Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 48796 (19)
- University: Sharif University of Technology
- Department: Computer Engineering
- Advisor(s): Habibi, Jafar
- Abstract:
- Digital image analysis of histological datasets is a currently expanding field of research. Histological images are inherently complex in nature and contain a wide variety of visual information. Graph-based methods have recently gained immense popularity, as these methods can effectively describe tissue architecture and provide adequate numeric information for subsequent computer-based analysis. Graphs have the ability to represent spatial arrangements and neighborhood relationships of different tissue components, which are essential characteristics observed visually by pathologists during the investigation of specimens. In this thesis, we proposed an automatic approach for classification histological images to benign or malignant. The contributions of this thesis are using an automatic method for cell segmentation, extracting image features with graph features and using a regression-based method instead of classification. The aforementioned contributions yield better accuracy (%96:8) rather than the other approaches. In recent, we designed and proposed a framework for facilitating future works in this field of research that can be used for methods comparison
- Keywords:
- Classification ; Feature Extraction ; Cancer Diagnosis ; Microscopic Image ; Gragh Based Method ; Histological Images Analysis ; Tissue Type Diagnosis
Digital Object List
محتواي کتاب
Bookmark