Loading...

MRI Semi-Supervised Segmentation

Izadi, Azadeh | 2010

700 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: English
  3. Document No: 41105 (55)
  4. University: Sharif University of Technology, International Campus, Kish Island
  5. Department: Science and Engineering
  6. Advisor(s): Bagheri Shouraki, Saeed
  7. Abstract:
  8. Image segmentation is a technique which divides an image into significant parts. The accuracy of this technique plays an important role when it applies on medical images. Among various image segmentation methods, clustering methods have been extensively investigated and used. Since it is an unsupervised method, the existence of a small amount of side-information which is extracted from a specific application (in this case, medical image) could improve its accuracy. Using this side-information in clustering methods introduces a new generation of clustering approaches called semi-supervised clustering. This information usually has a format of pair-wise constraints and can be prepared easily with low cost. Medical image clustering does not make an exception to this rule. The side-information for improving the results and detecting human brain’s layers in a multispectral magnetic resonance images (MRI) could be achieved by an expert. The main goal of this thesis is to indicate that machine learning is a practical approach for many problems in medical image segmentation. In our proposed method, input data along with the supervision information are used to create an appropriate space that provides more accurate clustering results. To reach this goal, we first evaluate some existing unsupervised methods and the effect of having supervision information on MRI segmentation. Once the effectiveness of using supervision data is proven, a machine learning technique, i.e., kernel learning base technique, is used along with the supervision information to find the most efficient clusters in MRI. The obtained results are compared with the case where no supervision information is used. Our comparison results are highly promising and clearly show that the semi-supervised techniques can be successfully applied to segment high-dimensional data such as MRI
  9. Keywords:
  10. Image Segmentation ; Semi-Supervised Algorithm ; Magnetic Resonance Imagin (MRI) ; Semi-Supervised Clustering ; Normalized Cut Method

 Digital Object List

  • محتواي پايان نامه
  •   view

 Bookmark

No TOC