Soft computing method based computer aided screening of cervical cancer using cervigram

Abstract

Cervical cancer is the second dangerous disease in women around the world, especially in developing countries. At present, image processing techniques are used to detect and diagnose the cervical cancer in computer aided automatic approach. Most conventional methods used single cervical image to detect the cancer regions. The cancer segmentation accuracy of these conventional methods was low for the inclined cervical image cases. To overcome such issues in conventional methods, the image registration technique is proposed to increase the cancer detection and segmentation accuracy. This research work proposes a cervical cancer detection and segmentation methodology using Adaptive Neuro Fuzzy Inference System (ANFIS) classifier. This method consists of image registration, feature extraction, classification and segmentation. Image registration is applied on the source cervical image with respect to reference cervical image. Here FFT is used for Image registration technique. The features are extracted from registered cervical image in order to differentiate the normal image from abnormal image. Then, these extracted features are classified (cancerous or non-cancerous cervical image) using ANFIS classifier based on the comparison with the trained data. Performance analysis of this method results in 97.42% Sensitivity, 99.36 % Specificity and 99.36% segmentation accuracy. newline