||Dynamic contrast enhanced (DCE) MR of the liver contains valuable information regarding liver function and anatomy. Automatic image processing and analysis of these images can aid the radiologist in diagnosis and treatment planning. My work focusses on the automatic detection and classification of liver lesions. It involves motion correction between the contrast phases of the DCE MR images, automatic liver segmentation and lesion detection using convolutional neural networks, and lesion classification using extremely random forests with features derived from DCE MR and T2-weighted MR images. Ongoing work aims to register two longitudinal DCE MR scans to ease the follow-up on lesions and access the treatment result.