Neuromuscular fiber segmentation through particle filtering and discrete optimization

Abstract

We present an algorithm to segment a set of parallel, intertwined and bifurcating bers from 3D images, targeted for the identi cation of neuronal bers in very large sets of 3D confocal microscopy images. The method consists of preprocessing, local calculation of ber probabilities, seed detection, tracking by particle ltering, global supervised seed clustering and nal voxel segmentation. The preprocessing uses a novel random local probability ltering (RLPF). The ber probabilities computation is performed by means of SVM using steerable lters and the RLPF outputs as features. The global segmentation is solved by discrete optimization. The combination of global and local approaches makes the segmentation robust, yet the individual data blocks can be processed sequentially, limiting memory consumption. The method is automatic but e cient manual interactions are possible if needed. The method is validated on the Neuromuscular Projection Fibers dataset from the Diadem Challenge. On the 15 rst blocks presented, our method has a 99.4% detection rate. We also compare our segmentation results to a state-of-the-art method. On average, the performances of our method are either higher or equivalent to that of the state-of-the-art method but less user interactions are needed in our approach.