21st Annual Conference on Medical Image Understanding and Analysis, MIUA 2017

The proceedings contain 81 papers. The special focus in this conference is on Medical Image Understanding and Analysis. The topics include: End-to-end learning of a conditional random field for intra-retinal layer segmentation in optical coherence tomography; superpixel-based line operator for retinal blood vessel segmentation; automatic detection and identification of retinal vessel junctions in colour fundus photography; fast optic disc segmentation in retinal images using polar transform; a novel technique for splat generation and patch level prediction in diabetic retinopathy; deep residual networks for quantification of muscle fiber orientation and curvature from ultrasound images; multi-level trainable segmentation for measuring gestational and yolk sacs from ultrasound images; weakly supervised learning of placental ultrasound images with residual networks; edge aware geometric filter for ultrasound image ENHA; tissues classification of the cardiovascular system using texture descriptors; multidimensional assessments of abdominal aortic aneurysms by magnetic resonance against ultrasound diameter measurements; comparison of automatic vessel segmentation techniques for whole body magnetic resonance angiography with limited ground truth data; evaluating classifiers for atherosclerotic plaque component segmentation in MRI; classification of cross-sections for vascular skeleton extraction using convolutional neural networks; improved CTA coronary segmentation with a volume-specific intensity threshold; multi task fully convolutional network for brain tumour segmentation; classification of cervical-cancer using pap-smear images; new level set model in follow up radiotherapy image analysis and topological analysis of the vasculature of angiopoietin-expressing tumours through scale-space tracing.