Medical Images Texture Analysis Using Waveles.ppt

Medical Images Texture Analysis Using Waveles Why Texture Analysis? Method for differentiation between normal and abnormal tissue. Contrast between malignant tissue and normal tissue may be present but below the threshold of human perception. Avoid biopsy intervention to identify the tissue. Diagnosis (early!!!!) and tracking of diseases. (Diffuse diseases). How to do it? ….image processing Techniques to extract image features and enhances these features with respect to their surroundings…. ……Statistical, Fuzzy Logic, Fourier, Wavelets… Why Use Wavelets? ……. ……Wavelet-based processing algorithms were superior….Ability of wavelets to discriminate different frequencies and to preserve signal details at different resolutions. …‘zoom in’ and ‘zoom out’ capability of the wavelet filters which can translate themeselves to a location of a signal that is of interest and dilate themselves properly to preserve the resolution of that portion of the signal.[3] 2D Wavelets S=aφ+dψ Applications Detection of Microcalcifications in Digital Mammograms. Characterization of Liver tissue. MR Image texture analysis. Detection of Microcalcifications in Digital Mammograms (1) Detection of Microcalcifications in Digital Mammograms (2) Detection of Microcalcifications in Digital Mammograms(3) Using BAUBECHIES Wavelet Filters (try Matlab db4) Decomposition level 2 MC correspond to highest frequencies Set the background signal to zero (lowest frequencies) Reconstruct the image Detection of MC in …Some lifting (4) Visibility is improved by using nonlinear thresholding (nonlinear mapping based on arc-tangent). Try Matlab!!! Detection of MC Organigram (5) Detection of MC Organigram (6) Detection of MC in … Discussion(7) The resolution of the original digital mammogram Intensity contrast (S/N??) Artifacts!!! Wavelet Selection Effect of threshold. Characterization of Liver tissue(1) Using nonseparable Wavelet. Quincunx transform: nonseparable and nonoriented filters. Filter based on McClel