A Convolutional Neural Network Approach for The Diagnosis of Breast Cancer

Most common cancer in females is found to be Breast cancer which is a widespread disease. One out of eight females worldwide are affected by this cancer only. We can detect this cancer by detecting malignancy from breast tissues. There are various types of computer-aided techniques and approaches which are used by doctors for detecting cancer. The major objective of this paper is to build a well-defined model for the recognition of breast cancer by expending various parameters. Different types of machine learning and deep learning methodologies are used for the classification of malignant and benign tissues. In this we are using a dataset that obtains 569 samples with 30 features, this dataset is majorly called the Wisconsin dataset. Many techniques are implemented on this dataset we are using deep convolutional neural network (CNN) and Machine learning methodology (KNN) for the diagnosis and training purpose and then compare the results of both the techniques. Deep convolutional NN is implemented on the google platform called the Google Colab on the other side KNN is implemented on the Anaconda Spyder platform. The best accuracy achieved from KNN is 96.49%. To improve the performance and accuracy we implemented CNN on the same dataset and then achieved 99.41% accuracy. Deep learning is extensively useful in getting the best and optimal results in other performance matrics such as precision, recall, F1-score and AVC-ROC - 98.64%,97.61 %, 98.08%, 97.61% respectively.