Hidden Markov Modeling for Maximum Likelihood Neuron Reconstruction

Recent advances in brain clearing and imaging have made it possible to image entire mammalian brains at sub-micron resolution. These images offer the potential to assemble brain-wide atlases of projection neuron morphology, but manual neuron reconstruction remains a bottleneck. In this paper we present a probabilistic method which combines a hidden Markov state process that encodes neuron geometric properties with a random field appearance model of the flourescence process. Our method utilizes dynamic programming to efficiently compute the global maximizers of what we call the "most probable" neuron path. We applied our algorithm to the output of image segmentation models where false negatives severed neuronal processes, and showed that it can follow axons in the presence of noise or nearby neurons. Our method has the potential to be integrated into a semi or fully automated reconstruction pipeline. Additionally, it creates a framework for conditioning the probability to fixed start and endpoints through which users can intervene with hard constraints to, for example, rule out certain reconstructions, or assign axons to particular cell bodies.