A 3D semantic segmentation network for accurate neuronal soma segmentation
Neuronal soma segmentation plays a crucial role in neuroscience applications. However, the fine structure, such as boundaries, small-volume neuronal somata and fibers, are commonly present in cell images, which pose a challenge for accurate segmentation. In this paper, we propose a 3D semantic segme...
Saved in:
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
World Scientific Publishing
2025-01-01
|
Series: | Journal of Innovative Optical Health Sciences |
Subjects: | |
Online Access: | https://www.worldscientific.com/doi/10.1142/S1793545824500184 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
_version_ | 1832585091278176256 |
---|---|
author | Li Ma Qi Zhong Yezi Wang Xiaoquan Yang Qian Du |
author_facet | Li Ma Qi Zhong Yezi Wang Xiaoquan Yang Qian Du |
author_sort | Li Ma |
collection | DOAJ |
description | Neuronal soma segmentation plays a crucial role in neuroscience applications. However, the fine structure, such as boundaries, small-volume neuronal somata and fibers, are commonly present in cell images, which pose a challenge for accurate segmentation. In this paper, we propose a 3D semantic segmentation network for neuronal soma segmentation to address this issue. Using an encoding-decoding structure, we introduce a Multi-Scale feature extraction and Adaptive Weighting fusion module (MSAW) after each encoding block. The MSAW module can not only emphasize the fine structures via an upsampling strategy, but also provide pixel-wise weights to measure the importance of the multi-scale features. Additionally, a dynamic convolution instead of normal convolution is employed to better adapt the network to input data with different distributions. The proposed MSAW-based semantic segmentation network (MSAW-Net) was evaluated on three neuronal soma images from mouse brain and one neuronal soma image from macaque brain, demonstrating the efficiency of the proposed method. It achieved an F1 score of 91.8% on Fezf2-2A-CreER dataset, 97.1% on LSL-H2B-GFP dataset, 82.8% on Thy1-EGFP-Mline dataset, and 86.9% on macaque dataset, achieving improvements over the 3D U-Net model by 3.1%, 3.3%, 3.9%, and 2.3%, respectively. |
format | Article |
id | doaj-art-b2bb7f6acbf54459b56830e37aee7515 |
institution | Kabale University |
issn | 1793-5458 1793-7205 |
language | English |
publishDate | 2025-01-01 |
publisher | World Scientific Publishing |
record_format | Article |
series | Journal of Innovative Optical Health Sciences |
spelling | doaj-art-b2bb7f6acbf54459b56830e37aee75152025-01-27T05:49:53ZengWorld Scientific PublishingJournal of Innovative Optical Health Sciences1793-54581793-72052025-01-01180110.1142/S1793545824500184A 3D semantic segmentation network for accurate neuronal soma segmentationLi Ma0Qi Zhong1Yezi Wang2Xiaoquan Yang3Qian Du4School of Mechanical Engineering and Electronic Information, China University of Geosciences, Wuhan 430074, P. R. ChinaSchool of Mechanical Engineering and Electronic Information, China University of Geosciences, Wuhan 430074, P. R. ChinaSchool of Mechanical Engineering and Electronic Information, China University of Geosciences, Wuhan 430074, P. R. ChinaBritton Chance Center for Biomedical Photonics, Wuhan National Laboratory for Optoelectronics, MoE Key Laboratory for Biomedical Photonics, Huazhong University of Science and Technology, Wuhan 430074, P. R. ChinaDepartment of Electrical and Computer Engineering, Mississippi State University, Mississippi State, MS 39762, USANeuronal soma segmentation plays a crucial role in neuroscience applications. However, the fine structure, such as boundaries, small-volume neuronal somata and fibers, are commonly present in cell images, which pose a challenge for accurate segmentation. In this paper, we propose a 3D semantic segmentation network for neuronal soma segmentation to address this issue. Using an encoding-decoding structure, we introduce a Multi-Scale feature extraction and Adaptive Weighting fusion module (MSAW) after each encoding block. The MSAW module can not only emphasize the fine structures via an upsampling strategy, but also provide pixel-wise weights to measure the importance of the multi-scale features. Additionally, a dynamic convolution instead of normal convolution is employed to better adapt the network to input data with different distributions. The proposed MSAW-based semantic segmentation network (MSAW-Net) was evaluated on three neuronal soma images from mouse brain and one neuronal soma image from macaque brain, demonstrating the efficiency of the proposed method. It achieved an F1 score of 91.8% on Fezf2-2A-CreER dataset, 97.1% on LSL-H2B-GFP dataset, 82.8% on Thy1-EGFP-Mline dataset, and 86.9% on macaque dataset, achieving improvements over the 3D U-Net model by 3.1%, 3.3%, 3.9%, and 2.3%, respectively.https://www.worldscientific.com/doi/10.1142/S1793545824500184Neuronal soma segmentationsemantic segmentation networkmulti-scale feature extractionadaptive weighting fusion |
spellingShingle | Li Ma Qi Zhong Yezi Wang Xiaoquan Yang Qian Du A 3D semantic segmentation network for accurate neuronal soma segmentation Journal of Innovative Optical Health Sciences Neuronal soma segmentation semantic segmentation network multi-scale feature extraction adaptive weighting fusion |
title | A 3D semantic segmentation network for accurate neuronal soma segmentation |
title_full | A 3D semantic segmentation network for accurate neuronal soma segmentation |
title_fullStr | A 3D semantic segmentation network for accurate neuronal soma segmentation |
title_full_unstemmed | A 3D semantic segmentation network for accurate neuronal soma segmentation |
title_short | A 3D semantic segmentation network for accurate neuronal soma segmentation |
title_sort | 3d semantic segmentation network for accurate neuronal soma segmentation |
topic | Neuronal soma segmentation semantic segmentation network multi-scale feature extraction adaptive weighting fusion |
url | https://www.worldscientific.com/doi/10.1142/S1793545824500184 |
work_keys_str_mv | AT lima a3dsemanticsegmentationnetworkforaccurateneuronalsomasegmentation AT qizhong a3dsemanticsegmentationnetworkforaccurateneuronalsomasegmentation AT yeziwang a3dsemanticsegmentationnetworkforaccurateneuronalsomasegmentation AT xiaoquanyang a3dsemanticsegmentationnetworkforaccurateneuronalsomasegmentation AT qiandu a3dsemanticsegmentationnetworkforaccurateneuronalsomasegmentation AT lima 3dsemanticsegmentationnetworkforaccurateneuronalsomasegmentation AT qizhong 3dsemanticsegmentationnetworkforaccurateneuronalsomasegmentation AT yeziwang 3dsemanticsegmentationnetworkforaccurateneuronalsomasegmentation AT xiaoquanyang 3dsemanticsegmentationnetworkforaccurateneuronalsomasegmentation AT qiandu 3dsemanticsegmentationnetworkforaccurateneuronalsomasegmentation |