Use of Novel Thermography Features of Extraction and Different Artificial Neural Network Algorithms in Breast Cancer Screening

The breast thermography process is a physiological investigation that gives data dependent on the heat variations in the breast. It accounts for the heat circulation of a body utilizing the infrared radiation produced by the outside of that body. Precancerous tissue and the zone around a carcinogenic tumor have greater heats because of angiogenesis, and higher substance and blood vein action than a healthy breast; consequently, breast thermography can possibly recognize early strange changes in breast tissues. Thermography can identify the earliest indication of cancer initiation before mammography can notice. For the extraction of the scarce data from the breast, features like Energy, Effective information, Multi quadratic, Sigmoid, and Age of the patient are determined and applied to the neural network as inputs. Resilient backpropagation algorithm (RBPA) is a worldwide methodology managing weights; it is hard to get better subtleties from the breast image. To overcome the problems of RBPA, the artificial neural network (ANN) classifier is being used as a new derived Extension of Resilient backpropagation algorithm (ERBPA) for validation purposes. In this paper three ANN-based algorithms are used: Gradient descent, RBPA, and ERBPA are discussed and compared. As per the outcomes, the recently determined ERBPA is a progressively exact methodology to classify benign and malignant pathology. An accuracy of 99.90% has been obtained to bring about an effective strategy, which can recognize and cure breast cancer at an early stage.