GAN, CNN and ELM Based Breast Cancer Detection
Vandana Kate, R Ushasree, R.M. Tharsanee, Mrs. Taranjeet Kukreja, Shailendra Saraf, Bhoopathy Varadharajan
Abstract
The most prevalent kind of cancer among females is known as breast cancer. One of the most significant challenges facing modern medicine is improving the accuracy of breast cancer screenings. In addition to lowering costs associated with treatment, accurate diagnosis of cancer, regardless of whether it is benign or malignant, is of utmost importance. Image processing is a tool that can detect breast cancer in its early stages without causing any damage. The processing of images comes with a number of inherent uncertainties that can be caused by a variety of factors, including the sampling to noise ratio, the initial digitalization, the intensity, and the special domain. We put the Extreme Learning Machine (ELM) to work classifying, identifying, and instructing. This research suggests a GAN-CNN-ELM-based technique for detecting breast cancer. This method produces an accuracy of about 99% which outperforms the GAN-CNN model.