Downloads: 93 | Views: 145

M.Tech / M.E / PhD Thesis | Electronics & Communication Engineering | India | Volume 4 Issue 2, February 2015

Robust Segmentation and Classification of Histopathological Image

Jincy Wilson | Vipin V. R.

Abstract: Cancer is an uncontrolled cell growth caused due to abnormal cell division in any region of the body. When cancer is diagnosed at an early stage the treatment is simpler and effective. There are different types of cancer and each is classified by the type of cell that is initially affected. In this paper, early diagnosis of colon cancer is considered. In the proposed method, segmentation of tissue is done by graph cut, after the pre-processing of the histopathological image. Next, the features are extracted from segmented image, by structural and statistical methods. In structural approach, intensity histogram based features such as mean, variance etc. are computed. In statistical feature extraction, gray level co-occurrence matrix are used to find out the relation between pixels and thus energy, entropy, contrast etc. are measured. These extracted features are given to Support Vector Machine (SVM) classifier to classify them as cancerous and noncancerous. This system can be used in many real time problems like bioinformatics also.

Keywords: Cancer, Histopathology, Structural, Statistical, Graph cut, Gray Level Co-occurrence Matrix, Support Vector Machine

Edition: Volume 4 Issue 2, February 2015,

Pages: 1335 - 1340