Follicle Detection in Digital Ultrasound Images Using BEMD and Adaptive Clustering Algorithms

Abstract

Ultrasound imaging is one of the techniques used to study inside the human body with images generated using high-frequency sounds waves. The applications of ultrasound images include an examination of human body parts such as kidney, liver, heart, and ovaries. This paper mainly concentrates on ultrasound images of ovaries. Monitoring of follicle is important in human reproduction. This paper presents a method for follicle detection in ultrasound images using adaptive data clustering algorithms. The main requirements for any clustering algorithm are the number of clusters K. Estimating the value of K is a difficult task for given data. This paper presents an adaptive data clustering algorithm which generates accurate segmentation results with simple operation and avoids the interactive input K (number of clusters) value for segmentation of ultrasound image. The qualitative and quantitative results show that adaptive data clustering algorithms are more efficient than normal data clustering algorithms in segmenting the ultrasound image. After segmentation, using the region properties of the image, the follicles in the ovary image are identified. The proposed algorithm is tested on sample ultrasound images of ovaries for identification of follicles and with the region properties, the ovaries are classified into three categories, normal ovary, cystic ovary, and polycystic ovary with its properties. The experiment results are compared qualitatively with inferences drawn by medical expert manually and this data can be used to classify the ovary images.