A Novel method for IDC Prediction in Breast Cancer Histopathology images using Deep Conventional Neural Network

Ali Noshad¹ Sanaz Moradi²

1) Student of Computer Engineering,Iran University of Salman Farsi Kazerun Email:
2) Student of Computer Engineering,Iran University of Salman Farsi Kazerun Email:

Publication : 2nd. International Congress on science & Engineering - paris(parisconf.com)
Abstract :
Breast cancer as a global problem has affected the lives of many women around the world for decades. Breast cancer is a major global health challenge, according to researches. IDC is the most common type of breast cancer, accounting for 80 percent of all breast cancers, according to the American Cancer Society. The leading cause of death among women with this type of cancer is the lack of early detection of invasive IDC. Therefore, for the successful treatment of breast cancer, this diagnosis is of particular importance. In this case, the diagnosis of IDC in histological images requires microscopy and manual study of various slides, so that they can be classified as positive or negative cancer. Despite this time-consuming and tedious process, it is obvious that the result of this manual process to identify breast cancer is not error-free and leads to misdiagnosis due to human cognitive limitations. Therefore, early detection of breast cancer is a challenging task it can use the development of diagnostic models. The study introduces a new Conventional Neural Networks (CNN) architecture, as a deep learning approach to detect IDC from histological images. RGB microscopic images of breast cancer histology were used to train and test the model. The proposed model presented in this study, in the classification of positive and negative IDCs, achieved an accuracy of 82% and according to the evaluation results, the proposed model, based on accuracy, precision, sensitivity and F-score, compared to existing methods based on Machine learning has performed better in classifying positive and negative IDCs.
Keywords : Image Procession Deep Learning Machine Learning Conventional Neural Networks Breast Cancer