An Automated Approach to Detect Breast Cancer Tissue Using Ultrasound Images

Cancer is a kind of cell that develops out of control of the regular drives that normalizes growth. Breast cancer is one of the deadly forms of cancers for the human. Initial detection of cancerous tissues is likely to monitor the patient’s strength to deal with the disease and live more. Electronic modalities are applied to diagnose breast cancers. Due to low cost and safety, ultrasound scanning is considered as one of the most frequently used electronic modalities for breast cancer detection. This research study proposed an automated approach for breast cancer detection using ultrasound gray-scale images. To this end, enhancement operation was used at the beginning to minimize the grayscale color shifts. Median filter was used to facilitate better segmentation removing maximum unwanted noises. Threshold-based segmentation (Threshold based OTSU) was used as the test images used in this study are gray-scale images. Finally, necessary feature information that is used in the pathology laboratory by pathology experts is used in this study to identify the ROI (Region of interests) to detect breast cancer tissues. This study proposed approach achieved an average accuracy rate of 93.11%. The depicted experimental outcomes of the proposed approach outperform other contemporary existing available approaches in terms of accuracy while maintaining the medical experts’ acceptable accuracy rate.