# Synaptic Partner Assignment Using Attentional Voxel Association Networks

## TL;DR

This paper introduces an attentional voxel association network that directly predicts presynaptic and postsynaptic partners of synapses from electron microscopy images, improving connectomics mapping accuracy.

## Contribution

It presents a novel convolutional network approach that uses cleft masks as attention signals to accurately identify synaptic partners from volumetric data.

## Key findings

- Performs well on mouse somatosensory cortex dataset
- Effective in predicting both clefts and synaptic connections
- Enhances connectomics analysis accuracy

## Abstract

Connectomics aims to recover a complete set of synaptic connections within a dataset imaged by volume electron microscopy. Many systems have been proposed for locating synapses, and recent research has included a way to identify the synaptic partners that communicate at a synaptic cleft. We re-frame the problem of identifying synaptic partners as directly generating the mask of the synaptic partners from a given cleft. We train a convolutional network to perform this task. The network takes the local image context and a binary mask representing a single cleft as input. It is trained to produce two binary output masks: one which labels the voxels of the presynaptic partner within the input image, and another similar labeling for the postsynaptic partner. The cleft mask acts as an attentional gating signal for the network. We find that an implementation of this approach performs well on a dataset of mouse somatosensory cortex, and evaluate it as part of a combined system to predict both clefts and connections.