A Cascaded Feature Extraction for Diagnosis of Ovarian Cancer in CT Images

This paper proposed ovarian cancer detection in the image using joint feature extraction and an efficient Net model. The noise of input is filtered by Improved NLM (Improved Non-Local Means) filtering. deep features are extracted Deep CNN_RSO (Deep Convolutional Neural Network Rat Swarm Optimization), low-level texture ILBP (Interpolated Local Binary Pattern or Interpolated LBP). To improve reduce error, use a cascading technique for extraction. RSO also helps to efficiently optimize DCNN from images. Finally, classified Efficient classifier, which performs global average summary classification (normal abnormal). system’s performance implemented on Cancer Genome Atlas Ovarian (TCGA-OV) dataset. performance, like sensitivity, specificity, accuracy error rates, shows better with respect other techniques.