A Scalable Inference Pipeline for 3D Axon Tracing Algorithms

2022 IEEE High Performance Extreme Computing Conference (HPEC) 19-23 September 2022 Waltham, MA, USA

Bibliographic Details
Main Authors: Fenelon, Benjamin, Gjesteby, Lars A., Guan, Webster, Park, Juhyuk, Chung, Kwanghun, Brattain, Laura J.
Other Authors: Lincoln Laboratory
Format: Article
Language:English
Published: IEEE|2022 IEEE High Performance Extreme Computing Conference (HPEC) 2024
Online Access:https://hdl.handle.net/1721.1/155787
_version_ 1811081859936813056
author Fenelon, Benjamin
Gjesteby, Lars A.
Guan, Webster
Park, Juhyuk
Chung, Kwanghun
Brattain, Laura J.
author2 Lincoln Laboratory
author_facet Lincoln Laboratory
Fenelon, Benjamin
Gjesteby, Lars A.
Guan, Webster
Park, Juhyuk
Chung, Kwanghun
Brattain, Laura J.
author_sort Fenelon, Benjamin
collection MIT
description 2022 IEEE High Performance Extreme Computing Conference (HPEC) 19-23 September 2022 Waltham, MA, USA
first_indexed 2024-09-23T11:53:37Z
format Article
id mit-1721.1/155787
institution Massachusetts Institute of Technology
language English
last_indexed 2024-09-23T11:53:37Z
publishDate 2024
publisher IEEE|2022 IEEE High Performance Extreme Computing Conference (HPEC)
record_format dspace
spelling mit-1721.1/1557872024-09-20T17:29:39Z A Scalable Inference Pipeline for 3D Axon Tracing Algorithms Fenelon, Benjamin Gjesteby, Lars A. Guan, Webster Park, Juhyuk Chung, Kwanghun Brattain, Laura J. Lincoln Laboratory Massachusetts Institute of Technology. Institute for Medical Engineering & Science 2022 IEEE High Performance Extreme Computing Conference (HPEC) 19-23 September 2022 Waltham, MA, USA High inference times of machine learning-based axon tracing algorithms pose a significant challenge to the practical analysis and interpretation of large-scale brain imagery. This paper explores a distributed data pipeline that employs a SLURM-based job array to run multiple machine learning algorithm predictions simultaneously. Image volumes were split into N (1–16) equal chunks that are each handled by a unique compute node and stitched back together into a single 3D prediction. Preliminary results comparing the inference speed of 1 versus 16 node job arrays demonstrated a 90.95% decrease in compute time for 32 GB input volume and 88.41% for 4 GB input volume. The general pipeline may serve as a baseline for future improved implementations on larger input volumes which can be tuned to various application domains. 2024-07-25T14:20:14Z 2024-07-25T14:20:14Z 2022-09-19 2024-07-25T13:42:40Z Article http://purl.org/eprint/type/ConferencePaper https://hdl.handle.net/1721.1/155787 Fenelon, Benjamin, Gjesteby, Lars A., Guan, Webster, Park, Juhyuk, Chung, Kwanghun et al. 2022. "A Scalable Inference Pipeline for 3D Axon Tracing Algorithms." 00. en 10.1109/hpec55821.2022.9926403 Creative Commons Attribution-Noncommercial-ShareAlike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf IEEE|2022 IEEE High Performance Extreme Computing Conference (HPEC) Author
spellingShingle Fenelon, Benjamin
Gjesteby, Lars A.
Guan, Webster
Park, Juhyuk
Chung, Kwanghun
Brattain, Laura J.
A Scalable Inference Pipeline for 3D Axon Tracing Algorithms
title A Scalable Inference Pipeline for 3D Axon Tracing Algorithms
title_full A Scalable Inference Pipeline for 3D Axon Tracing Algorithms
title_fullStr A Scalable Inference Pipeline for 3D Axon Tracing Algorithms
title_full_unstemmed A Scalable Inference Pipeline for 3D Axon Tracing Algorithms
title_short A Scalable Inference Pipeline for 3D Axon Tracing Algorithms
title_sort scalable inference pipeline for 3d axon tracing algorithms
url https://hdl.handle.net/1721.1/155787
work_keys_str_mv AT fenelonbenjamin ascalableinferencepipelinefor3daxontracingalgorithms
AT gjestebylarsa ascalableinferencepipelinefor3daxontracingalgorithms
AT guanwebster ascalableinferencepipelinefor3daxontracingalgorithms
AT parkjuhyuk ascalableinferencepipelinefor3daxontracingalgorithms
AT chungkwanghun ascalableinferencepipelinefor3daxontracingalgorithms
AT brattainlauraj ascalableinferencepipelinefor3daxontracingalgorithms
AT fenelonbenjamin scalableinferencepipelinefor3daxontracingalgorithms
AT gjestebylarsa scalableinferencepipelinefor3daxontracingalgorithms
AT guanwebster scalableinferencepipelinefor3daxontracingalgorithms
AT parkjuhyuk scalableinferencepipelinefor3daxontracingalgorithms
AT chungkwanghun scalableinferencepipelinefor3daxontracingalgorithms
AT brattainlauraj scalableinferencepipelinefor3daxontracingalgorithms