Exploring New Frontiers in Cervical Cancer Detection: A Survey and Preliminary Framework Results
Kiranmai Peddakasula, Raju Anitha · 2024
This study provides a comprehensive survey and proposes a novel framework for detecting cervical cancer through image analysis, focusing on the stages from initialization to edge detection. The survey component provides an in-depth examination of existing methodologies, emphasizing the challenges and advancements in cervix cancer detection. The proposed framework consists of ten major steps, beginning with Initialization and Setup, where parameters for cervical cancer detection are configured. Following that, image selection, pre-processing, segmentation, and edge detection are performed using techniques such as the Median Filter, binary conversion, and the Canny edge detection algorithm. The framework combines traditional and deep learning techniques, including the use of a Convolutional Neural Network (CNN) for complex feature extraction. To categorize images into Normal, Benign, or Malignant classes, the classification stage employs k-Nearest Neighbors (kNN) and Support Vector Machines (SVM). Eventually, this paper provides a survey of cervical cancer detection methodologies as well as an introduction to a robust framework that includes critical steps from initialization to edge detection. The results show that the proposed framework is effective at improving image quality and extracting relevant features for accurate classification, contributing to the advancement of cervical cancer detection methods.