Early Glaucoma Detection by Using Style Transfer to Predict Retinal Nerve Fiber Layer Thickness Distribution on the Fundus Photograph
Objective: We aimed to develop a deep learning (DL)–based algorithm for early glaucoma detection based on color fundus photographs that provides information on defects in the retinal nerve fiber layer (RNFL) and its thickness from the mapping and translating relations of spectral domain OCT (SD-OCT)...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-09-01
|
| Series: | Ophthalmology Science |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666914522000690 |
| _version_ | 1811206526611750912 |
|---|---|
| author | Henry Shen-Lih Chen, MD, MBA Guan-An Chen, MSc Jhen-Yang Syu, MSc Lan-Hsin Chuang, MD Wei-Wen Su, MD Wei-Chi Wu, MD, PhD Jian-Hong Liu, MSc Jian-Ren Chen, PhD Su-Chen Huang, MSc Eugene Yu-Chuan Kang, MD |
| author_facet | Henry Shen-Lih Chen, MD, MBA Guan-An Chen, MSc Jhen-Yang Syu, MSc Lan-Hsin Chuang, MD Wei-Wen Su, MD Wei-Chi Wu, MD, PhD Jian-Hong Liu, MSc Jian-Ren Chen, PhD Su-Chen Huang, MSc Eugene Yu-Chuan Kang, MD |
| author_sort | Henry Shen-Lih Chen, MD, MBA |
| collection | DOAJ |
| description | Objective: We aimed to develop a deep learning (DL)–based algorithm for early glaucoma detection based on color fundus photographs that provides information on defects in the retinal nerve fiber layer (RNFL) and its thickness from the mapping and translating relations of spectral domain OCT (SD-OCT) thickness maps. Design: Developing and evaluating an artificial intelligence detection tool. Subjects: Pretraining paired data of color fundus photographs and SD-OCT images from 189 healthy participants and 371 patients with early glaucoma were used. Methods: The variational autoencoder (VAE) network training architecture was used for training, and the correlation between the fundus photographs and RNFL thickness distribution was determined through the deep neural network. The reference standard was defined as a vertical cup-to-disc ratio of ≥0.7, other typical changes in glaucomatous optic neuropathy, and RNFL defects. Convergence indicates that the VAE has learned a distribution that would enable us to produce corresponding synthetic OCT scans. Main Outcome Measures: Similarly to wide-field OCT scanning, the proposed model can extract the results of RNFL thickness analysis. The structural similarity index measure (SSIM) and peak signal-to-noise ratio (PSNR) were used to assess signal strength and the similarity in the structure of the color fundus images converted to an RNFL thickness distribution model. The differences between the model-generated images and original images were quantified. Results: We developed and validated a novel DL-based algorithm to extract thickness information from the color space of fundus images similarly to that from OCT images and to use this information to regenerate RNFL thickness distribution images. The generated thickness map was sufficient for clinical glaucoma detection, and the generated images were similar to ground truth (PSNR: 19.31 decibels; SSIM: 0.44). The inference results were similar to the OCT-generated original images in terms of the ability to predict RNFL thickness distribution. Conclusions: The proposed technique may aid clinicians in early glaucoma detection, especially when only color fundus photographs are available. |
| first_indexed | 2024-04-12T03:49:51Z |
| format | Article |
| id | doaj.art-769208e35a8a4170a56d1f4cef5b2dd6 |
| institution | Directory Open Access Journal |
| issn | 2666-9145 |
| language | English |
| last_indexed | 2024-04-12T03:49:51Z |
| publishDate | 2022-09-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Ophthalmology Science |
| spelling | doaj.art-769208e35a8a4170a56d1f4cef5b2dd62022-12-22T03:49:02ZengElsevierOphthalmology Science2666-91452022-09-0123100180Early Glaucoma Detection by Using Style Transfer to Predict Retinal Nerve Fiber Layer Thickness Distribution on the Fundus PhotographHenry Shen-Lih Chen, MD, MBA0Guan-An Chen, MSc1Jhen-Yang Syu, MSc2Lan-Hsin Chuang, MD3Wei-Wen Su, MD4Wei-Chi Wu, MD, PhD5Jian-Hong Liu, MSc6Jian-Ren Chen, PhD7Su-Chen Huang, MSc8Eugene Yu-Chuan Kang, MD9Department of Ophthalmology, Chang Gung Memorial Hospital, Linkou Medical Center, Taoyuan, Taiwan; College of Medicine, Chang Gung University, Taoyuan, Taiwan; Henry Shen-Lih Chen, MD, MBA, Department of Ophthalmology, Chang Gung Memorial Hospital, No. 5, Fu-Hsin Rd., Taoyuan 333, Taiwan.Healthcare Service Division, Department of Intelligent Medical & Healthcare, Service Systems Technology Center, Industrial Technology Research Institute, Hsinchu, TaiwanHealthcare Service Division, Department of Intelligent Medical & Healthcare, Service Systems Technology Center, Industrial Technology Research Institute, Hsinchu, TaiwanCollege of Medicine, Chang Gung University, Taoyuan, Taiwan; Department of Ophthalmology, Keelung Chang Gung Memorial Hospital, Keelung, TaiwanDepartment of Ophthalmology, Chang Gung Memorial Hospital, Linkou Medical Center, Taoyuan, Taiwan; College of Medicine, Chang Gung University, Taoyuan, TaiwanDepartment of Ophthalmology, Chang Gung Memorial Hospital, Linkou Medical Center, Taoyuan, Taiwan; College of Medicine, Chang Gung University, Taoyuan, TaiwanHealthcare Service Division, Department of Intelligent Medical & Healthcare, Service Systems Technology Center, Industrial Technology Research Institute, Hsinchu, TaiwanHealthcare Service Division, Department of Intelligent Medical & Healthcare, Service Systems Technology Center, Industrial Technology Research Institute, Hsinchu, TaiwanHealthcare Service Division, Department of Intelligent Medical & Healthcare, Service Systems Technology Center, Industrial Technology Research Institute, Hsinchu, TaiwanDepartment of Ophthalmology, Chang Gung Memorial Hospital, Linkou Medical Center, Taoyuan, Taiwan; College of Medicine, Chang Gung University, Taoyuan, Taiwan; Graduate Institute of Clinical Medical Sciences, College of Medicine, Chang Gung University, Taoyuan, Taiwan; Correspondence: Eugene Yu-Chuan Kang, MD, Department of Ophthalmology, Chang Gung Memorial Hospital, No. 5, Fu-Hsin Rd., Taoyuan 333, Taiwan.Objective: We aimed to develop a deep learning (DL)–based algorithm for early glaucoma detection based on color fundus photographs that provides information on defects in the retinal nerve fiber layer (RNFL) and its thickness from the mapping and translating relations of spectral domain OCT (SD-OCT) thickness maps. Design: Developing and evaluating an artificial intelligence detection tool. Subjects: Pretraining paired data of color fundus photographs and SD-OCT images from 189 healthy participants and 371 patients with early glaucoma were used. Methods: The variational autoencoder (VAE) network training architecture was used for training, and the correlation between the fundus photographs and RNFL thickness distribution was determined through the deep neural network. The reference standard was defined as a vertical cup-to-disc ratio of ≥0.7, other typical changes in glaucomatous optic neuropathy, and RNFL defects. Convergence indicates that the VAE has learned a distribution that would enable us to produce corresponding synthetic OCT scans. Main Outcome Measures: Similarly to wide-field OCT scanning, the proposed model can extract the results of RNFL thickness analysis. The structural similarity index measure (SSIM) and peak signal-to-noise ratio (PSNR) were used to assess signal strength and the similarity in the structure of the color fundus images converted to an RNFL thickness distribution model. The differences between the model-generated images and original images were quantified. Results: We developed and validated a novel DL-based algorithm to extract thickness information from the color space of fundus images similarly to that from OCT images and to use this information to regenerate RNFL thickness distribution images. The generated thickness map was sufficient for clinical glaucoma detection, and the generated images were similar to ground truth (PSNR: 19.31 decibels; SSIM: 0.44). The inference results were similar to the OCT-generated original images in terms of the ability to predict RNFL thickness distribution. Conclusions: The proposed technique may aid clinicians in early glaucoma detection, especially when only color fundus photographs are available.http://www.sciencedirect.com/science/article/pii/S2666914522000690Image-to-image translationGlaucomaAutoencoderOCTRetinal nerve fiber layer thickness |
| spellingShingle | Henry Shen-Lih Chen, MD, MBA Guan-An Chen, MSc Jhen-Yang Syu, MSc Lan-Hsin Chuang, MD Wei-Wen Su, MD Wei-Chi Wu, MD, PhD Jian-Hong Liu, MSc Jian-Ren Chen, PhD Su-Chen Huang, MSc Eugene Yu-Chuan Kang, MD Early Glaucoma Detection by Using Style Transfer to Predict Retinal Nerve Fiber Layer Thickness Distribution on the Fundus Photograph Ophthalmology Science Image-to-image translation Glaucoma Autoencoder OCT Retinal nerve fiber layer thickness |
| title | Early Glaucoma Detection by Using Style Transfer to Predict Retinal Nerve Fiber Layer Thickness Distribution on the Fundus Photograph |
| title_full | Early Glaucoma Detection by Using Style Transfer to Predict Retinal Nerve Fiber Layer Thickness Distribution on the Fundus Photograph |
| title_fullStr | Early Glaucoma Detection by Using Style Transfer to Predict Retinal Nerve Fiber Layer Thickness Distribution on the Fundus Photograph |
| title_full_unstemmed | Early Glaucoma Detection by Using Style Transfer to Predict Retinal Nerve Fiber Layer Thickness Distribution on the Fundus Photograph |
| title_short | Early Glaucoma Detection by Using Style Transfer to Predict Retinal Nerve Fiber Layer Thickness Distribution on the Fundus Photograph |
| title_sort | early glaucoma detection by using style transfer to predict retinal nerve fiber layer thickness distribution on the fundus photograph |
| topic | Image-to-image translation Glaucoma Autoencoder OCT Retinal nerve fiber layer thickness |
| url | http://www.sciencedirect.com/science/article/pii/S2666914522000690 |
| work_keys_str_mv | AT henryshenlihchenmdmba earlyglaucomadetectionbyusingstyletransfertopredictretinalnervefiberlayerthicknessdistributiononthefundusphotograph AT guananchenmsc earlyglaucomadetectionbyusingstyletransfertopredictretinalnervefiberlayerthicknessdistributiononthefundusphotograph AT jhenyangsyumsc earlyglaucomadetectionbyusingstyletransfertopredictretinalnervefiberlayerthicknessdistributiononthefundusphotograph AT lanhsinchuangmd earlyglaucomadetectionbyusingstyletransfertopredictretinalnervefiberlayerthicknessdistributiononthefundusphotograph AT weiwensumd earlyglaucomadetectionbyusingstyletransfertopredictretinalnervefiberlayerthicknessdistributiononthefundusphotograph AT weichiwumdphd earlyglaucomadetectionbyusingstyletransfertopredictretinalnervefiberlayerthicknessdistributiononthefundusphotograph AT jianhongliumsc earlyglaucomadetectionbyusingstyletransfertopredictretinalnervefiberlayerthicknessdistributiononthefundusphotograph AT jianrenchenphd earlyglaucomadetectionbyusingstyletransfertopredictretinalnervefiberlayerthicknessdistributiononthefundusphotograph AT suchenhuangmsc earlyglaucomadetectionbyusingstyletransfertopredictretinalnervefiberlayerthicknessdistributiononthefundusphotograph AT eugeneyuchuankangmd earlyglaucomadetectionbyusingstyletransfertopredictretinalnervefiberlayerthicknessdistributiononthefundusphotograph |