NeuVue: A Framework and Workflows for High-Throughput Electron Microscopy Connectomics Proofreading

Abstract

1.AbstractNeuVue is a software platform created for large-scale proofreading of machine segmentation and neural circuit reconstruction in high-resolution electron microscopy connectomics datasets. The NeuVue platform provides a robust web-based interface for proofreaders to collaboratively view, annotate, and edit segmentation and connectivity data. A backend queuing service organizes proofreader tasks into purpose-driven task types and increases proofreader throughput by limiting proofreader actions to simple, atomic operations. A collection of analytical dashboards, data visualization tools, and Application Program Interface (API) capabilities provide stakeholders real-time access to proofreading progress at an individual proofreader level as well as insights on task generation priorities. NeuVue is agnostic to the underlying data being proofread and improves upon the traditional proofreader experience through quality-of-life features that streamline complex editing operations such as splitting and merging objects in dense nanoscale segmentation.NeuVue heavily leverages cloud resources to enable proofreaders to simultaneously access and edit data on the platform. Production-quality features such as load-balancing, auto-scaling, and pre-deployment testing are all integrated into the platform’s cloud architecture. Additionally, NeuVue is powered by well-supported open-source connectomics tools from the community such as Neuroglancer, PyChunkedGraph, and Connectomics Annotation Versioning Engine (CAVE). The modular design of NeuVue facilitates easy integration and adoption of useful community tools to allow proofreaders to take advantage of the latest improvements in data visualization, processing, and analysis.We demonstrate our framework through proofreading of the mouse visual cortex data generated on the IARPA MICrONS Project. This effort has yielded over 40,000 proofreader edits across the 2 petavoxels of “Minnie” neuroimaging data. 44 unique proofreaders of various skill levels have logged a cumulative 3,740 proofreading hours, and we have been able to validate the improved connectivity of thousands of neurons in the volume. With sustained development on the platform, new integrated error detection and error correction capabilities, and continuous improvements to the proofreader model, we believe that the NeuVue framework can enable high-throughput proofreading for large-scale connectomics datasets of the future.

NEURD offers automated proofreading and feature extraction for connectomics

“…A key method for both validating and applying automatic edits was the functionality in Neuroglancer (Perlman, 2019) which allows the placement of point annotations to define a split in the PyChunkedGraph segmentation (Dorkenwald et al, 2022a,b). To facilitate the proofreading process, APL created a web-based interface called NeuVue (Xenes et al, 2022) that allows for the efficient queuing, review and execution of split suggestions in Neuroglancer. We built the logic required to translate mesh errors identified by NEURD into split point annotations that can be executed by the NeuVue pipeline.…”

“…Manual validation of these rules was performed in the context of standard proofreading and multi-soma splitting using the NeuVue Proofreading Platform (Xenes et al, 2022). We provided the proofreading team at John Hopkins University Applied Physics Laboratory (APL) with suggested error locations in the MICrONS volume, and experienced proofreaders evaluated each proposed split for accuracy (Fig.…”

“…To facilitate the proofreading process, APL created a web-based interface called NeuVue (Xenes et al, 2022) that allows for the efficient queuing, review and execution of split suggestions in Neuroglancer. We built the logic required to translate mesh errors identified by NEURD into split point annotations that can be executed by the NeuVue pipeline.…”

Exploiting large neuroimaging datasets to create connectome-constrained approaches for more robust, efficient, and adaptable artificial intelligence

“…Continued research into tools to assist and automate user queries and discovery using such data is critical to realize their full potential, such as scalable pipelines for proofreading and quality assessment. 40 Despite these challenges, the increasing scale, number, and diversity of large neuroimaging datasets represents an exciting opportunity to extract principles of structure and function from biological neural networks to incorporate into next generation artificial neural networks. We have demonstrated several ways biological insight can be extracted from these large datasets to design new machine learning approaches or augment existing computational models, overcoming the challenges discussed in Section 2.1.…”

“…39 Our team has extensive experience with storage, proofreading, and analysis of such datasets. 28,40,41 Given the speed with which such datasets are being collected and publicly disseminated by the US BRAIN Initiative and other efforts, our team has sought multiple approaches which can utilize this large scale structural information (and co-registered information about neuron function and cell properties) to influence the design of more efficient or capable neural network architectures by incorporating principles of neural connectivity (reflected in the pipeline in Fig. 3).…”