DOC PREVIEW
NMT EE 552 - Adipose Quantification of CT Scans using Image Processing in Matlab

This preview shows page 1-2-24-25 out of 25 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 25 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 25 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 25 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 25 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 25 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

Slide 1Slide 2Slide 3Slide 4Slide 5Slide 6Slide 7Slide 8Slide 9Slide 10Slide 11Slide 12Slide 13Slide 14Slide 15Slide 16Slide 17Slide 18Slide 19Slide 20Slide 21Slide 22Slide 23Slide 24Slide 25Adipose Quantification of CT Scans using Image Processing in MatlabTed Schuler-SandyBrianna Klein30th April, 2009Sponsor: Dr. Michaelann TartisAdviser: Dr. Hector ErivesPurposeThe project is to use image processing techniques to accurately quantify the amount of fat in a mouse CT scan.Background●Why is this being done?●To test the effectiveness of treatments for diabetes and obesity in laboratory mice. ●Overview of computed tomographyViews of MiceCoronalSagittalTransverseBackground - Previous WorkSpecifications●Automated●Accurately identify fat in an image with minimum input from the user●Calculate number of fat pixels on mouse imageScope of Project●Adult mice only●Skinny, Medium, and Fat mice●One view - coronal●One image at a time●Focusing on three specific fat padsWhere is the fat?Fat Mouse ExampleWhere is the fat?Skinny Mouse ExampleChallenges/Goals●Maximum automation and ease for user●Removal of non-fat highlights●Skin●Gas Pockets●Lungs●Noise reduction not necessary●Initial images are cleanDesign SolutionsOverview●Obtain usable images*●Intensity range identification*●False identification reduction/removal●Skin●Gas Pockets*●Fat percent calculation●Intensity Highlighting*Requires user input.Design SolutionsImage Acquisition●AMIDE●Select image slices (coronal, sagittal, transverse)●Automatic Histogram Processing●Convert to JPEGNo Histogram ProcessingIntensity HighlightingDesign SolutionsIntensity Range Identification●The fat lies within a specific intensity range, which can vary from image to image. ●Dynamic Intensity Range Selection ●User Input ●Maximum and minimum intensity selection●Problem: Other non-fat parts can lie within this intensity range, causing false identification.False Identification ExampleUnhighlighted ImageFat HighlightingDesign SolutionsFalse Identification Reduction/Removal●Skin Highlighting Removal●Zero-padding●Scan rows and columns to remove a threshold of pixels from selected intensities●Alternate solutions●Gas Pocket Removal●Manual IdentificationSkin Removal ExampleSkin HighlightingSkin Highlighting RemovedSkin RemovalAlternativesBoundary Trace using the Image Processing ToolboxRemoval by UserThe lungs and gas pockets can easily be removed from the identified pixels. Gas pockets HighlightedHighlighting RemovedDesign SolutionsFat Percent Calculation●Method similar to skin removal counts total pixels of mouseFat percent=NfatNtotal∗100Non-Ideal PicturesTwisted Coronal ImageHighlighted FatOther ViewsTransverseTransverse images highlight the bed where the mouse is sitting.Transverse ImageHighlighted FatOther ViewsSagittalSagittal Mouse ImageHighlighted FatResults●Extra conversion step before processing●High sensitivity to intensity selection●Requires user input●Automated calculationConclusion●Specifications met●Automated highlighting●Fat pixel count calculation●Further Work:●Removal of test bed in Sagittal and Transverse●Three dimensional image processingAdipose Quantification of CT Scans using Image Processing in MatlabTed Schuler-SandyBrianna Klein30th April, 2009Sponsor: Dr. Michaelann TartisAdviser: Dr. Hector


View Full Document

NMT EE 552 - Adipose Quantification of CT Scans using Image Processing in Matlab

Download Adipose Quantification of CT Scans using Image Processing in Matlab
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Adipose Quantification of CT Scans using Image Processing in Matlab and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Adipose Quantification of CT Scans using Image Processing in Matlab 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?