Abstract:

A computer-aided diagnosis system is the most important step for implementing an automated cell nuclei segmentation using cancer cells. Breast cancer is one of the world's major public health issues affecting women. It has two states: benign and malignant. Benign states are slow to develop, rarely spread to other parts of the body, and have well-defined borders. Malignant state, on the other hand, has a tendency to grow faster and is life threatening. Histopathological Images (HIs) are the gold standard for evaluating certain types of tumors for cancer diagnosis. Image segmentation techniques aim to identify and extract foreground objects in an image, resulting in individual segments. Image segmentation is fundamentally different from one type of image to the next because each has its own context and different geometrical properties, posing a challenge in designing a generic algorithmic procedure. In this paper, an effort is made to compare and study the efficiency of colour image segmentation using Fuzzy C-Means Clustering, segmentation using K-Means Clustering, Watershed Segmentation using Gradient, and proposed H&E Stained based Marker-Controlled towards tumor detection segmentation. The analysis concluded that the performance of Watershed Segmentation using Marker-Controlled produced better results than other techniques.

Keywords: