Breast cancer is the leading cause of death for women in most countries. The high levels of mortality relate mostly to late diagnosis and to the direct proportionally relationship between breast density and breast cancer development. Therefore, the correct assessment of breast density is important to provide better screening for higher risk patients. However, in modern digital mammography the discrimination among breast densities is highly complex due to increased contrast and visual information for all densities. Thus, a computational system for classifying breast density might be a useful tool for aiding medical staff. Several machine-learning algorithms are already capable of classifying small number of classes with good accuracy. However, machinelearning algorithms main constraint relates to the set of features extracted and used for classification. Although well-known feature extraction techniques might provide a good set of features, it is a complex task to select an initial set during design of a classifier. Thus, we propose feature extraction using a Convolutional Neural Network (CNN) for classifying breast density by a usual machine-learning classifier. We used 307 mammographic images downsampled to 260x200 pixels to train a CNN and extract features from a deep layer. After training, the activation of 8 neurons from a deep fully connected layer are extracted and used as features. Then, these features are feedforward to a single hidden layer neural network that is cross-validated using 10-folds to classify among four classes of breast density. The global accuracy of this method is 98.4%, presenting only 1.6% of misclassification. However, the small set of samples and memory constraints required the reuse of data in both CNN and MLP-NN, therefore overfitting might have influenced the results even though we cross-validated the network. Thus, although we presented a promising method for extracting features and classifying breast density, a greater database is still required for evaluating the results.
As the technology evolves, the analog mammography systems are being replaced by digital systems. The digital system uses video monitors as the display of mammographic images instead of the previously used screen-film and negatoscope for analog images. The change in the way of visualizing mammographic images may require a different approach for training the health care professionals in diagnosing the breast cancer with digital mammography. Thus, this paper presents a computational approach to train the health care professionals providing a smooth transition between analog and digital technology also training to use the advantages of digital image processing tools to diagnose the breast cancer. This computational approach consists of a software where is possible to open, process and diagnose a full mammogram case from a database, which has the digital images of each of the mammographic views. The software communicates with a gold standard digital mammogram cases database. This database contains the digital images in Tagged Image File Format (TIFF) and the respective diagnoses according to BI-RADSTM, these files are read by software and shown to the user as needed. There are also some digital image processing tools that can be used to provide better visualization of each single image. The software was built based on a minimalist and a user-friendly interface concept that might help in the smooth transition. It also has an interface for inputting diagnoses from the professional being trained, providing a result feedback. This system has been already completed, but hasn’t been applied to any professional training yet.