Planning of radiotherapy involves accurate segmentation of a large number of organs at risk, i.e. organs for which irradiation doses should be minimized to avoid important side effects of the therapy. We propose a deep learning method for segmentation of organs at risk inside the brain region, from Magnetic Resonance (MR) images. Our system performs segmentation of eight structures: eye, lens, optic nerve, optic chiasm, pituitary gland, hippocampus, brainstem and brain. We propose an efficient algorithm to train neural networks for an end-to-end segmentation of multiple and nonexclusive classes, addressing problems related to computational costs and missing ground truth segmentations for a subset of classes. We enforce anatomical consistency of the result in a postprocessing step, in particular we introduce a graph-based algorithm for segmentation of the optic nerves, enforcing the connectivity between the eyes and the optic chiasm. We report cross-validated quantitative results on a database of 44 contrast-enhanced T1-weighted MRIs with provided segmentations of the considered organs at risk, which were originally used for radiotherapy planning. In addition, the segmentations produced by our model on an independent test set of 50 MRIs are evaluated by an experienced radiotherapist in order to qualitatively assess their accuracy. The mean distances between produced segmentations and the ground truth ranged from 0.1 mm to 0.7 mm across different organs. A vast majority (96 %) of the produced segmentations were found acceptable for radiotherapy planning.

Over the last years, the oligometastatic disease state has gained more and more interest, and randomized trials are now suggesting an added value of stereotactic radiotherapy on all macroscopic disease in oligometastatic patients; but what barriers could impede widespread disease in some patients? In this review, we first discuss the concept of oligometastatic disease and some examples of clinical evidence. We then explore the route to dissemination: the hurdles a tumoral clone has to overtake before it can produce efficient and widespread dissemination. The spectrum theory argues that the range of metastatic patterns encountered in the clinic is the consequence of gradually obtained metastatic abilities of the tumor cells. Tumor clones can obtain these capabilities by Darwinian evolution, hence early in their genetic progression tumors might produce only a limited number of metastases. We illustrate selective dissemination by discussing organ tropism, the preference of different cancer (sub)types to metastasize to certain organs. Finally we discuss biomarkers that may help to distinguish the oligometastatic state.

No abstract