SUD-GAN : Deep Convolution Generative Adversarial Network Combined with Short Connection and Dense Block for Retinal Vessel Segmentation
Since morphology of retinal blood vessels plays a key role in ophthalmological disease diagnosis, retinal vessel segmentation is an indispensable step for the screening and diagnosis of retinal diseases with fundus images. In this paper, deep convolution adversarial network combined with short connection and dense block is proposed to separate blood vessels from fundus image, named SUD-GAN. The generator adopts U-shape encode-decode structure and adds short connection block between convolution layers to prevent gradient dispersion caused by deep convolution network. The discriminator is all composed of convolution block, and dense connection structure is added to the middle part of the convolution network to strengthen the spread of features and enhance the network discrimination ability. The proposed method is evaluated on two publicly available databases, the DRIVE and STARE. The results show that the proposed method outperforms the state-of-the-art performance in sensitivity and specificity, which were 0.8340 and 0.9820, and 0.8334 and 0.9897 respectively on DRIVE and STARE, and can detect more tiny vessels and locate the edge of blood vessels more accurately.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2020 |
|---|---|
| Erschienen: |
2020 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:33 |
|---|---|
| Enthalten in: |
Journal of digital imaging - 33(2020), 4 vom: 22. Aug., Seite 946-957 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Yang, Tiejun [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
Dense block |
|---|
| Anmerkungen: |
Date Completed 16.08.2021 Date Revised 12.11.2023 published: Print Citation Status MEDLINE |
|---|
| doi: |
10.1007/s10278-020-00339-9 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM309074746 |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM309074746 | ||
| 003 | DE-627 | ||
| 005 | 20231225133119.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231225s2020 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1007/s10278-020-00339-9 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1030.xml |
| 035 | |a (DE-627)NLM309074746 | ||
| 035 | |a (NLM)32323089 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Yang, Tiejun |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a SUD-GAN |b Deep Convolution Generative Adversarial Network Combined with Short Connection and Dense Block for Retinal Vessel Segmentation |
| 264 | 1 | |c 2020 | |
| 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 16.08.2021 | ||
| 500 | |a Date Revised 12.11.2023 | ||
| 500 | |a published: Print | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a Since morphology of retinal blood vessels plays a key role in ophthalmological disease diagnosis, retinal vessel segmentation is an indispensable step for the screening and diagnosis of retinal diseases with fundus images. In this paper, deep convolution adversarial network combined with short connection and dense block is proposed to separate blood vessels from fundus image, named SUD-GAN. The generator adopts U-shape encode-decode structure and adds short connection block between convolution layers to prevent gradient dispersion caused by deep convolution network. The discriminator is all composed of convolution block, and dense connection structure is added to the middle part of the convolution network to strengthen the spread of features and enhance the network discrimination ability. The proposed method is evaluated on two publicly available databases, the DRIVE and STARE. The results show that the proposed method outperforms the state-of-the-art performance in sensitivity and specificity, which were 0.8340 and 0.9820, and 0.8334 and 0.9897 respectively on DRIVE and STARE, and can detect more tiny vessels and locate the edge of blood vessels more accurately | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a Research Support, Non-U.S. Gov't | |
| 650 | 4 | |a Dense block | |
| 650 | 4 | |a Generative adversarial network | |
| 650 | 4 | |a Retinal vessel segmentation | |
| 650 | 4 | |a Short connection block | |
| 700 | 1 | |a Wu, Tingting |e verfasserin |4 aut | |
| 700 | 1 | |a Li, Lei |e verfasserin |4 aut | |
| 700 | 1 | |a Zhu, Chunhua |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Journal of digital imaging |d 1988 |g 33(2020), 4 vom: 22. Aug., Seite 946-957 |w (DE-627)NLM014390140 |x 1618-727X |7 nnns |
| 773 | 1 | 8 | |g volume:33 |g year:2020 |g number:4 |g day:22 |g month:08 |g pages:946-957 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1007/s10278-020-00339-9 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 33 |j 2020 |e 4 |b 22 |c 08 |h 946-957 | ||