Deep learning techniques on texture analysis of chest and breast images

The deep learning techniques can automatically learn texture and image context features from training data without the need of explicit feature engineering. With proper and sufficient training data, useful features can be learned to support various medical image analysis applications. In this chapter, we discuss recent deep learning studies on the analysis of chest and breast images. The specific elaborated computerized techniques are computer-aided detection, computer-aided diagnosis, and automatic semantic mapping. We show that the feature learned with the deep learning techniques can be helpful to boost the performances of these computerized applications for the analysis of medical images.