A Novel and Robust Semi-Automated Best Frame Selection Method from Strain Videos

Abstract

Breast cancer is the second leading cause of death of women worldwide. Accurate lesion boundary detection is important for breast cancer diagnosis. Since many crucial features for discriminating benign and malignant lesions are based on the contour, shape, and texture of the lesion, an accurate segmentation method is essential for a successful diagnosis. Ultrasound is an effective screening tool and primarily useful for differentiating benign and malignant lesions. However, due to inherent speckle noise and the low contrast of breast ultrasound imaging, automatic lesion segmentation is still a challenging task. This research focuses on developing a novel, effective, and fully automatic lesion segmentation method for breast ultrasound images. By incorporating empirical domain knowledge of breast structure, a region of interest is generated. Then, a novel enhancement algorithm (using a novel phase feature) and a newly developed neutrosophic clustering method is developed to detect the precise lesion boundary. Neutrosophy is a recently introduced branch of philosophy that deals with paradoxes, contradictions, antitheses, and antinomies. When neutrosophy is used to segment images with vague boundaries, its unique ability to deal with uncertainty is brought to bear. In this work, we apply neutrosophy to breast ultrasound image segmentation and propose a new clustering method named neutrosophic l-means. We compare the proposed method with traditional fuzzy c-means clustering and three other well-developed segmentation methods for breast ultrasound images, using the same database. Both accuracy and time complexity are analyzed. The proposed method achieves the best accuracy (TP rate is 94.36%, FP rate is 8.08%, and the similarity rate is 87.39%) with a fairly rapid processing speed (about 20 seconds). Sensitivity analysis shows the robustness of the proposed a method as well. Cases with multiple-lesions and severe shadowing effect (shadow areas having similar intensity values of the lesion and tightly connected with the lesion) is not included in this study. Ultrasound is one of the ways of detecting and identifying breast tumor. From the raw US signal (known as RF data), we get the B-mode image, in which the tumor might not be that much visible. For better tumor visibility, we need to form a strain image. From pre and post compressed US image of the breast, we can estimate strain and form a strain video. Each frame of the video does not have good tumor visibility. It is difficult and time consuming for a doctor to accurately detect the shape of the tumor from rapidly changing frames. Selecting the frames where the tumor is comparably more visible will help the

doctor/radiologist to detect the tumor more easily. In this paper, a method of semi-automated best frame selection from a strain video is proposed. The method involves two ways to select the required frames and to show the best output frames in the form of a video. It is based on Mean Pixel Difference (MPD) and contrast as the Image Descriptors.