As in many areas of computer vision, deep networks now deliver state-of-the-art results for delineation tasks, such as finding axons and dendrites in 3D light microscopy images. Most of the existing approaches rely on convolutional networks to extract from images binary masks denoting which voxels belong to neurites and which do not. Unfortunately, they do not guarantee that the connectivity of the produced masks corresponds to that of the real neurite network. This is because these methods are trained to minimize losses, such as cross-entropy and mean squared error, that do not explicitly enforce topological consistency. When the annotations do not perfectly coincide with the imaged structures, networks trained with the per-voxel losses produce results plagued by topological errors, such as interruptions and false positive connections.
The aim of this project is to implement and test a 3D connectivity-oriented loss function in order to prevent topological inconsistencies, by using a similar approach to this paper.
The candidate should have Python programming experience. Previous experience with deep learning, particularly using PyTorch, is a plus.
For further information, send an email to [email protected].