image regression

Mammography Image Quality Assurance Using Deep Learning

Author(s)3: Tobias Kretz, Klaus-Robert Müller, Tobias Schaeffter, Clemens Elster
Mammography Image Quality Assurance Using Deep Learning 170 177 IEEE Transactions on Biomedical Engineering (TBME)
Image quality assurance is crucial in mammography to ensure reliable breast cancer diagnostics. Analyzing images of a technical phantom allows to routinely and reliably assess image quality. Current state-of-the-art analysis determines local image quality features by applying pre-processing and regression procedures for a set of repeatedly recorded images. This proof of concept paper demonstrates that mammography image quality assessment can benefit from deep learning. A neural network is trained on a large database of phantom images, and it is shown that the trained net retrieves the local image quality features already from single images without cumbersome pre-processing. This allows to maintain quality standards at significantly less labor. read more