1 / 3

P ROPOSED M ETHOD

A N O PTIMIZATION M ETHOD F OR S LICE I NTERPOLATION O F M EDICAL I MAGES Ahmadreza Baghaie , Prof. Zeyun Yu EE Dept., CS Dept., UW-Milwaukee.

treva
Télécharger la présentation

P ROPOSED M ETHOD

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. AN OPTIMIZATION METHOD FOR SLICE INTERPOLATION OF MEDICAL IMAGESAhmadrezaBaghaie, Prof. Zeyun YuEE Dept., CS Dept., UW-Milwaukee • Slice interpolation is a well-known research topic in bio-medical applications. With modern imaging modalities (CT, MRI, light/electron microscopy etc.), a sequence of 2D images can be provided and used in building 3D models. However, the resolutions of the images are often not identical in all three dimensions. This asymmetry in the resolution causes problems such as step shaped iso-surfaces and discontinuity in structures in 3D reconstructed models. Therefore utilizing a slice interpolation algorithm to augment the 3D data into a symmetric one is of high demand. In general, slice interpolation methods can be divided into two groups: intensity-based interpolation and object based interpolation. • In the first category, the final result of interpolation is directly computed from the intensity values of input images. Despite their advantages due to their low computational complexity, these methods suffer from blurring effects on object boundaries. • In object-based methods, on the other hand, the extracted information from objects contained in input images is used in order to guide the interpolation into more accurate results. An increasingly important group of approaches for object-based image interpolation is based on image registration.

  2. PROPOSED METHOD • Given two images (reference and template), image registration is to find a spatial transformation such that the transformed template matches the reference, subject to a suitable distance measure. • Rigid methods • Affine Methods • Non-rigid methods • Similarity measures: Sum of Squared Differences (SSD), Mutual Information (MI), Normalized Mutual Information (NMI), Correlation Coefficient (CC), Normalized Correlation Coefficient (NCC), Normalized Gradient Fields (NGF) etc. • Regularization: Elastic registration, Fluid registration, Diffusion registration, Curvature registration etc.

  3. RESULTS Fig. 2. Row #1: Three consecutive images. The first and third images are used for interpolation. Row #2: results of interpolation by using the linear, non-modified and proposed methods respectively. Row#3: difference images of the results and original image. Fig. 1. Row #1: Three consecutive images. The first and third images are used for interpolation. Row #2: computed displacement fields after optimization. Row #3: interpolation results for linear and proposed method. Row #4: difference images of the results and reference image.

More Related