Breast Cancerous Nucleion Digital Histogram Segmentation
Ranveer Pratap Lal, Mamta Patankar, Vijayshri Chaurasia, Madhu Shandilya, Ana Kumar, Vikas Gupta · 2023
Most frequent cancer which happens in women is breast cancer. One of the newest fields in today's diagnostic system is digital pathology. Image segmentation is an important step before cancer classification. When there are touching or overlapping nuclei for separation, the main issue arises. The thresholding technique is used in this thesis to present an image segmentation method. The proposed method employs OTSU thresholding followed by binarization and morphological operations. In post-processing, the unwanted region in the image is removed. The experimental results on the used datasets validate the efficiency of the suggested approach. It offers improved segmentation of breast histopathology images with high accuracy and minimal complexity as compared to state-of-the-art techniques.