Deep Convolutional Neural Networks (DCNN), are used in several areas of a medical image, making possible identification of an irrelevant object exactly where it appears in an image, its orientation or even its scaling. In this work we propose a DCNN model with detailed ontology of different types of the most common forms of cancer and integrate it with our PPIR module for simulation in medical imaging. Our focus is to train it separately for each particular form of cancer and each particular view. The DCNN model will have as input a sequence of images, known images obtained from TCIA (The Cancer Imaging Archive). We aim to use PyTorch to load the dataset and preprocess to prepare for training, specify activation functions, regularization, and other parameters. To train the DCNN we use the training dataset in Matlab and PyTorch's optimization module by specifying a loss function, such as cross-entropy loss, and an optimizer, such as stochastic gradient descent, the learning rate and number of epochs. Finally, we want to evaluate the performance of the trained DCNN using the validation dataset. In our study we used a dataset of MRI patients’ examinations. Even though we are in the first steps of building up a model, we aim in the future to use the same DCNN model trained for different forms and levels of cancer