1
|
Gende M, Castelo L, de Moura J, Novo J, Ortega M. Intra- and Inter-expert Validation of an Automatic Segmentation Method for Fluid Regions Associated with Central Serous Chorioretinopathy in OCT Images. JOURNAL OF IMAGING INFORMATICS IN MEDICINE 2024; 37:107-122. [PMID: 38343245 DOI: 10.1007/s10278-023-00926-6] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 06/15/2023] [Revised: 10/16/2023] [Accepted: 10/16/2023] [Indexed: 03/02/2024]
Abstract
Central Serous Chorioretinopathy (CSC) is a retinal disorder caused by the accumulation of fluid, resulting in vision distortion. The diagnosis of this disease is typically performed through Optical Coherence Tomography (OCT) imaging, which displays any fluid buildup between the retinal layers. Currently, these fluid regions are manually detected by visual inspection a time-consuming and subjective process that can be prone to errors. A series of six deep learning-based automatic segmentation architectural configurations of different levels of complexity were trained and compared in order to determine the best model intended for the automatic segmentation of CSC-related lesions in OCT images. The best performing models were then evaluated in an external validation study. Furthermore, an intra- and inter-expert analysis was conducted in order to compare the manual segmentation performed by expert ophthalmologists with the automatic segmentation provided by the models. Test results of the best performing configuration achieved a mean Dice of 0.868 ± 0.056 in the internal dataset. In the external validation set, these models achieved a level of agreement with human experts of up to 0.960 in terms of Kappa coefficient, contrasting with a value of 0.951 for agreement between human experts. Overall, the models reached a better agreement with either of the human experts than these experts with each other, suggesting that automatic segmentation models for the detection of CSC-related lesions in OCT imaging can be useful tools for assessing this disease, reducing the workload of manual inspection and leading to a more robust and objective diagnosis method.
Collapse
Affiliation(s)
- Mateo Gende
- Grupo, VARPA, Instituto de Investigación Biomédica de A Coruña (INIBIC), Universidade da Coruña, Xubias de Arriba, 84, 15006, A Coruña, Spain
- Centro de investigación, CITIC, Universidade da Coruña, Campus de Elviña s/n, 15071, A Coruña, Spain
| | - Lúa Castelo
- Grupo, VARPA, Instituto de Investigación Biomédica de A Coruña (INIBIC), Universidade da Coruña, Xubias de Arriba, 84, 15006, A Coruña, Spain
- Centro de investigación, CITIC, Universidade da Coruña, Campus de Elviña s/n, 15071, A Coruña, Spain
| | - Joaquim de Moura
- Grupo, VARPA, Instituto de Investigación Biomédica de A Coruña (INIBIC), Universidade da Coruña, Xubias de Arriba, 84, 15006, A Coruña, Spain.
- Centro de investigación, CITIC, Universidade da Coruña, Campus de Elviña s/n, 15071, A Coruña, Spain.
| | - Jorge Novo
- Grupo, VARPA, Instituto de Investigación Biomédica de A Coruña (INIBIC), Universidade da Coruña, Xubias de Arriba, 84, 15006, A Coruña, Spain
- Centro de investigación, CITIC, Universidade da Coruña, Campus de Elviña s/n, 15071, A Coruña, Spain
| | - Marcos Ortega
- Grupo, VARPA, Instituto de Investigación Biomédica de A Coruña (INIBIC), Universidade da Coruña, Xubias de Arriba, 84, 15006, A Coruña, Spain
- Centro de investigación, CITIC, Universidade da Coruña, Campus de Elviña s/n, 15071, A Coruña, Spain
| |
Collapse
|
2
|
Gawęcki M, Grzybowski A. Ganglion Cell Loss in the Course of Central Serous Chorioretinopathy. Ophthalmol Ther 2023; 12:517-533. [PMID: 36510030 PMCID: PMC9834473 DOI: 10.1007/s40123-022-00625-5] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/13/2022] [Accepted: 11/24/2022] [Indexed: 12/14/2022] Open
Abstract
INTRODUCTION Central serous chorioretinopathy (CSCR) as a clinical entity is potentially damaging and may significantly affect retinal morphology and function, especially in the chronic form. Our study aimed to determine the amount of deficit of best corrected visual acuity (BCVA) and individual retinal layers, including ganglion cells, in different types of CSCR and with reference to its duration. METHODS The retrospective analysis included 69 eyes of patients with resolved CSCR managed in Dobry Wzrok Ophthalmological Clinic between 1 January 2019 and 30 June 2022. The diagnosis of CSCR was based on the criteria outlined by the Central Serous Chorioretinopathy International Group. The analysis included data obtained from medical history, BCVA testing, and spectral domain optical coherence tomography (SD-OCT) measurements, with specific thickness values for individual retinal layers. The results were compared among affected eyes, unaffected fellow eyes, and healthy controls. RESULTS BCVA values were significantly lower in acute (0.08 ± 0.12 logMAR) and chronic (0.26 ± 0.19 logMAR) cases versus controls (0.0 logMAR). The thickness of all retinal layers (central subfoveal thickness, CST; inner retina with ganglion cell complex, GC; outer retina, ORT; and photoreceptor outer segments, POS) and macular volume (MV) were significantly decreased in chronic eyes versus controls (p < 0.01). Acute eyes had significant thinning of the outer retina and POS only compared to control eyes (p < 0.01). The amount of deficit in CST, ORT, GC, and MV was strongly correlated with poorer BCVA (p < 0.001), and the deficit of CST, ORT, and GC was correlated with disease duration (p < 0.05). The subfoveal choroidal thickness was significantly greater in affected and fellow eyes versus controls (p < 0.001). CONCLUSION Damage to the outer retina and photoreceptors occurs early in the course of CSCR, with a deficit in ganglion cells noted adjunctively in chronic forms of the disease. Further studies are required to precisely determine correlation between visual loss in CSCR and deficits in individual retinal layers.
Collapse
Affiliation(s)
- Maciej Gawęcki
- Dobry Wzrok Ophthalmological Clinic, 80-822 Gdańsk, Poland
| | - Andrzej Grzybowski
- Department of Ophthalmology, University of Warmia and Mazury, 10-719 Olsztyn, Poland ,Institute for Research in Ophthalmology, Foundation for Ophthalmology Development, Okulistyka 21, ul. Mickiewicza 24 Lok 3B, 61-553 Poznan, Poland
| |
Collapse
|
3
|
Yusef YN, Kurguzova AG, Budzinskaya MV. [Choroidal neovascularization in central serous chorioretinopathy]. Vestn Oftalmol 2023; 139:152-157. [PMID: 38235642 DOI: 10.17116/oftalma2023139061152] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/19/2024]
Abstract
The literature review discusses the features of the pathogenesis, differential diagnosis and antiangiogenic therapy of choroidal neovascularization (CNV) associated with central serous chorioretinopathy (CSC), with particular attention given to the choice of antiangiogenic drug and therapy regimen to achieve optimal anatomical and functional outcomes in patients with CSC complicated by CNV.
Collapse
Affiliation(s)
- Yu N Yusef
- Krasnov Research Institute of Eye Diseases, Moscow, Russia
- I.M. Sechenov First Moscow State Medical University (Sechenov University), Moscow, Russia
| | - A G Kurguzova
- Krasnov Research Institute of Eye Diseases, Moscow, Russia
| | | |
Collapse
|
4
|
Liu Y, Chen D, Xu J, Tan Y, Wang Y, Zhao H, Li H, Liu H, Gu Y, Qiu H. Quantitative assessment of vascular features in port wine stains through optical coherence tomography angiography. Photodiagnosis Photodyn Ther 2021; 36:102607. [PMID: 34706276 DOI: 10.1016/j.pdpdt.2021.102607] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/09/2021] [Revised: 10/16/2021] [Accepted: 10/22/2021] [Indexed: 11/18/2022]
Abstract
BACKGROUND Vascular lesions such as port wine stains (PWS) lead to facial and psychological problems, which require careful and precise treatments. The key point of treating PWS is to selectively destroy the abnormal blood vessels. Hence, the in vivo monitoring of targeted vessels is crucial. Optical coherence tomography angiography (OCTA), an emerging label-free imaging tool, facilitates the evaluation of skin structure and vasculature at a high resolution. In this study, we utilised OCTA to capture the structural and vascular morphology in patients with PWS. Moreover, we quantitatively characterised the morphological features of different types of PWS. METHODS This observational clinical study was conducted on 3 patients with flat PWS and 3 patients with thickened PWS. The age range was 4-27 years, and all of them had not received any treatment before this study. The OCTA images of the PWS lesions and contralateral skin were compared. Vascular morphology was characterized, and ectatic vessel depth was quantified according to the OCTA images. RESULTS The blood vessels of the PWS lesions tend to had larger diameters and higher densities than those in the contralateral normal skin. The vessel diameters of PWS lesions were 73 ± 14 μm, with high heterogeneity ranging from 10 to >150 μm, however, the vessel diameters of normal skin were 28 ± 2 μm, ranging from 10 μm to 60 μm. In terms of different PWS lesions, the thickened type showed a trend of larger vessel diameter and higher density than those of the purplish red type. The ectatic vessels were located at the depth of 216 ± 13 μm in the PWS skin. CONCLUSIONS OCTA can facilitate the in vivo three-dimensional visualization of structure and vasculature for PWS lesions. Various quantitative analysis parameters, such as vessel diameter, density, and depth, are typically measured using OCTA. This fact demonstrates the superior capability of OCTA for the precise and comprehensive assessment of PWS lesions.
Collapse
Affiliation(s)
- Yidi Liu
- Medical School of Chinese PLA, Beijing 100853, China; Department of Laser Medicine, the First Medical Centre, Chinese PLA General Hospital, Beijing 100853, China
| | - Defu Chen
- Institute of Engineering Medicine, Beijing Institute of Technology, Beijing 100081, China
| | - Jingjiang Xu
- School of Physics and Optoelectronic Engineering, Foshan University, Foshan 528000, China
| | - Yizhou Tan
- Department of Laser Medicine, the First Medical Centre, Chinese PLA General Hospital, Beijing 100853, China
| | - Ying Wang
- Department of Laser Medicine, the First Medical Centre, Chinese PLA General Hospital, Beijing 100853, China
| | - Hongyou Zhao
- Institute of Engineering Medicine, Beijing Institute of Technology, Beijing 100081, China
| | - Hui Li
- Medical School of Chinese PLA, Beijing 100853, China; Department of Laser Medicine, the First Medical Centre, Chinese PLA General Hospital, Beijing 100853, China
| | - Haolin Liu
- Medical School of Chinese PLA, Beijing 100853, China; Department of Laser Medicine, the First Medical Centre, Chinese PLA General Hospital, Beijing 100853, China
| | - Ying Gu
- Department of Laser Medicine, the First Medical Centre, Chinese PLA General Hospital, Beijing 100853, China; Institute of Engineering Medicine, Beijing Institute of Technology, Beijing 100081, China; Precision laser medical diagnosis and treatment Innovation unit, Chinese Academy of Medical Sciences, Beijing 100000, China.
| | - Haixia Qiu
- Department of Laser Medicine, the First Medical Centre, Chinese PLA General Hospital, Beijing 100853, China.
| |
Collapse
|