심층 학습을 이용한 삼출성 나이관련황반변성 환자의 치료 12개월 후 시력 악화 예측
Post-treatment Visual Acuity Prediction Using Deep Learning in Age-related Macular Degeneration
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
목적: 삼출성 나이관련황반변성 환자의 치료 초기 임상 정보를 바탕으로 12개월 후 시력 악화 여부를 예측하는 심층 학습 모형을구현하고자 한다. 대상과 방법: 2007년 1월부터 2020년 12월까지 삼출성 나이관련황반변성으로 진단되어 유리체 내 항혈관내피성장인자 주입술을 시행받은 환자 330안의 정보를 후향적으로 확보하였다. 수치 정보는 다층 퍼셉트론으로, 영상 정보는 합성곱신경망으로 처리하여 12개월후 시력이 초기 시력에 비해 logarithm of the minimum angle of resolution 0.3 이상 저하하였는지를 예측하였다. 환자의 치료 전및 3회 치료 후 시력, 치료 계획, 치료 전 및 3회 치료 후 빛간섭단층촬영 영상을 입력 정보로 순차적으로 하나씩 추가하며 5가지심층 학습 모형을 구현하였다. 각 모형의 성능을 비교하여 추가된 정보들의 시력 예후 예측에 대한 기여도를 분석하였다. 결과: 치료 전 및 3회 치료 후 시력, 치료 전 및 3회 치료 후 빛간섭단층촬영 영상을 순차적으로 심층 학습 모형에 투입할수록 예측력의 향상을 보였다. 또한 치료 전 및 3회 치료 후 시력, 치료 계획, 치료 전 및 3회 치료 후 빛간섭단층촬영 영상 정보 모두를 포함하여학습한 모형의 시력 예후 예측 성능이 area under the curve (AUC) 0.79로 가장 우수하였다. 결론: 삼출성 나이관련황반변성 환자의 치료 초기 임상 정보를 통해 12개월 후 시력 예후를 예측하는 심층 학습 모형은 우수한 성능을보였다. 또한 수치 정보에 빛간섭단층촬영 영상 정보를 추가한 경우 시력 예후 예측 결과가 더 좋았다.
Abstract
Purpose: To develop a deep learning model to predict visual acuity (VA) outcomes after 12 months of anti-vascular endothelial growth factor (anti-VEGF) treatment. Methods: A total of 330 treatment-naive eyes of neovascular age-related macular degeneration patients, who underwent anti- VEGF therapy between 2007 and 2020 at Konkuk University medical center, were included. The network was trained using VA at baseline, VA after three loading doses of anti-VEGF, and treatment regimen data. It was also trained using 12,300 augmented optical coherence tomography (OCT) B-scan images at baseline and after three loading doses of anti-VEGF. We generated five deep learning models using sequentially input data (VA and OCT B-scan images at baseline and after three loading doses, and treatment regimen). Prediction of VA at 12 months was performed using deep learning algorithms, such as convolutional neural network and multilayer perceptron. The outcomes were dichotomized based on whether the decremental change in VA during the 12 months of treatment was more or less than logarithm of the minimum angle of resolution 0.3. Predictive efficiency was assessed by comparing the performance of deep learning models. Results: The best performing model was trained using input data, including VA at baseline and after three loading doses, treatment regimen, and OCT B-scan images at baseline and after three loading doses. The decremental outcome in VA after 12 months of anti-VEGF treatment was predicted as an area under the curve (AUC) of 0.79. The addition of OCT images at baseline and after three loading doses as input data improved the AUC, sensitivity, and negative predictive value (AUC 0.74-0.79, 0.58-0.86, and 0.90-0.95, respectively). Conclusions: Our deep learning model showed relatively good performance in classifying good or poor post-treatment VA based on combined clinical information including numerical and image data.
참고문헌 (27)
- Augood C, "Methods for a population-based study of the prevalence of and risk factors for age-related maculopathy and macular degeneration in elderly European populations : the EUREYE study", Ophthalmic Epidemiol, 11, pp.117~129 (2004)
- Friedman DS, "Prevalence of age-related macular degeneration in the United States", Arch ophthalmol, 122, pp.564~572 (2004)
- Park SJ, "Age-related macular degeneration : prevalence and risk factors from Korean National Health and Nutrition Examination Survey, 2008 through 2011", Ophthalmology, 121, pp.1756~1765 (2014)
- Brown DM, "Ranibizumab versus verteporfin for neovascular age-related macular degeneration", N Engl J Med, 355, pp.1432~1444 (2006)
- Rosenfeld PJ, "Ranibizumab for neovascular age-related macular degeneration", N Engl J Med, 355, pp.1419~1431 (2006)
- Berg K, "Comparison of ranibizumab and bevacizumab for neovascular age-related macular degeneration according to LUCAS treat-and-extend protocol", 122, pp.146~152 (2015)
- Finger RP, "Treatment patterns, visual acuity and quality-of-life outcomes of the WAVE study-a noninterventional study of ranibizumab treatment for neovascular age-related macular degeneration in Germany", Acta Ophthalmol, 91, pp.540~546 (2013)
- Holz FG, "Multi-country real-life experience of anti-vascular endothelial growth factor therapy for wet age-related macular degeneration", Br J Ophthalmol, 99, pp.220~226 (2015)
- Rohm M, "Predicting visual acuity by using machine learning in patients treated for neovascular age-related macular degeneration", Ophthalmology, 125, pp.1028~1036 (2018)
- Schmidt-Erfurth U, "A paradigm shift in imaging biomarkers in neovascular age-related macular degeneration", Prog Retin Eye Res, 50, pp.1~24 (2016)
- Waldstein SM, "Correlation of 3-dimensionally quantified intraretinal and subretinal fluid with visual acuity in neovascular age-related macular degeneration", JAMA Ophthalmol, 134, pp.182~190 (2016)
- Roberts PK, "Baseline predictors for subretinal fibrosis in neovascular age-related macular degeneration", Sci Rep, 12, pp.88~ (2022)
- Patil NS, "Association between visual acuity and residual retinal fluid following intravitreal anti-vascular endothelial growth factor treatment for neovascular age-related macular degeneration : a systematic review and meta-analysis", JAMA Ophthalmol, 140, pp.611~622 (2022)
- Llorente-González S, "The role of retinal fluid location in atrophy and fibrosis evolution of patients with neovascular age-related macular degeneration longterm treated in real world", Acta Ophthalmol, 100, pp.e521~31 (2022)
- Jaffe GJ, "Macular morphology and visual acuity in the comparison of age-related macular degeneration treatments trials", Ophthalmology, 120, pp.1860~1870 (2013)
- Mehta H, "Real-world outcomes in patients with neovascular age-related macular degeneration treated with intravitreal vascular endothelial growth factor inhibitors", Prog Retin Eye Res, 65, pp.127~146 (2018)
- Daien V, "Evolution of treatment paradigms in neovascular age-related macular degeneration : a review of real-world evidence", Br J Ophthalmol, 105, pp.1475~1479 (2021)
- Okada M, "Nonadherence or nonpersistence to intravitreal injection therapy for neovascular age-related macular degeneration : a mixed-methods systematic review", Ophthalmology, 128, pp.234~247 (2021)
- Haller JA, "Current anti-vascular endothelial growth factor dosing regimens : benefits and burden", Ophthalmology, 120, pp.S3~7 (2013)
- Schmidt-Erfurth U, "Artificial intelligence in retina", Prog Retin Eye Res, 67, pp.1~29 (2018)
- Fu DJ, "Predicting incremental and future visual change in neovascular age-related macular degeneration using deep learning", Ophthalmol Retina, 5, pp.1074~1084 (2021)
- Liu B, "Automatic prediction of treatment outcomes in patients with diabetic macular edema using ensemble machine learning", Ann Transl Med, 9, pp.43~ (2021)
- Bogunović H, "Prediction of anti-VEGF treatment requirements in neovascular AMD using a machine learning approach", Invest Ophthalmol Vis Sci, 58, pp.3240~3248 (2017)
- Romo-Bucheli D, "End-to-end deep learning model for predicting treatment requirements in neovascular AMD from longitudinal retinal OCT imaging", IEEE J Biomed Health Inform, 24, pp.3456~3465 (2020)
- Lee H, "Post-treatment prediction of optical coherence tomography using a conditional generative adversarial network in age-related macular degeneration", Retina, 41, pp.572~580 (2021)
- Bradley AP, "The use of the area under the ROC curve in the evaluation of machine learning algorithms", Pattern Recognit, 30, pp.1145~1159 (1997)
- Almuhtaseb H, "One-year real-world outcomes in patients receiving fixed-dosing aflibercept for neovascular age-related macular degeneration", Eye(Lond), 31, pp.878~883 (2017)