / Line 77: @@
 = Key Investigators =
-*MIT: Mert R. Sabuncu, B.T. Thomas Yeo, Koen Van Leemput, Michal Depa and Polina Golland
+*MIT: Mert R. Sabuncu, B.T. Thomas Yeo, Koen Van Leemput, Michal Depa, Christian Wachinger and Polina Golland
 *Harvard: Koen Van Leemput and Bruce Fischl
 = Publications =
 [http://www.na-mic.org/publications/pages/display?search=Projects%3ANonparametricSegmentation&submit=Search&words=all&title=checked&keywords=checked&authors=checked&abstract=checked&sponsors=checked&searchbytag=checked| NA-MIC Publications Database on Nonparametric Models for Supervised Image Segmentation]

@@ Line 54: / Line 54: @@
 In a third experiment, we evaluated the spectral label fusion. We automatically segment the left atrium of the heart in a set of 16 electro-cardiogram gated (0.2 mmol/kg) Gadolinium- DTPA contrast-enhanced cardiac MRA images (CIDA sequence, TR=4.3ms, TE=2.0ms, θ = 40º, in-plane resolution varying from 0.51mm to 0.68mm, slice thickness varying from 1.2mm to 1.7mm, 512 × 512 × 96, -80 kHz bandwidth, atrial diastolic ECG timing to counteract considerable volume changes of the left atrium). The left atrium was manually segmented in each image by an expert. We set the UCM threshold (see scheme) to ρ = 0.2 for the 2D and ρ = 0 for the 3D watershed. We perform leave-one-out experiments by treating one subject as the test image and the remaining 15 subjects as the training set. We use the Dice score and the modified (average) Hausdorff distance between the automatic and expert segmentations as quantitative measures of segmentation quality. We compare our method to majority voting (MV) and intensity-weighted label fusion (IW).
-In the figures below, we present dice volume overlap and modified Hausdorff distance for each algorithm. The improvements in segmentation accuracy between the proposed method and IW are statistically significant (p<10−5).
+In the figures below, we present dice volume overlap (left) and modified Hausdorff distance (right) for each algorithm. The improvements in segmentation accuracy between the proposed method and IW are statistically significant (p<10−5).
 [[File:boxplotDice.png|400px]] [[File:boxplotHausdorff.png|400px]]

@@ Line 54: / Line 54: @@
 In a third experiment, we evaluated the spectral label fusion. We automatically segment the left atrium of the heart in a set of 16 electro-cardiogram gated (0.2 mmol/kg) Gadolinium- DTPA contrast-enhanced cardiac MRA images (CIDA sequence, TR=4.3ms, TE=2.0ms, θ = 40º, in-plane resolution varying from 0.51mm to 0.68mm, slice thickness varying from 1.2mm to 1.7mm, 512 × 512 × 96, -80 kHz bandwidth, atrial diastolic ECG timing to counteract considerable volume changes of the left atrium). The left atrium was manually segmented in each image by an expert. We set the UCM threshold (see scheme) to ρ = 0.2 for the 2D and ρ = 0 for the 3D watershed. We perform leave-one-out experiments by treating one subject as the test image and the remaining 15 subjects as the training set. We use the Dice score and the modified (average) Hausdorff distance between the automatic and expert segmentations as quantitative measures of segmentation quality. We compare our method to majority voting (MV) and intensity-weighted label fusion (IW).
-Fig. 3 presents dice volume overlap and modified Hausdorff distance for each algorithm. The improvements in segmentation accuracy between the proposed method and IW are statistically significant (p<10−5).
+In the figures below, we present dice volume overlap and modified Hausdorff distance for each algorithm. The improvements in segmentation accuracy between the proposed method and IW are statistically significant (p<10−5).
 [[File:boxplotDice.png|400px]] [[File:boxplotHausdorff.png|400px]]

@@ Line 56: / Line 56: @@
 Fig. 3 presents dice volume overlap and modified Hausdorff distance for each algorithm. The improvements in segmentation accuracy between the proposed method and IW are statistically significant (p<10−5).
-[[File:boxplotDice.png|200px]] [[File:boxplotHausdorff.png|200px]]
+[[File:boxplotDice.png|400px]] [[File:boxplotHausdorff.png|400px]]
 == Conclusion ==

@@ Line 53: / Line 53: @@
 [[File:HippocampalVolume.png]]
-In a third experiment, we evaluated the spectral label fusion. We automatically segment the left atrium of the heart in a set of 16 electro-cardiogram gated (0.2 mmol/kg) Gadolinium- DTPA contrast-enhanced cardiac MRA images (CIDA sequence, TR=4.3ms, TE=2.0ms, θ = 40º, in-plane resolution varying from 0.51mm to 0.68mm, slice thickness varying from 1.2mm to 1.7mm, 512 × 512 × 96, -80 kHz bandwidth, atrial diastolic ECG timing to counteract considerable volume changes of the left atrium). The left atrium was manually segmented in each image by an expert. For all the experiments we set γ = 2.5, giving higher weight to the spectral component. We set ρ = 0.2 for the 2D and ρ = 0 for the 3D watershed after inspecting the UCM. We perform leave-one-out experiments by treating one subject as the test image and the remaining 15 subjects as the training set. We use the Dice score and the modified (average) Hausdorff distance between the automatic and expert segmentations as quantitative measures of segmentation quality. We compare our method to majority voting (MV) and intensity-weighted label fusion (IW) [11].
+In a third experiment, we evaluated the spectral label fusion. We automatically segment the left atrium of the heart in a set of 16 electro-cardiogram gated (0.2 mmol/kg) Gadolinium- DTPA contrast-enhanced cardiac MRA images (CIDA sequence, TR=4.3ms, TE=2.0ms, θ = 40º, in-plane resolution varying from 0.51mm to 0.68mm, slice thickness varying from 1.2mm to 1.7mm, 512 × 512 × 96, -80 kHz bandwidth, atrial diastolic ECG timing to counteract considerable volume changes of the left atrium). The left atrium was manually segmented in each image by an expert. We set the UCM threshold (see scheme) to ρ = 0.2 for the 2D and ρ = 0 for the 3D watershed. We perform leave-one-out experiments by treating one subject as the test image and the remaining 15 subjects as the training set. We use the Dice score and the modified (average) Hausdorff distance between the automatic and expert segmentations as quantitative measures of segmentation quality. We compare our method to majority voting (MV) and intensity-weighted label fusion (IW).
 == Conclusion ==

← Older revision		Revision as of 20:04, 28 November 2012
Line 53:		Line 53:
	[[File:HippocampalVolume.png]]		[[File:HippocampalVolume.png]]

		+	In a third experiment, we evaluated the spectral label fusion. We automatically segment the left atrium of the heart in a set of 16 electro-cardiogram gated (0.2 mmol/kg) Gadolinium- DTPA contrast-enhanced cardiac MRA images (CIDA sequence, TR=4.3ms, TE=2.0ms, θ = 40º, in-plane resolution varying from 0.51mm to 0.68mm, slice thickness varying from 1.2mm to 1.7mm, 512 × 512 × 96, -80 kHz bandwidth, atrial diastolic ECG timing to counteract considerable volume changes of the left atrium). The left atrium was manually segmented in each image by an expert. For all the experiments we set γ = 2.5, giving higher weight to the spectral component. We set ρ = 0.2 for the 2D and ρ = 0 for the 3D watershed after inspecting the UCM. We perform leave-one-out experiments by treating one subject as the test image and the remaining 15 subjects as the training set. We use the Dice score and the modified (average) Hausdorff distance between the automatic and expert segmentations as quantitative measures of segmentation quality. We compare our method to majority voting (MV) and intensity-weighted label fusion (IW) [11].

	== Conclusion ==		== Conclusion ==

@@ Line 33: / Line 33: @@
 '''(1):'''  The first step extracts the boundaries from the image and label map, joins them in the spectral framework, and produces weighted contours. The boundary extraction we employ concepts based on spectral clustering, as presented in~[5].
-'''(2):'''In the second step, the extracted contours give rise to regions, partitioning the image. We obtain the parcellation of the image with the watershed algorithm.
+'''(2):''' In the second step, the extracted contours give rise to regions, partitioning the image. We obtain the parcellation of the image with the watershed algorithm.
-In the third step, we assign a label to each region based on the input label map, producing the final segmentation. For the region-based voting, we replace the assumption of independent voxel samples from the probabilistic framework with the Markov property. Crucial is the selection of image-specific neighborhoods that capture the relevant information, as done in the second step.
+'''(3):''' In the third step, we assign a label to each region based on the input label map, producing the final segmentation. For the region-based voting, we replace the assumption of independent voxel samples from the probabilistic framework with the Markov property. Crucial is the selection of image-specific neighborhoods that capture the relevant information, as done in the second step.
 [[File:scheme5.png|800px]]