Article Preview
Top1. Introduction
In every country of the world, Cancer ranks as one of the leading causes of death. Ovarian cancer is the third most common gynaecological cancer in Indian women. From age 35, the risk of Ovarian cancer increases and reaches a peak between 55-64 (Labidi-Galy et al., 2012). Ovarian cancer has the worst prognosis as it is diagnosed in advanced stages such as stage III or IV. Since the cause of ovarian cancer is unknown, effective screening strategies for it are not found. The process of ovarian cancer segmentation using MRI and CT images has received considerable amount of attention in biomedical image segmentation and classification. As per the National Cancer Institute statistics, around 19,880 people were diagnosed with ovarian cancer and approximately 12,810 deaths were due to ovarian cancer by 2022 (National Cancer Institute, 2023). Greater efforts have been made for the early detection of ovarian cancer. Computer-aided diagnosis system is less time consuming and also gives more accurate and reliable results. With the advancement in the Image Processing techniques, Machine Learning algorithms, Deep Learning models and more importantly the computational devices, it is more evident to there is tremendous developments in the CAD systems. In the literature there are many works which has focused on the development of CADs using various medical images (Alharbi & Tchier, 2017; Bron et al., 2017; Chang et al., 2017; Chen et al., 2014; de Carvalho Filho et al., 2017; Nishio & Nagashima, 2017; Wang et al., 2016; Yilmaz et al., 2017). There are CADs which are developed for diagnosing the various cancer such as liver, brain, breast etc using CT and other modalities using Image Processing techniques (Chang et al., 2017), automatic image analysis for detection of various cancers and classifying as benign and malignant using classical machine learning algorithms (Nishio & Nagashima, 2017), there are many works which are carried out for the detection of ovarian cancer with the help of Deep Convolutional Neural Networks (Alharbi & Tchier, 2017; Bron et al., 2017; Chen et al., 2014; de Carvalho Filho et al., 2017; Suzuki, 2017; Wang et al., 2016; Yilmaz et al., 2017).
The rest of the paper is organized as follows. Section 2 presents the literature survey, with the detailed methodology and the description of algorithms is presented in Section 3, with the experimental results reported in Section 3, Section 4 provides conclusion.