Breast Cancer Classification Using Deep Neural Networks Technique

Breast cancer is one of the important reasons for the demise of women globally. Precise and initial recognition of breast cancer can confirm lasting survival for the patients. Deep neural network-based techniques can assist pathologists and doctors in precisely detecting abnormalities. The efficacy of our proposed artificial deep neural network A-DNN model is tested on Breast Cancer data sets (Wisconsin Diagnostic-WDBC), and the outcomes demonstrate that the A-DNN model outstrips other evaluation approaches. The key objective of this research work was to apply a proposed approach in an actual medical investigative system and thus assist medical doctors in making exact and actual decisions in the future. The A-DNN model was studied in experimental work at numerous training-testing partitions of the WDBC data set. The A-DNN model obtains a micro average classification accuracy of 99%, precision of 99%, and recall of 99% on the 70-30% partitioning of a WDBC dataset. The A-DNN classifier model gives an accuracy of 99% representing promising results compared to the other research study.