Breast Cancer Detection From Histopathological Images Using Deep Learning

The cancer is the most dangerous diseases in the world, its mainly effective for women. So, our prime target must be curing the cancer through scientific investigation and the second main target should be early detection of cancer because the early detection of cancer can be helpful for remove the cancer completed. After reviewed 41 papers we found that several techniques are available for cancer detection. In this paper we proposed Deep Leaning algorithm convolutional neural network for diagnosed breast cancer using Mammograph MIAS database. The paper shows how we can use deep learning technology for diagnosis breast cancer using MIAS Dataset. Because deep learning techniques almost used for high task objective Computer Vision, Image processing, Medical Diagnosis, Neural Language Processing. But in this paper, we are applying deep learning technology on the MIAS Database and we have seen that is very beneficial for us for diagnosis breast cancer with accuracy 98%. This paper is divided in three parts first we have collect dataset and applied pre-processing algorithm for scaled and filter data then we have split dataset in training and testing purpose and generate some graph for visualization data. In last implement model on training dataset and achieved accuracy 98%. So, we have seen deep learning technology is a good way for diagnosis breast cancer with MIAS Dataset. This database provides200images and 12 features in the dataset. In this paper we have used 12 features for diagnosis breast cancer that we have got after pre-processing. But before train model we have applied some pre-processing algorithm like Watershed Segmentation, Colour based segmentation and Adaptive Mean Filters for scaled dataset then applied model and achieved accuracy. In this paper we also compare deep learning algorithm with other machine learning and seen our proposed system is proved best from others machine learning algorithm.