Enhancing the data processing speed of a deep-learning-based three-dimensional single molecule localization algorithm (FD-DeepLoc) with a combination of feature compression and pipeline programming

Three-dimensional (3D) single molecule localization microscopy (SMLM) plays an important role in biomedical applications, but its data processing is very complicated. Deep learning is a potential tool to solve this problem. As the state of art 3D super-resolution localization algorithm based on deep...

Full description

Saved in:
Bibliographic Details
Main Authors: Shuhao Guo, Jiaxun Lin, Yingjun Zhang, Zhen-Li Huang
Format: Article
Language:English
Published: World Scientific Publishing 2025-03-01
Series:Journal of Innovative Optical Health Sciences
Subjects:
Online Access:https://www.worldscientific.com/doi/10.1142/S1793545824500251
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849389142069215232
author Shuhao Guo
Jiaxun Lin
Yingjun Zhang
Zhen-Li Huang
author_facet Shuhao Guo
Jiaxun Lin
Yingjun Zhang
Zhen-Li Huang
author_sort Shuhao Guo
collection DOAJ
description Three-dimensional (3D) single molecule localization microscopy (SMLM) plays an important role in biomedical applications, but its data processing is very complicated. Deep learning is a potential tool to solve this problem. As the state of art 3D super-resolution localization algorithm based on deep learning, FD-DeepLoc algorithm reported recently still has a gap with the expected goal of online image processing, even though it has greatly improved the data processing throughput. In this paper, a new algorithm Lite-FD-DeepLoc is developed on the basis of FD-DeepLoc algorithm to meet the online image processing requirements of 3D SMLM. This new algorithm uses the feature compression method to reduce the parameters of the model, and combines it with pipeline programming to accelerate the inference process of the deep learning model. The simulated data processing results show that the image processing speed of Lite-FD-DeepLoc is about twice as fast as that of FD-DeepLoc with a slight decrease in localization accuracy, which can realize real-time processing of [Formula: see text] pixels size images. The results of biological experimental data processing imply that Lite-FD-DeepLoc can successfully analyze the data based on astigmatism and saddle point engineering, and the global resolution of the reconstructed image is equivalent to or even better than FD-DeepLoc algorithm.
format Article
id doaj-art-c3ffca2196a24a4dac59e0c492b80403
institution Kabale University
issn 1793-5458
1793-7205
language English
publishDate 2025-03-01
publisher World Scientific Publishing
record_format Article
series Journal of Innovative Optical Health Sciences
spelling doaj-art-c3ffca2196a24a4dac59e0c492b804032025-08-20T03:42:02ZengWorld Scientific PublishingJournal of Innovative Optical Health Sciences1793-54581793-72052025-03-01180210.1142/S1793545824500251Enhancing the data processing speed of a deep-learning-based three-dimensional single molecule localization algorithm (FD-DeepLoc) with a combination of feature compression and pipeline programmingShuhao Guo0Jiaxun Lin1Yingjun Zhang2Zhen-Li Huang3Key Laboratory of Biomedical Engineering of Hainan Province, School of Biomedical Engineering, Hainan University, Sanya 572025, P. R. ChinaKey Laboratory of Biomedical Engineering of Hainan Province, School of Biomedical Engineering, Hainan University, Sanya 572025, P. R. ChinaKey Laboratory of Biomedical Engineering of Hainan Province, School of Biomedical Engineering, Hainan University, Sanya 572025, P. R. ChinaKey Laboratory of Biomedical Engineering of Hainan Province, School of Biomedical Engineering, Hainan University, Sanya 572025, P. R. ChinaThree-dimensional (3D) single molecule localization microscopy (SMLM) plays an important role in biomedical applications, but its data processing is very complicated. Deep learning is a potential tool to solve this problem. As the state of art 3D super-resolution localization algorithm based on deep learning, FD-DeepLoc algorithm reported recently still has a gap with the expected goal of online image processing, even though it has greatly improved the data processing throughput. In this paper, a new algorithm Lite-FD-DeepLoc is developed on the basis of FD-DeepLoc algorithm to meet the online image processing requirements of 3D SMLM. This new algorithm uses the feature compression method to reduce the parameters of the model, and combines it with pipeline programming to accelerate the inference process of the deep learning model. The simulated data processing results show that the image processing speed of Lite-FD-DeepLoc is about twice as fast as that of FD-DeepLoc with a slight decrease in localization accuracy, which can realize real-time processing of [Formula: see text] pixels size images. The results of biological experimental data processing imply that Lite-FD-DeepLoc can successfully analyze the data based on astigmatism and saddle point engineering, and the global resolution of the reconstructed image is equivalent to or even better than FD-DeepLoc algorithm.https://www.worldscientific.com/doi/10.1142/S1793545824500251Real-time data processingfeature compressionpipeline programming
spellingShingle Shuhao Guo
Jiaxun Lin
Yingjun Zhang
Zhen-Li Huang
Enhancing the data processing speed of a deep-learning-based three-dimensional single molecule localization algorithm (FD-DeepLoc) with a combination of feature compression and pipeline programming
Journal of Innovative Optical Health Sciences
Real-time data processing
feature compression
pipeline programming
title Enhancing the data processing speed of a deep-learning-based three-dimensional single molecule localization algorithm (FD-DeepLoc) with a combination of feature compression and pipeline programming
title_full Enhancing the data processing speed of a deep-learning-based three-dimensional single molecule localization algorithm (FD-DeepLoc) with a combination of feature compression and pipeline programming
title_fullStr Enhancing the data processing speed of a deep-learning-based three-dimensional single molecule localization algorithm (FD-DeepLoc) with a combination of feature compression and pipeline programming
title_full_unstemmed Enhancing the data processing speed of a deep-learning-based three-dimensional single molecule localization algorithm (FD-DeepLoc) with a combination of feature compression and pipeline programming
title_short Enhancing the data processing speed of a deep-learning-based three-dimensional single molecule localization algorithm (FD-DeepLoc) with a combination of feature compression and pipeline programming
title_sort enhancing the data processing speed of a deep learning based three dimensional single molecule localization algorithm fd deeploc with a combination of feature compression and pipeline programming
topic Real-time data processing
feature compression
pipeline programming
url https://www.worldscientific.com/doi/10.1142/S1793545824500251
work_keys_str_mv AT shuhaoguo enhancingthedataprocessingspeedofadeeplearningbasedthreedimensionalsinglemoleculelocalizationalgorithmfddeeplocwithacombinationoffeaturecompressionandpipelineprogramming
AT jiaxunlin enhancingthedataprocessingspeedofadeeplearningbasedthreedimensionalsinglemoleculelocalizationalgorithmfddeeplocwithacombinationoffeaturecompressionandpipelineprogramming
AT yingjunzhang enhancingthedataprocessingspeedofadeeplearningbasedthreedimensionalsinglemoleculelocalizationalgorithmfddeeplocwithacombinationoffeaturecompressionandpipelineprogramming
AT zhenlihuang enhancingthedataprocessingspeedofadeeplearningbasedthreedimensionalsinglemoleculelocalizationalgorithmfddeeplocwithacombinationoffeaturecompressionandpipelineprogramming