Deep Learning for Motion Artifact-Suppressed OCTA Image Generation from Both Repeated and Adjacent OCT Scans
Optical coherence tomography angiography (OCTA) is a popular technique for imaging microvascular networks, but OCTA image quality is commonly affected by motion artifacts. Deep learning (DL) has been used to generate OCTA images from structural OCT images, yet limitations persist, such as low label...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-01-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/12/3/446 |
| _version_ | 1827354743936122880 |
|---|---|
| author | Zhefan Lin Qinqin Zhang Gongpu Lan Jingjiang Xu Jia Qin Lin An Yanping Huang |
| author_facet | Zhefan Lin Qinqin Zhang Gongpu Lan Jingjiang Xu Jia Qin Lin An Yanping Huang |
| author_sort | Zhefan Lin |
| collection | DOAJ |
| description | Optical coherence tomography angiography (OCTA) is a popular technique for imaging microvascular networks, but OCTA image quality is commonly affected by motion artifacts. Deep learning (DL) has been used to generate OCTA images from structural OCT images, yet limitations persist, such as low label image quality caused by motion and insufficient use of neighborhood information. In this study, an attention-based U-Net incorporating both repeated and adjacent structural OCT images in network input and high-quality label OCTA images in training was proposed to generate high-quality OCTA images with motion artifact suppression. A sliding-window correlation-based adjacent position (SWCB-AP) image fusion method was proposed to generate high-quality OCTA label images with suppressed motion noise. Six different DL schemes with various configurations of network inputs and label images were compared to demonstrate the superiority of the proposed method. Motion artifact severity was evaluated by a motion noise index in B-scan (MNI-B) and in en-face (MNI-C) OCTA images, which were specifically defined in this study for the purpose of evaluating various DL models’ capability in motion noise suppression. Experimental results on a nailfold OCTA image dataset showed that the proposed DL method generated the best results with a peak signal-to-noise ratio (PSNR) of 32.666 ± 7.010 dB, structural similarity (SSIM) of 0.926 ± 0.051, mean absolute error (MAE) of 1.798 ± 1.575, and MNI-B of 0.528 ± 0.124 in B-scan OCTA images and a contrast-to-noise ratio (CNR) of 1.420 ± 0.291 and MNI-C of 0.156 ± 0.057 in en-face OCTA images. Our proposed DL approach generated OCTA images with improved blood flow contrast and reduced motion artifacts, which could be used as a fundamental signal processing module in generating high-quality OCTA images from structural OCT images. |
| first_indexed | 2024-03-08T03:53:06Z |
| format | Article |
| id | doaj.art-d6a68559c03f429aad2365fa8125b5ac |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-08T03:53:06Z |
| publishDate | 2024-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-d6a68559c03f429aad2365fa8125b5ac2024-02-09T15:18:22ZengMDPI AGMathematics2227-73902024-01-0112344610.3390/math12030446Deep Learning for Motion Artifact-Suppressed OCTA Image Generation from Both Repeated and Adjacent OCT ScansZhefan Lin0Qinqin Zhang1Gongpu Lan2Jingjiang Xu3Jia Qin4Lin An5Yanping Huang6Guangdong-Hong Kong-Macao Joint Laboratory for Intelligent Micro-Nano Optoelectronic Technology, School of Physics and Optoelectronic Engineering, Foshan University, Foshan 528000, ChinaGuangdong-Hong Kong-Macao Joint Laboratory for Intelligent Micro-Nano Optoelectronic Technology, School of Physics and Optoelectronic Engineering, Foshan University, Foshan 528000, ChinaGuangdong-Hong Kong-Macao Joint Laboratory for Intelligent Micro-Nano Optoelectronic Technology, School of Physics and Optoelectronic Engineering, Foshan University, Foshan 528000, ChinaGuangdong-Hong Kong-Macao Joint Laboratory for Intelligent Micro-Nano Optoelectronic Technology, School of Physics and Optoelectronic Engineering, Foshan University, Foshan 528000, ChinaInnovation and Entrepreneurship Teams Project of Guangdong Provincial Pearl River Talents Program, Guangdong Weiren Meditech Co., Ltd., Foshan 528015, ChinaInnovation and Entrepreneurship Teams Project of Guangdong Provincial Pearl River Talents Program, Guangdong Weiren Meditech Co., Ltd., Foshan 528015, ChinaGuangdong-Hong Kong-Macao Joint Laboratory for Intelligent Micro-Nano Optoelectronic Technology, School of Physics and Optoelectronic Engineering, Foshan University, Foshan 528000, ChinaOptical coherence tomography angiography (OCTA) is a popular technique for imaging microvascular networks, but OCTA image quality is commonly affected by motion artifacts. Deep learning (DL) has been used to generate OCTA images from structural OCT images, yet limitations persist, such as low label image quality caused by motion and insufficient use of neighborhood information. In this study, an attention-based U-Net incorporating both repeated and adjacent structural OCT images in network input and high-quality label OCTA images in training was proposed to generate high-quality OCTA images with motion artifact suppression. A sliding-window correlation-based adjacent position (SWCB-AP) image fusion method was proposed to generate high-quality OCTA label images with suppressed motion noise. Six different DL schemes with various configurations of network inputs and label images were compared to demonstrate the superiority of the proposed method. Motion artifact severity was evaluated by a motion noise index in B-scan (MNI-B) and in en-face (MNI-C) OCTA images, which were specifically defined in this study for the purpose of evaluating various DL models’ capability in motion noise suppression. Experimental results on a nailfold OCTA image dataset showed that the proposed DL method generated the best results with a peak signal-to-noise ratio (PSNR) of 32.666 ± 7.010 dB, structural similarity (SSIM) of 0.926 ± 0.051, mean absolute error (MAE) of 1.798 ± 1.575, and MNI-B of 0.528 ± 0.124 in B-scan OCTA images and a contrast-to-noise ratio (CNR) of 1.420 ± 0.291 and MNI-C of 0.156 ± 0.057 in en-face OCTA images. Our proposed DL approach generated OCTA images with improved blood flow contrast and reduced motion artifacts, which could be used as a fundamental signal processing module in generating high-quality OCTA images from structural OCT images.https://www.mdpi.com/2227-7390/12/3/446optical coherence tomography angiographydeep learningimage fusionimage generationneighborhood informationtraining scheme |
| spellingShingle | Zhefan Lin Qinqin Zhang Gongpu Lan Jingjiang Xu Jia Qin Lin An Yanping Huang Deep Learning for Motion Artifact-Suppressed OCTA Image Generation from Both Repeated and Adjacent OCT Scans Mathematics optical coherence tomography angiography deep learning image fusion image generation neighborhood information training scheme |
| title | Deep Learning for Motion Artifact-Suppressed OCTA Image Generation from Both Repeated and Adjacent OCT Scans |
| title_full | Deep Learning for Motion Artifact-Suppressed OCTA Image Generation from Both Repeated and Adjacent OCT Scans |
| title_fullStr | Deep Learning for Motion Artifact-Suppressed OCTA Image Generation from Both Repeated and Adjacent OCT Scans |
| title_full_unstemmed | Deep Learning for Motion Artifact-Suppressed OCTA Image Generation from Both Repeated and Adjacent OCT Scans |
| title_short | Deep Learning for Motion Artifact-Suppressed OCTA Image Generation from Both Repeated and Adjacent OCT Scans |
| title_sort | deep learning for motion artifact suppressed octa image generation from both repeated and adjacent oct scans |
| topic | optical coherence tomography angiography deep learning image fusion image generation neighborhood information training scheme |
| url | https://www.mdpi.com/2227-7390/12/3/446 |
| work_keys_str_mv | AT zhefanlin deeplearningformotionartifactsuppressedoctaimagegenerationfrombothrepeatedandadjacentoctscans AT qinqinzhang deeplearningformotionartifactsuppressedoctaimagegenerationfrombothrepeatedandadjacentoctscans AT gongpulan deeplearningformotionartifactsuppressedoctaimagegenerationfrombothrepeatedandadjacentoctscans AT jingjiangxu deeplearningformotionartifactsuppressedoctaimagegenerationfrombothrepeatedandadjacentoctscans AT jiaqin deeplearningformotionartifactsuppressedoctaimagegenerationfrombothrepeatedandadjacentoctscans AT linan deeplearningformotionartifactsuppressedoctaimagegenerationfrombothrepeatedandadjacentoctscans AT yanpinghuang deeplearningformotionartifactsuppressedoctaimagegenerationfrombothrepeatedandadjacentoctscans |