Active Learning Pipeline for Brain Mapping in a High Performance Computing Environment

TL;DR

This paper presents a scalable active learning pipeline for large-scale brain mapping that leverages high performance computing to significantly increase throughput and reduce manual annotation effort.

Contribution

The paper introduces a novel high-performance computing pipeline for brain mapping that achieves over 100x throughput improvements with minimal processing time increase.

Findings

Achieved 100x increase in image processing throughput

Pipeline reduces manual annotation workload

Demonstrated robust scalability on Xeon CPUs

Abstract

This paper describes a scalable active learning pipeline prototype for large-scale brain mapping that leverages high performance computing power. It enables high-throughput evaluation of algorithm results, which, after human review, are used for iterative machine learning model training. Image processing and machine learning are performed in a batch layer. Benchmark testing of image processing using pMATLAB shows that a 100$\times$ increase in throughput (10,000%) can be achieved while total processing time only increases by 9% on Xeon-G6 CPUs and by 22% on Xeon-E5 CPUs, indicating robust scalability. The images and algorithm results are provided through a serving layer to a browser-based user interface for interactive review. This pipeline has the potential to greatly reduce the manual annotation burden and improve the overall performance of machine learning-based brain mapping.