1
|
Li X, Wen X, Shang X, Liu J, Zhang L, Cui Y, Luo X, Zhang G, Xie J, Huang T, Chen Z, Lyu Z, Wu X, Lan Y, Meng Q. Identification of diabetic retinopathy classification using machine learning algorithms on clinical data and optical coherence tomography angiography. Eye (Lond) 2024:10.1038/s41433-024-03173-3. [PMID: 38871934 DOI: 10.1038/s41433-024-03173-3] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/02/2023] [Revised: 04/10/2024] [Accepted: 06/06/2024] [Indexed: 06/15/2024] Open
Abstract
BACKGROUND To apply machine learning (ML) algorithms to perform multiclass diabetic retinopathy (DR) classification using both clinical data and optical coherence tomography angiography (OCTA). METHODS In this cross-sectional observational study, clinical data and OCTA parameters from 203 diabetic patients (203 eye) were used to establish the ML models, and those from 169 diabetic patients (169 eye) were used for independent external validation. The random forest, gradient boosting machine (GBM), deep learning and logistic regression algorithms were used to identify the presence of DR, referable DR (RDR) and vision-threatening DR (VTDR). Four different variable patterns based on clinical data and OCTA variables were examined. The algorithms' performance were evaluated using receiver operating characteristic curves and the area under the curve (AUC) was used to assess predictive accuracy. RESULTS The random forest algorithm on OCTA+clinical data-based variables and OCTA+non-laboratory factor-based variables provided the higher AUC values for DR, RDR and VTDR. The GBM algorithm produced similar results, albeit with slightly lower AUC values. Leading predictors of DR status included vessel density, retinal thickness and GCC thickness, as well as the body mass index, waist-to-hip ratio and glucose-lowering treatment. CONCLUSIONS ML-based multiclass DR classification using OCTA and clinical data can provide reliable assistance for screening, referral, and management DR populations.
Collapse
Affiliation(s)
- Xiaoli Li
- Department of Ophthalmology, Guangdong Eye Institute, Guangdong Provincial People's Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, China
| | - Xin Wen
- Department of Ophthalmology, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, Guangzhou, China
| | - Xianwen Shang
- Department of Ophthalmology, Guangdong Eye Institute, Guangdong Provincial People's Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, China
| | - Junbin Liu
- Department of Ophthalmology, Guangdong Eye Institute, Guangdong Provincial People's Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, China
| | - Liang Zhang
- Department of Ophthalmology, Guangdong Eye Institute, Guangdong Provincial People's Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, China
| | - Ying Cui
- Department of Ophthalmology, Guangdong Eye Institute, Guangdong Provincial People's Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, China
| | - Xiaoyang Luo
- Department of Ophthalmology, Guangdong Eye Institute, Guangdong Provincial People's Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, China
| | - Guanrong Zhang
- Statistics Section, Guangdong Provincial People's Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, Guangdong, China
| | - Jie Xie
- Department of Ophthalmology, Heyuan People's Hospital, Heyuan, China
| | - Tian Huang
- Department of Ophthalmology, Zhujiang Hospital, Southern Medical University, Guangzhou, China
| | - Zhifan Chen
- Department of Ophthalmology, The Fourth Affiliated Hospital of Guangzhou Medical University, Guangzhou, China
| | - Zheng Lyu
- Department of Ophthalmology, Guangdong Eye Institute, Guangdong Provincial People's Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, China
| | - Xiyu Wu
- Department of Ophthalmology, Guangdong Eye Institute, Guangdong Provincial People's Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, China
| | - Yuqing Lan
- Department of Ophthalmology, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, Guangzhou, China.
| | - Qianli Meng
- Department of Ophthalmology, Guangdong Eye Institute, Guangdong Provincial People's Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, China.
| |
Collapse
|
2
|
Li Z, Lu T, Zhou L, Huang C, Zhao H, Liang J, Li C, Cong Q, Lan Y, Jin C. Retinal and Choroidal Alterations in Diabetic Retinopathy Treatment using Subthreshold Panretinal Photocoagulation with Endpoint Management Algorithm: A Secondary Analysis of a Randomized Clinical Trial. Ophthalmol Ther 2023; 12:1867-1880. [PMID: 37120774 PMCID: PMC10287611 DOI: 10.1007/s40123-023-00713-0] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/15/2023] [Accepted: 04/04/2023] [Indexed: 05/01/2023] Open
Abstract
INTRODUCTION The aim of this study was to compare retinal and choroidal alterations in eyes with severe nonproliferative diabetic retinopathy (NPDR) after panretinal photocoagulation (PRP), using conventional pattern scan laser (PASCAL) and PASCAL with endpoint management (EPM). METHODS This was a post hoc analysis of a paired randomized clinical trial. Bilateral treatment-naïve eyes of an individual with symmetric severe NPDR were randomly allocated into the threshold PRP group and subthreshold EPM PRP group. Patients had follow-up visits at 1, 3, 6, 9, and 12 months post-treatment. The retinal thickness (RT), choroidal thickness (CT), choroidal area, and choroidal vascularity index (CVI) were compared between the two groups and among different time points within the same group. RESULTS Seventy eyes of 35 patients with diabetes mellitus (DM) were finally included for analysis at the 6- and 12-month visits, respectively. At 3 and 6 months post-treatment, the RT in the subthreshold EPM PRP group was significantly thinner than that in the threshold PRP group. CT, stromal area, and luminal area were reduced earlier in the threshold PRP group than in the subthreshold EPM PRP group. CVI was not significantly different within the same group or between groups at most time points. CONCLUSION At 12 months post-treatment, retinal thickening and choroidal disturbance may be slightly less severe and more delayed in eyes receiving PRP using PASCAL with EPM than in those receiving PRP using conventional PASCAL. The EPM algorithm may be a good alternative in PRP when treating severe NPDR. TRIAL REGISTRATION ClinicalTrials.gov identifier, NCT01759121.
Collapse
Affiliation(s)
- Zijing Li
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-Sen University, 54 South Xianlie Road, Guangzhou, 510060, China
- Sun Yat-Sen Memorial Hospital, Sun Yat-Sen University, 107 Yanjiang West Road, Guangzhou, 510020, China
| | - Tu Lu
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-Sen University, 54 South Xianlie Road, Guangzhou, 510060, China
| | - Lijun Zhou
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-Sen University, 54 South Xianlie Road, Guangzhou, 510060, China
| | - Chuangxin Huang
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-Sen University, 54 South Xianlie Road, Guangzhou, 510060, China
| | - Hongkun Zhao
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-Sen University, 54 South Xianlie Road, Guangzhou, 510060, China
| | - Jiandong Liang
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-Sen University, 54 South Xianlie Road, Guangzhou, 510060, China
| | - Cong Li
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-Sen University, 54 South Xianlie Road, Guangzhou, 510060, China
| | - Qifeng Cong
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-Sen University, 54 South Xianlie Road, Guangzhou, 510060, China
| | - Yuqing Lan
- Sun Yat-Sen Memorial Hospital, Sun Yat-Sen University, 107 Yanjiang West Road, Guangzhou, 510020, China
| | - Chenjin Jin
- State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-Sen University, 54 South Xianlie Road, Guangzhou, 510060, China.
| |
Collapse
|
3
|
Li F, Liu L, Rao CH, Gao J. Comparison of Chorioretinal Parameters in Diabetic Retinopathy with or without Pan-Retinal Photocoagulation Using Ultrawide-Field Swept-Source Optical Coherence Tomography Angiography. Ophthalmic Res 2023; 66:538-549. [PMID: 36709752 DOI: 10.1159/000529335] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/30/2022] [Accepted: 01/16/2023] [Indexed: 01/28/2023]
Abstract
INTRODUCTION To compare structural and vascular parameters in the central and peripheral retina and choroid of eyes diagnosed with severe nonproliferative diabetic retinopathy (NPDR) or proliferative diabetic retinopathy (PDR) with or without pan-retinal photocoagulation (PRP) using ultrawide-field swept-source optical coherence tomography angiography (SS-OCTA). METHODS All participants underwent SS-OCTA imaging centered on the fovea. Retinal and choroidal thickness, vessel density of the superficial capillary plexus (SCP), deep capillary plexus (DCP), choroidal capillary plexus (CCP), and choroidal Sattler's and Haller's layer (CSHL) were analyzed in nine grids using built-in angiography analysis software. RESULTS A total of 82 eyes from 82 participants were enrolled in this study, including 40 eyes diagnosed with severe NPDR or PDR without PRP and 42 eyes with PRP. Retinal thickness in part grids was higher in eyes with PRP than in eyes without PRP. Vessel density of the SCP in nasal-superior (p = 0.003) grid was lower in eyes with PRP than in eyes without PRP. No significant difference was found in the vessel density of the DCP between the two groups (all p > 0.05). The choroidal thickness and vessel density of the CSHL of all grids were significantly lower in eyes with PRP than in eyes without PRP (all p < 0.05). A statistically significant decrease in vessel density in CCP was found in the superior (p = 0.043), nasal-superior grid (p = 0.003), macular grid (p < 0.001), and optic disc grid (p = 0.001) of eyes with PRP, compared to eyes without PRP. A significant decrease in the vessel density of CSHL was observed in all grids of PDR eyes with PRP compared to PDR eyes without PRP (all p < 0.05). Significant decrease in choroidal thickness was observed in most grids of PDR eyes with PRP, except for macular grid (p = 0.090) and optic disc grid (p = 0.057). CONCLUSION Structural and vascular parameters in the central and peripheral retinal and choroidal layers in eyes diagnosed with severe NPDR or PDR with or without PRP could be quantified using a ultrawide-field SS-OCTA. Eyes with PRP showed a significant decrease in choroidal thickness and vessel density of CCP and CSHL, compared with eyes without PRP. This trend was more obvious in eyes with PDR.
Collapse
Affiliation(s)
- Fang Li
- Department of Ophthalmology, the First Affiliated Hospital of Anhui Medical University, Hefei, China
| | - Lun Liu
- Department of Ophthalmology, the First Affiliated Hospital of Anhui Medical University, Hefei, China
| | - Cai-Hua Rao
- Department of Ophthalmology, the First Affiliated Hospital of Anhui Medical University, Hefei, China
| | - Jian Gao
- Department of Ophthalmology, the First Affiliated Hospital of Anhui Medical University, Hefei, China
| |
Collapse
|
4
|
Raman R, Ramasamy K, Shah U. A Paradigm Shift in the Management Approaches of Proliferative Diabetic Retinopathy: Role of Anti-VEGF Therapy. Clin Ophthalmol 2022; 16:3005-3017. [PMID: 36106093 PMCID: PMC9467443 DOI: 10.2147/opth.s374165] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/10/2022] [Accepted: 08/22/2022] [Indexed: 11/23/2022] Open
Abstract
Diabetic retinopathy (DR) is considered one of the leading causes of vision loss globally. It principally causes upregulation of pro-angiogenic, proinflammatory, and vascular permeability factors such as vascular endothelial growth factor (VEGF), leading to neovascularisation. The advanced stage of DR or proliferative diabetic retinopathy (PDR) is of more concern, as it leads to vitreous haemorrhage and traction retinal detachment. Various risk factors associated with PDR include hyperglycemia, hypertension, neuropathy, dyslipidemia, anaemia, nephropathy, and retinal complications of drugs used for diabetes. Current management approaches for PDR have been stratified and involve pan-retinal photocoagulation, vitrectomy, and anti-VEGF agents. Given the emerging role of anti-VEGF agents as a favourable adjunct or alternative therapy, they have a critical role in the management of PDR. The review emphasises current management approaches for PDR focusing on anti-VEGF therapy. The review also highlights the risk/benefit evaluation of the various approaches employed for PDR management in various clinical scenarios.
Collapse
Affiliation(s)
- Rajiv Raman
- Shri Bhagwan Mahavir Vitreoretinal Services, Chennai, Tamil Nadu, India
- Correspondence: Rajiv Raman, Shri Bhagwan Mahavir Vitreoretinal Services, Chennai, Tamil Nadu, India, Tel +91-9444288708, Email
| | - Kim Ramasamy
- Aravind Eye Hospital, Madurai, Tamil Nadu, India
| | - Utkarsh Shah
- Novartis Healthcare Private Limited, Mumbai, India
| |
Collapse
|