AN AUTOMATED IMAGES-TO-GRAPHS PIPELINE FOR HIGH RESOLUTION CONNECTOMICS

Reconstructing a map of neuronal connectivity is a critical challenge in contemporary neuroscience. Recent advances in high-throughput electron microscopy have produced mas-sive images of nanoscale brain tissue for the first time. This resolution allows for individual neurons and their synaptic connections to be directly observed. Manually tracing each neuronal process at this scale is unmanageable, and there-fore researchers are developing automated image processing modules. Thus far, state-of-the-art algorithms focus only on the solution to a particular task (e.g., neuron segmentation or synapse identification). We have created the first fully automated images-to-graphs pipeline (i.e., a pipeline that begins with an imaged volume of neural tissue and produces a brain graph without any human interaction). To evaluate overall performance and select the best parameters and methods, we also de-velop a metric to assess the quality of the output graphs. We demonstrate our pipeline by selecting a set of algorithms and parameters, and search possible operating points to identify the best available brain graph. New algorithms can be easily integrated into our scalable pipeline, and used to improve results and facilitate analysis