2016 Winter Project Week/Projects/ImageRestoration - Revision history

Adalca at 14:09, 8 January 2016

2016-01-08T14:09:22Z

2016-01-04T18:43:52Z

2016-01-04T18:42:34Z

Project Description

2016-01-04T18:22:12Z

Project Description

2016-01-04T03:28:28Z

2015-12-24T13:59:43Z

Project Description

2015-12-18T22:38:10Z

2015-12-18T22:18:59Z

2015-12-18T21:46:01Z

2015-12-17T22:14:29Z

@@ Line 4: / Line 4: @@
 Image:WMH_T1.png|Clinical Stroke Image
 Image:gmm_interp.png|Running result: Original (linear-interp) data | our gmm-based interp | "true" high res data
 </gallery>
@@ Line 34: / Line 35: @@
+|
 <!-- Fill this out at the end of Project Week; describe what you did this week and what you plan to do next -->
+* We've built infrastructure for deploying the training at each location in parallele
 |}

@@ Line 3: / Line 3: @@
 Image:PW-MIT2016.png|[[2016_Winter_Project_Week#Projects|Projects List]]
 Image:WMH_T1.png|Clinical Stroke Image
-Image:gmm_interp.png|Current Result: Original (linear-interpolated) data | our gmm-based interpolation | "true" high resolution data
+Image:gmm_interp.png|Running result: Original (linear-interp) data | our gmm-based interp | "true" high res data
 </gallery>

← Older revision		Revision as of 18:42, 4 January 2016
Line 18:		Line 18:

	To improve results for large datasets of clinical-quality data, we are investigating restoration methods without training datasets. Here, we are using a patch-based Gaussian Mixture Model approach with MRF priors and utilizing only the current dataset, without an external training dataset.		To improve results for large datasets of clinical-quality data, we are investigating restoration methods without training datasets. Here, we are using a patch-based Gaussian Mixture Model approach with MRF priors and utilizing only the current dataset, without an external training dataset.
		+
		+	<strong>Essentially, we explore a model where for a given location in all volumes of a dataset, we model those image patches as drawn from a particular mixture model.</strong>

	{\| class="wikitable"		{\| class="wikitable"

@@ Line 29: / Line 29: @@
+|
 <!-- Add a bulleted list of key points -->
+* We've implemented a small version of the model without regularization and that only works on a small section of a volume. We'll try to deploy it on large-scale dataset on entire volumes.
+|
 <!-- Fill this out at the end of Project Week; describe what you did this week and what you plan to do next -->
+*
 |}

@@ Line 3: / Line 3: @@
 Image:PW-MIT2016.png|[[2016_Winter_Project_Week#Projects|Projects List]]
 Image:WMH_T1.png|Clinical Stroke Image
 </gallery>

@@ Line 19: / Line 19: @@
 {| class="wikitable"
 |- style="vertical-align:top;"
-! style="text-align: left; width:27%" |       Objective
-<!-- Add a bulleted list of key points -->
+<!-- Objective bullet points -->
 * We will investigate a current model for learning and using Patch-based Gaussian Mixture Model to restore sparse-slice data.
-! style="text-align: left; width:27%" |       Approach, Plan
 <!-- Add a bulleted list of key points -->
+*
-−
+|
 <!-- Fill this out at the end of Project Week; describe what you did this week and what you plan to do next -->
+*
 |}

← Older revision		Revision as of 22:38, 18 December 2015
Line 32:		Line 32:

	\|}		\|}
−
−	~~==Background and References==~~
−	~~<!-- Use this space for information that may help people better understand your project, like links to papers, source code, or data -->~~

@@ Line 18: / Line 18: @@
 To improve results for large datasets of clinical-quality data, we are investigating restoration methods without training datasets. Here, we are using a patch-based Gaussian Mixture Model approach with MRF priors and utilizing only the current dataset, without an external training dataset.
-<div style="width: 27%; float: left; padding-right: 3%;">
+! style="text-align: left; width:27%" |      Progress
-<h3>Objective</h3>
+<!-- Fill this out at the end of Project Week; describe what you did this week and what you plan to do next -->
-* We will investigate a current model for learning and using Patch-based Gaussian Mixture Model to restore sparse-slice data.
-</div>
-<div style="width: 27%; float: left; padding-right: 3%;">
+|}
-</div>
+==Background and References==
-<div style="width: 27%; float: left; padding-right: 3%;">
+<!-- Use this space for information that may help people better understand your project, like links to papers, source code, or data -->

@@ Line 6: / Line 6: @@
 ==Key Investigators==
-- Adrian Dalca, Katie Bouman, Polina Golland, MIT
+*Adrian Dalca (MIT)
 ==Project Description==
@@ Line 19: / Line 21: @@
 <div style="width: 27%; float: left; padding-right: 3%;">
 <h3>Objective</h3>
-* We will investigate a current model for learning and using Patch-based Gaussian Mixture Model to restore sparse-slice data..
+* We will investigate a current model for learning and using Patch-based Gaussian Mixture Model to restore sparse-slice data.
 </div>
 <div style="width: 27%; float: left; padding-right: 3%;">

← Older revision	Revision as of 22:14, 17 December 2015
(No difference)