Difference between revisions of "2016 Winter Project Week/Projects/ImageRestoration"
From NAMIC Wiki
| Line 29: | Line 29: | ||
| | | | ||
<!-- Add a bulleted list of key points --> | <!-- 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 --> | <!-- Fill this out at the end of Project Week; describe what you did this week and what you plan to do next --> | ||
* | * | ||
|} | |} | ||
Revision as of 18:22, 4 January 2016
Home < 2016 Winter Project Week < Projects < ImageRestoration
Clinical Stroke Image
"true" high resolution data
Key Investigators
- Adrian Dalca (MIT)
- Katie Bouman (MIT)
- Polina Golland (MIT)
Project Description
Most synthesis, in-painting or super-resolution methods require a training dataset which includes the desired-quality images. Unfortunately, in the clinical setting this is often now available.
Due to the low quality of clinical images (often with many artifacts, 7mm thick slices, etc), most standard algorithms, such as those for registration, segmentation, analysis, will fail.
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.
| Objective | Approach and Plan | Progress and Next Steps |
|---|---|---|
|
|
|
