Detection And Classification of Uterine Fibroid Using Ultrasound Images

Abstract

Ultrasound imaging technique is the most effective tool used for detecting fibroids present in uterus. Mostly women’s are affected from the ovarian cyst and uterine fibroid diseases after their age 50. Generally fibroid affects the uterus. By the use of this Ultrasound imaging technique we easily found the size and type of fibroids. It is an initial method, because 90% of fibroids size was easily found and here no harmful radiation will be produced from the device. In the existing method histogram equalization are used in the preprocessing step. Then gray thresholding is applied to convert the image into the binary one. Then active contour-based image segmentation is done. After that Then active contour-based image segmentation and HOG feature extraction. Pre-processing is done for adjusting image intensities. In this project Histogram Equalization technique is used to adjust the intensity of image. By using active contour image segmentation, the contour energy is minimized and specifies the curve on the image that moves to find the object boundaries. The mask argument specifies the initial state of the contour. HOG (Histogram Oriented Gradient) feature extraction is performed on individual objects to compute HOG on each cell. After feature extraction the images are classified using Artificial Neural Network. Index Terms – Ultrasound, fibroid, Uterine Fibroid, HOG (Histogram Oriented Gradient).