Deep Neural Network-Based Semantic Segmentation of Microvascular Decompression Images
Image semantic segmentation has been applied more and more widely in the fields of satellite remote sensing, medical treatment, intelligent transportation, and virtual reality. However, in the medical field, the study of cerebral vessel and cranial nerve segmentation based on true-color medical imag...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-02-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/21/4/1167 |
id |
doaj-7146b08f7a6b430da65e1f1babe568a5 |
---|---|
record_format |
Article |
spelling |
doaj-7146b08f7a6b430da65e1f1babe568a52021-02-08T00:02:07ZengMDPI AGSensors1424-82202021-02-01211167116710.3390/s21041167Deep Neural Network-Based Semantic Segmentation of Microvascular Decompression ImagesRuifeng Bai0Shan Jiang1Haijiang Sun2Yifan Yang3Guiju Li4Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, ChinaChangchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, ChinaChangchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, ChinaChangchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, ChinaChangchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, ChinaImage semantic segmentation has been applied more and more widely in the fields of satellite remote sensing, medical treatment, intelligent transportation, and virtual reality. However, in the medical field, the study of cerebral vessel and cranial nerve segmentation based on true-color medical images is in urgent need and has good research and development prospects. We have extended the current state-of-the-art semantic-segmentation network DeepLabv3+ and used it as the basic framework. First, the feature distillation block (FDB) was introduced into the encoder structure to refine the extracted features. In addition, the atrous spatial pyramid pooling (ASPP) module was added to the decoder structure to enhance the retention of feature and boundary information. The proposed model was trained by fine tuning and optimizing the relevant parameters. Experimental results show that the encoder structure has better performance in feature refinement processing, improving target boundary segmentation precision, and retaining more feature information. Our method has a segmentation accuracy of 75.73%, which is 3% better than DeepLabv3+.https://www.mdpi.com/1424-8220/21/4/1167microvascular decompression imagesemantic segmentationDeepLabv3+encoder structuredecoder structure |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Ruifeng Bai Shan Jiang Haijiang Sun Yifan Yang Guiju Li |
spellingShingle |
Ruifeng Bai Shan Jiang Haijiang Sun Yifan Yang Guiju Li Deep Neural Network-Based Semantic Segmentation of Microvascular Decompression Images Sensors microvascular decompression image semantic segmentation DeepLabv3+ encoder structure decoder structure |
author_facet |
Ruifeng Bai Shan Jiang Haijiang Sun Yifan Yang Guiju Li |
author_sort |
Ruifeng Bai |
title |
Deep Neural Network-Based Semantic Segmentation of Microvascular Decompression Images |
title_short |
Deep Neural Network-Based Semantic Segmentation of Microvascular Decompression Images |
title_full |
Deep Neural Network-Based Semantic Segmentation of Microvascular Decompression Images |
title_fullStr |
Deep Neural Network-Based Semantic Segmentation of Microvascular Decompression Images |
title_full_unstemmed |
Deep Neural Network-Based Semantic Segmentation of Microvascular Decompression Images |
title_sort |
deep neural network-based semantic segmentation of microvascular decompression images |
publisher |
MDPI AG |
series |
Sensors |
issn |
1424-8220 |
publishDate |
2021-02-01 |
description |
Image semantic segmentation has been applied more and more widely in the fields of satellite remote sensing, medical treatment, intelligent transportation, and virtual reality. However, in the medical field, the study of cerebral vessel and cranial nerve segmentation based on true-color medical images is in urgent need and has good research and development prospects. We have extended the current state-of-the-art semantic-segmentation network DeepLabv3+ and used it as the basic framework. First, the feature distillation block (FDB) was introduced into the encoder structure to refine the extracted features. In addition, the atrous spatial pyramid pooling (ASPP) module was added to the decoder structure to enhance the retention of feature and boundary information. The proposed model was trained by fine tuning and optimizing the relevant parameters. Experimental results show that the encoder structure has better performance in feature refinement processing, improving target boundary segmentation precision, and retaining more feature information. Our method has a segmentation accuracy of 75.73%, which is 3% better than DeepLabv3+. |
topic |
microvascular decompression image semantic segmentation DeepLabv3+ encoder structure decoder structure |
url |
https://www.mdpi.com/1424-8220/21/4/1167 |
work_keys_str_mv |
AT ruifengbai deepneuralnetworkbasedsemanticsegmentationofmicrovasculardecompressionimages AT shanjiang deepneuralnetworkbasedsemanticsegmentationofmicrovasculardecompressionimages AT haijiangsun deepneuralnetworkbasedsemanticsegmentationofmicrovasculardecompressionimages AT yifanyang deepneuralnetworkbasedsemanticsegmentationofmicrovasculardecompressionimages AT guijuli deepneuralnetworkbasedsemanticsegmentationofmicrovasculardecompressionimages |
_version_ |
1724280408829329408 |