Graph Attention U-Net for Retinal Layer Surface Detection and Choroid Neovascularization Segmentation in OCT Images
Choroidal neovascularization (CNV) is a typical symptom of age-related macular degeneration (AMD) and is one of the leading causes for blindness. Accurate segmentation of CNV and detection of retinal layers are critical for eye disease diagnosis and monitoring. In this paper, we propose a novel graph attention U-Net (GA-UNet) for retinal layer surface detection and CNV segmentation in optical coherence tomography (OCT) images. Due to retinal layer deformation caused by CNV, it is challenging for existing models to segment CNV and detect retinal layer surfaces with the correct topological order. We propose two novel modules to address the challenge. The first module is a graph attention encoder (GAE) in a U-Net model that automatically integrates topological and pathological knowledge of retinal layers into the U-Net structure to achieve effective feature embedding. The second module is a graph decorrelation module (GDM) that takes reconstructed features by the decoder of the U-Net as inputs, it then decorrelates and removes information unrelated to retinal layer for improved retinal layer surface detection. In addition, we propose a new loss function to maintain the correct topological order of retinal layers and the continuity of their boundaries. The proposed model learns graph attention maps automatically during training and performs retinal layer surface detection and CNV segmentation simultaneously with the attention maps during inference. We evaluated the proposed model on our private AMD dataset and another public dataset. Experiment results show that the proposed model outperformed the competing methods for retinal layer surface detection and CNV segmentation and achieved new state of the arts on the datasets.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2023 |
|---|---|
| Erschienen: |
2023 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:42 |
|---|---|
| Enthalten in: |
IEEE transactions on medical imaging - 42(2023), 11 vom: 26. Nov., Seite 3140-3154 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Shen, Yuhe [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
|---|
| Anmerkungen: |
Date Completed 30.10.2023 Date Revised 12.11.2023 published: Print-Electronic Citation Status MEDLINE |
|---|
| doi: |
10.1109/TMI.2023.3240757 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM355269619 |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM355269619 | ||
| 003 | DE-627 | ||
| 005 | 20231226064008.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231226s2023 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1109/TMI.2023.3240757 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1184.xml |
| 035 | |a (DE-627)NLM355269619 | ||
| 035 | |a (NLM)37022267 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Shen, Yuhe |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Graph Attention U-Net for Retinal Layer Surface Detection and Choroid Neovascularization Segmentation in OCT Images |
| 264 | 1 | |c 2023 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Completed 30.10.2023 | ||
| 500 | |a Date Revised 12.11.2023 | ||
| 500 | |a published: Print-Electronic | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a Choroidal neovascularization (CNV) is a typical symptom of age-related macular degeneration (AMD) and is one of the leading causes for blindness. Accurate segmentation of CNV and detection of retinal layers are critical for eye disease diagnosis and monitoring. In this paper, we propose a novel graph attention U-Net (GA-UNet) for retinal layer surface detection and CNV segmentation in optical coherence tomography (OCT) images. Due to retinal layer deformation caused by CNV, it is challenging for existing models to segment CNV and detect retinal layer surfaces with the correct topological order. We propose two novel modules to address the challenge. The first module is a graph attention encoder (GAE) in a U-Net model that automatically integrates topological and pathological knowledge of retinal layers into the U-Net structure to achieve effective feature embedding. The second module is a graph decorrelation module (GDM) that takes reconstructed features by the decoder of the U-Net as inputs, it then decorrelates and removes information unrelated to retinal layer for improved retinal layer surface detection. In addition, we propose a new loss function to maintain the correct topological order of retinal layers and the continuity of their boundaries. The proposed model learns graph attention maps automatically during training and performs retinal layer surface detection and CNV segmentation simultaneously with the attention maps during inference. We evaluated the proposed model on our private AMD dataset and another public dataset. Experiment results show that the proposed model outperformed the competing methods for retinal layer surface detection and CNV segmentation and achieved new state of the arts on the datasets | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a Research Support, Non-U.S. Gov't | |
| 700 | 1 | |a Li, Jiang |e verfasserin |4 aut | |
| 700 | 1 | |a Zhu, Weifang |e verfasserin |4 aut | |
| 700 | 1 | |a Yu, Kai |e verfasserin |4 aut | |
| 700 | 1 | |a Wang, Meng |e verfasserin |4 aut | |
| 700 | 1 | |a Peng, Yuanyuan |e verfasserin |4 aut | |
| 700 | 1 | |a Zhou, Yi |e verfasserin |4 aut | |
| 700 | 1 | |a Guan, Liling |e verfasserin |4 aut | |
| 700 | 1 | |a Chen, Xinjian |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t IEEE transactions on medical imaging |d 1982 |g 42(2023), 11 vom: 26. Nov., Seite 3140-3154 |w (DE-627)NLM082855269 |x 1558-254X |7 nnns |
| 773 | 1 | 8 | |g volume:42 |g year:2023 |g number:11 |g day:26 |g month:11 |g pages:3140-3154 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1109/TMI.2023.3240757 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 42 |j 2023 |e 11 |b 26 |c 11 |h 3140-3154 | ||