Improvements to a GLCM-based machine-learning approach for quantifying posterior capsule opacification.
cataract surgery
grey level co-occurrence matrix
machine learning
posterior capsular opacification quantification
Journal
Journal of applied clinical medical physics
ISSN: 1526-9914
Titre abrégé: J Appl Clin Med Phys
Pays: United States
ID NLM: 101089176
Informations de publication
Date de publication:
23 Jan 2024
23 Jan 2024
Historique:
revised:
26
12
2023
received:
08
06
2023
accepted:
02
01
2024
medline:
23
1
2024
pubmed:
23
1
2024
entrez:
23
1
2024
Statut:
aheadofprint
Résumé
Posterior capsular opacification (PCO) is a common complication following cataract surgery that leads to visual disturbances and decreased quality of vision. The aim of our study was to employ a machine-learning methodology to characterize and validate enhancements applied to the grey-level co-occurrence matrix (GLCM) while assessing its validity in comparison to clinical evaluations for evaluating PCO. One hundred patients diagnosed with age-related cataracts who were scheduled for phacoemulsification surgery were included in the study. Following mydriasis, anterior segment photographs were captured using a high-resolution photographic system. The GLCM was utilized as the feature extractor, and a supported vector machine as the regressor. Three variations, namely, GLCM, GLCM+C (+axial information), and GLCM+V (+regional voting), were analyzed. The reference value for regression was determined by averaging clinical scores obtained through subjective analysis. The relationships between the predicted PCO outcome scores and the ground truth were assessed using Pearson correlation analysis and a Bland-Altman plot, while agreement between them was assessed through the Bland-Altman plot. Relative to the ground truth, the GLCM, GLCM+C, and GLCM+V methods exhibited correlation coefficients of 0.706, 0.768, and 0.829, respectively. The relationship between the PCO score predicted by the GLCM+V method and the ground truth was statistically significant (p < 0.001). Furthermore, the GLCM+V method demonstrated competitive performance comparable to that of two experienced clinicians (r = 0.825, 0.843) and superior to that of two junior clinicians (r = 0.786, 0.756). Notably, a high level of agreement was observed between predictions and the ground truth, without significant evidence of proportional bias (p > 0.05). Overall, our findings suggest that a machine-learning approach incorporating the GLCM, specifically the GLCM+V method, holds promise as an objective and reliable tool for assessing PCO progression. Further studies in larger patient cohorts are warranted to validate these findings and explore their potential clinical applications.
Sections du résumé
BACKGROUND
BACKGROUND
Posterior capsular opacification (PCO) is a common complication following cataract surgery that leads to visual disturbances and decreased quality of vision. The aim of our study was to employ a machine-learning methodology to characterize and validate enhancements applied to the grey-level co-occurrence matrix (GLCM) while assessing its validity in comparison to clinical evaluations for evaluating PCO.
METHODS
METHODS
One hundred patients diagnosed with age-related cataracts who were scheduled for phacoemulsification surgery were included in the study. Following mydriasis, anterior segment photographs were captured using a high-resolution photographic system. The GLCM was utilized as the feature extractor, and a supported vector machine as the regressor. Three variations, namely, GLCM, GLCM+C (+axial information), and GLCM+V (+regional voting), were analyzed. The reference value for regression was determined by averaging clinical scores obtained through subjective analysis. The relationships between the predicted PCO outcome scores and the ground truth were assessed using Pearson correlation analysis and a Bland-Altman plot, while agreement between them was assessed through the Bland-Altman plot.
RESULTS
RESULTS
Relative to the ground truth, the GLCM, GLCM+C, and GLCM+V methods exhibited correlation coefficients of 0.706, 0.768, and 0.829, respectively. The relationship between the PCO score predicted by the GLCM+V method and the ground truth was statistically significant (p < 0.001). Furthermore, the GLCM+V method demonstrated competitive performance comparable to that of two experienced clinicians (r = 0.825, 0.843) and superior to that of two junior clinicians (r = 0.786, 0.756). Notably, a high level of agreement was observed between predictions and the ground truth, without significant evidence of proportional bias (p > 0.05).
CONCLUSIONS
CONCLUSIONS
Overall, our findings suggest that a machine-learning approach incorporating the GLCM, specifically the GLCM+V method, holds promise as an objective and reliable tool for assessing PCO progression. Further studies in larger patient cohorts are warranted to validate these findings and explore their potential clinical applications.
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
e14268Subventions
Organisme : Clinical Study of Shanghai Municipal Health Commission
ID : 20194Y0290
Informations de copyright
© 2024 The Authors. Journal of Applied Clinical Medical Physics published by Wiley Periodicals, LLC on behalf of The American Association of Physicists in Medicine.
Références
Ursell PG, Dhariwal M, O'Boyle D, Khan J, Venerus A. 5 year incidence of YAG capsulotomy and PCO after cataract surgery with single-piece monofocal intraocular lenses: a real-world evidence study of 20,763 eyes. Eye. 2019;34(5):960-968.
Wormstone IM, Wormstone YM, Smith AJO, Eldred JA. Posterior capsule opacification: what's in the bag? Prog Retin Eye Res. 2021;82:100905.
Awasthi N, Guo S, Wagner BJ. Posterior capsular opacification: a problem reduced but not yet eradicated. Arch Ophthalmol. 2009;127(4):555-562.
Nibourg LM, Gelens E, Kuijer R, Hooymans JM, van Kooten TG, Koopmans SA. Prevention of posterior capsular opacification. Exp Eye Res. 2015;136(1):100-115.
Hollo G, Aung T, Cantor LB, Aihara M. Cystoid macular edema related to cataract surgery and topical prostaglandin analogs: mechanism, diagnosis, and management. Surv Ophthalmol. 2020;65(5):496-512.
Gu X, Chen X, Jin G, et al. Early-onset posterior capsule opacification: incidence, severity, and risk factors. Ophthalmol Ther. 2021;11(1):113-123.
Apple DJ, Peng Q, Visessook N, et al. Eradication of posterior capsule opacification: documentation of a marked decrease in Nd:yAG laser posterior capsulotomy rates noted in an analysis of 5416 pseudophakic human eyes obtained postmortem. Ophthalmology. 2001;108(3):505-518.
Bhargava R, Kumar P, Phogat H, Chaudhary KP. Neodymium-yttrium aluminium garnet laser capsulotomy energy levels for posterior capsule opacification. J Ophthalmic Vis Res. 2015;10(1):37-42.
Hu C, Jing R, Wen C, et al. HNF4A negatively regulated posterior capsular opacification via transcriptional inhibition of MMP2. Curr Eye Res. 2023;48(7):627-638.
Li J, Xue W, Wang X, et al. HSP90 as a novel therapeutic target for posterior capsule opacification. Exp Eye Res. 2019:189.
Jing R, Hu C, Qi T, et al. FILIP1L-mediated cell apoptosis, epithelial-mesenchymal transition and extracellular matrix synthesis aggravate posterior capsular opacification. Life Sci. 2021:286.
Liu D, Wu Q, Chen W, et al. Nanoporous gold ring-integrated photothermal intraocular lens for active prevention of posterior capsular opacification. Small. 2022;18(34).
Maedel S, Evans JR, Harrer-Seely A, Findl O. Intraocular lens optic edge design for the prevention of posterior capsule opacification after cataract surgery. Cochrane Database Syst Rev. 2021;8(8):CD012516.
Zhang X, Wang J, Xu J, et al. Prophylaxis of posterior capsule opacification through autophagy activation with indomethacin-eluting intraocular lens. Bioact Mater. 2023:23539-23550.
Grewal D, Jain R, Brar GS, Grewal SP. Pentacam tomograms: a novel method for quantification of posterior capsule opacification. Invest Ophthalmol Vis Sci. 2008;49(5):2004-2008.
Tanaka Y, Kato S, Miyata K, et al. Limitation of Scheimpflug videophotography system in quantifying posterior capsule opacification after intraocular lens implantation. Am J Ophthalmol. 2004;137(4):732-735.
Aslam TMDB, Werghi N, Taguri A, Wadood A. Systems of analysis of posterior capsule opacification. Br J Ophthalmol. 2002;86(10):1181-1186.
Moreno-Montanes J, Alvarez A, Maldonado MJ. Objective quantification of posterior capsule opacification after cataract surgery, with optical coherence tomography. Invest Ophthalmol Vis Sci. 2005;46(11):3999-4006.
Tetz MR, Auffarth GU, Sperker M, Blum M, Völcker HE. Photographic image analysis system of posterior capsule opacification. J Cataract Refract Surg. 1997;23(10):1515-1520.
Barman SA, Hollick EJ, Boyce JF, et al. Quantification of posterior capsular opacification in digital images after cataract surgery. Invest Ophthalmol Vis Sci. 2000;41(12):3882-3892.
Bender L, Spalton DJ, Uyanonvara B, et al. POCOman: new system for quantifying posterior capsule opacification. J Cataract Refract Surg. 2004;30(10):2058-2063.
Findl O, Buehl W, Menapace R, et al. Comparison of 4 methods for quantifying posterior capsule opacification. J Cataract Refract Surg. 2003;29(1):106-111.
Aslam TM, Patton N, Graham J. A freely accessible, evidence based, objective system of analysis of posterior capsular opacification; evidence for its validity and reliability. BMC Ophthalmol. 2005;5(1):9.
Dai L, Wu L, Li H, et al. A deep learning system for detecting diabetic retinopathy across the disease spectrum. Nat Commun. 2021;12(1):3242.
Kronschlager M, Siegl H, Pinz A, et al. Automated qualitative and quantitative assessment of posterior capsule opacification by Automated Quantification of After-Cataract II (AQUA II) system. BMC Ophthalmol. 2019;19(1):114.
Huemer J, Kronschläger M, Ruiss M, et al. Diagnostic accuracy of code-free deep learning for detection and evaluation of posterior capsule opacification. BMJ Open Ophthalmol. 2022;7(1):e000992.
Tang X, Yang Y, Huang L, Qiu L. The application of texture feature analysis of rectus femoris based on local binary pattern (LBP) combined with gray-level co-occurrence matrix (GLCM) in sarcopenia. J Ultrasound Med. 2022;41(9):2169-2179.
Auffarth GU, Golescu A, Becker KA, Völcker HE. Quantification of posterior capsule opacification with round and sharp edge intraocular lenses. Ophthalmology. 2003;110(4):772-780.
Aslam TM, Patton N. Methods of assessment of patients for Nd:yAG laser capsulotomy that correlate with final visual improvement. BMC Ophthalmol. 2004;4(1):13.
Hayashi K, Hayashi H, Nakao F, Hayashi F. In vivo quantitative measurement of posterior capsule opacification after extracapsular cataract surgery. Am J Ophthalmol. 1998;125(6):837-843.
Shin H-C, Roth HR, Gao M, et al. Deep convolutional neural networks for computer-aided detection: CNN architectures, dataset characteristics and transfer learning. IEEE Trans Med Imaging. 2016;35(5):1285-1298.
Wang G, Li W, Zuluaga MA, et al. Interactive medical image segmentation using deep learning with image-specific fine tuning. IEEE Trans Med Imaging. 2018;37(7):1562-1573.
Gao F, Wu T, Chu X, Yoon H, Xu Y, Patel B. Deep residual inception encoder-decoder network for medical imaging synthesis. Ieee J Biomed Health Inf. 2020;24(1):39-49.
Benyahia S, Meftah B, Lézoray O. Multi-features extraction based on deep learning for skin lesion classification. Tissue Cell. 2022:74.
Wu W, Hu D, Niu C, Yu H, Vardhanabhuti V, Wang G. DRONE: Dual-Domain Residual-based Optimization NEtwork for sparse-view CT reconstruction. IEEE Trans Med Imaging. 2021;40(11):3002-3014.