Machine learning challenges serve as a benchmark for determining state-of-the-art results in medical imaging. They provide direct comparisons between algorithms, and realistic estimates of generalization capability. By participating in the Aneurysm Detection And segMentation (ADAM) challenge, we learnt the most effective deep learning design choices to adopt when tackling automated brain aneurysm detection on multi-site data. Adjusting patch overlap ratio during inference, using a hybrid loss, resampling to uniform voxel spacing, using a 3D neural network architecture, and correcting for bias field were the most effective. We show that, when adopting these expedients, our model drastically improves detection performances.
This abstract and the presentation materials are available to members only; a login is required.