Accuracy of a Machine-Learning Algorithm for Detecting and Classifying Choroidal Neovascularization on Spectral-Domain Optical Coherence Tomography
Andreas Maunz, Fethallah Benmansour, Yvonna Li, Thomas Albrecht, Yan-Ping Zhang, Filippo Arcadu, Yalin Zheng, Savita Madhusudhan, Jayashree Sahni, Andreas Maunz, Fethallah Benmansour, Yvonna Li, Thomas Albrecht, Yan-Ping Zhang, Filippo Arcadu, Yalin Zheng, Savita Madhusudhan, Jayashree Sahni
Abstract
Background: To evaluate the performance of a machine-learning (ML) algorithm to detect and classify choroidal neovascularization (CNV), secondary to age-related macular degeneration (AMD) on spectral-domain optical coherence tomography (SD-OCT) images.
Methods: Baseline fluorescein angiography (FA) and SD-OCT images from 1037 treatment-naive study eyes and 531 fellow eyes, without advanced AMD from the phase 3 HARBOR trial (NCT00891735), were used to develop, train, and cross-validate an ML pipeline combining deep-learning-based segmentation of SD-OCT B-scans and CNV classification, based on features derived from the segmentations, in a five-fold setting. FA classification of the CNV phenotypes from HARBOR was used for generating the ground truth for model development. SD-OCT scans from the phase 2 AVENUE trial (NCT02484690) were used to externally validate the ML model.
Results: The ML algorithm discriminated CNV absence from CNV presence, with a very high accuracy (area under the receiver operating characteristic [AUROC] = 0.99), and classified occult versus predominantly classic CNV types, per FA assessment, with a high accuracy (AUROC = 0.91) on HARBOR SD-OCT images. Minimally classic CNV was discriminated with significantly lower performance. Occult and predominantly classic CNV types could be discriminated with AUROC = 0.88 on baseline SD-OCT images of 165 study eyes, with CNV from AVENUE.
Conclusions: Our ML model was able to detect CNV presence and CNV subtypes on SD-OCT images with high accuracy in patients with neovascular AMD.
Keywords: age-related macular degeneration; choroidal neovascularization; classification; machine learning; optical coherence tomography.
Conflict of interest statement
A. Maunz, F. Benmansour, Y. Li, T. Albrecht, Y.-P. Zhang, F. Arcadu, and J. Sahni, F. Hoffmann-La Roche Ltd., Basel, Switzerland (E); Y. Zheng, F. Hoffmann-La Roche Ltd., Basel, Switzerland (F); S. Madhusudhan, F. Hoffmann-La Roche Ltd., Basel, Switzerland (F).
Figures
References
- Wong T.Y., Chakravarthy U., Klein R., Mitchell P., Zlateva G., Buggage R., Fahrbach K., Prost C., Sledge I. The natural history and prognosis of neovascular age-related macular degeneration: A systematic review of the literature and meta-analysis. Ophthalmology. 2008;115:116–126. doi: 10.1016/j.ophtha.2007.03.008.
- Gess A.J., Fung A.E., Rodriguez J.G. Imaging in neovascular age-related macular degeneration. Semin. Ophthalmol. 2011;26:225–233. doi: 10.3109/08820538.2011.582533.
- Castillo M.M., Mowatt G., Elders A., Lois N., Fraser C., Hernandez R., Amoaku W., Burr J.M., Lotery A., Ramsay C.R., et al. Optical coherence tomography for the monitoring of neovascular age-related macular degeneration: A systematic review. Ophthalmology. 2015;122:399–406. doi: 10.1016/j.ophtha.2014.07.055.
- Rosenfeld P.J. Optical coherence tomography and the development of antiangiogenic therapies in neovascular age-related macular degeneration. Investig. Ophthalmol. Vis. Sci. 2016;57:OCT14–OCT26. doi: 10.1167/iovs.16-19969.
- Schmidt-Erfurth U., Chong V., Loewenstein A., Larsen M., Souied E., Schlingemann R., Eldem B., Mones J., Richard G., Bandello F. Guidelines for the management of neovascular age-related macular degeneration by the European Society of Retina Specialists (EURETINA) Br. J. Ophthalmol. 2014;98:1144–1167. doi: 10.1136/bjophthalmol-2014-305702.
- Koh A., Lee W.K., Chen L.J., Chen S.J., Hashad Y., Kim H., Lai T.Y.Y., Pilz S., Ruamviboonsuk P., Tokaji E., et al. EVEREST study: Efficacy and safety of verteporfin photodynamic therapy in combination with ranibizumab or alone versus ranibizumab monotherapy in patients with symptomatic macular polypoidal choroidal vasculopathy. Retina. 2012;32:1453–1464. doi: 10.1097/IAE.0b013e31824f91e8.
- Koh A., Lai T.Y.Y., Takahashi K., Wong T.Y., Chen L.J., Ruamviboonsuk P., Tan C.S., Feller C., Margaron P., Lim T.H., et al. Efficacy and safety of ranibizumab with or without verteporfin photodynamic therapy for polypoidal choroidal vasculopathy: A randomized clinical trial. JAMA Ophthalmol. 2017;135:1206–1213. doi: 10.1001/jamaophthalmol.2017.4030.
- Lim T.H., Lai T.Y.Y., Takahashi K., Wong T.Y., Chen L.J., Ruamviboonsuk P., Tan C.S., Lee W.K., Cheung C.M.G., Ngah N.F., et al. Comparison of ranibizumab with or without verteporfin photodynamic therapy for polypoidal choroidal vasculopathy. The EVEREST II randomized clinical trial. JAMA Ophthalmol. 2020;138:935–942. doi: 10.1001/jamaophthalmol.2020.2443.
- Jaffe G.J., Ciulla T.A., Ciardella A.P., Devin F., Dugel P.U., Eandi C.M., Masonson H., Mones J., Pearlman J.A., Quaranta-El Maftouhi M., et al. Dual antagonism of PDGF and VEGF in neovascular age-related macular degeneration: A phase IIb, multicenter, randomized controlled trial. Ophthalmology. 2017;124:224–234. doi: 10.1016/j.ophtha.2016.10.010.
- Sahni J., Dugel P.U., Patel S.S., Chittum M.E., Berger B., Del Valle Rubido M., Sadikhov S., Szczesny P., Schwab D., Nogoceke E., et al. Safety and efficacy of different doses and regimens of faricimab vs ranibizumab in neovascular age-related macular degeneration: The AVENUE phase 2 randomized clinical trial. JAMA Ophthalmol. 2020;138:955–963. doi: 10.1001/jamaophthalmol.2020.2685.
- Khanani A.M., Patel S.S., Ferrone P.J., Osborne A., Sahni J., Grzeschik S., Basu K., Ehrlich J.S., Haskova Z., Dugel P.U. Efficacy of every four monthly and quarterly dosing of faricimab vs ranibizumab in neovascular age-related macular degeneration: The STAIRWAY phase 2 randomized clinical trial. JAMA Ophthalmol. 2020;138:964–972. doi: 10.1001/jamaophthalmol.2020.2699.
- Spaide R.F., Jaffe G.J., Sarraf D., Freund K.B., Sadda S.R., Staurenghi G., Waheed N.K., Chakravarthy U., Rosenfeld P.J., Holz F.G., et al. Consensus nomenclature for reporting neovascular age-related macular degeneration data: Consensus on Neovascular Age-Related Macular Degeneration Nomenclature Study Group. Ophthalmology. 2020;127:616–636. doi: 10.1016/j.ophtha.2019.11.004.
- Macular Photocoagulation Study Group Subfoveal neovascular lesions in age-related macular degeneration. Guidelines for evaluation and treatment in the Macular Photocoagulation Study. Arch. Ophthalmol. 1991;109:1242–1257. doi: 10.1001/archopht.1991.01080090066027.
- Busbee B.G., Ho A.C., Brown D.M., Heier J.S., Suner I.J., Li Z., Rubio R.G., Lai P. Twelve-month efficacy and safety of 0.5 mg or 2.0 mg ranibizumab in patients with subfoveal neovascular age-related macular degeneration. Ophthalmology. 2013;120:1046–1056. doi: 10.1016/j.ophtha.2012.10.014.
- Ho A.C., Busbee B.G., Regillo C.D., Wieland M.R., Van Everen S.A., Li Z., Rubio R.G., Lai P. Twenty-four-month efficacy and safety of 0.5 mg or 2.0 mg ranibizumab in patients with subfoveal neovascular age-related macular degeneration. Ophthalmology. 2014;121:2181–2192. doi: 10.1016/j.ophtha.2014.05.009.
- Abràmoff M.D., Garvin M.K., Sonka M. Retinal imaging and image analysis. IEEE Rev. Biomed. Eng. 2010;3:169–208. doi: 10.1109/RBME.2010.2084567.
- Van der Walt S., Schönberger J.L., Nunez-Iglesias J., Boulogne F., Warner J.D., Yager N., Gouillart E., Yu T. scikit-image: Image processing in Python. PeerJ. 2014;2:e453. doi: 10.7717/peerj.453.
- Ristau T., Keane P.A., Walsh A.C., Engin A., Mokwa N., Kirchhof B., Sadda S.R., Liakopoulos S. Relationship between visual acuity and spectral domain optical coherence tomography retinal parameters in neovascular age-related macular degeneration. Ophthalmologica. 2014;231:37–44. doi: 10.1159/000354551.
- Willoughby A.S., Ying G.S., Toth C.A., Maguire M.G., Burns R.E., Grunwald J.E., Ebenezer D., Jaffe G.J. Subretinal hyperreflective material in the Comparison of Age-Related Macular Degeneration Treatments Trials. Ophthalmology. 2015;122:1846–1853.e5. doi: 10.1016/j.ophtha.2015.05.042.
- Ronneberger O., Fischer F., Brox T. U-Net: Convolutional networks for biomedical image segmentation. arXiv. 20151505.04597
- Early Treatment Diabetic Retinopathy Study Research Group Grading diabetic retinopathy from stereoscopic color fundus photographs—an extension of the modified Airlie House classification. ETDRS report number 10. Ophthalmology. 1991;98:786–806. doi: 10.1016/S0161-6420(13)38012-9.
- Lundberg S., Lee S.-I. A unified approach to interpreting model predictions. arXiv. 20171705.07874
- Harding S.P. Neovascular age-related macular degeneration: Decision making and optimal management. Eye. 2010;24:497–505. doi: 10.1038/eye.2009.316.
- Khurana R.N., Dupas B., Bressler N.M. Agreement of time-domain and spectral-domain optical coherence tomography with fluorescein leakage from choroidal neovascularization. Ophthalmology. 2010;117:1376–1380. doi: 10.1016/j.ophtha.2009.11.039.
- Salinas-Alamán A., García-Layana A., Maldonado M.J., Sainz-Gómez C., Alvárez-Vidal A. Using optical coherence tomography to monitor photodynamic therapy in age related macular degeneration. Am. J. Ophthalmol. 2005;140:23–28. doi: 10.1016/j.ajo.2005.01.044.
- Van de Moere A., Sandhu S.S., Talks S.J. Correlation of optical coherence tomography and fundus fluorescein angiography following photodynamic therapy for choroidal neovascular membranes. Br. J. Ophthalmol. 2006;90:304–306. doi: 10.1136/bjo.2005.079947.
- Hee M.R., Baumal C.R., Puliafito C.A., Duker J.S., Wilkins J.R., Coker J.G., Schuman J.S., Swanson E.A., Fujimoto J.G. Optical coherence tomography of age-related macular degeneration and choroidal neovascularization. Ophthalmology. 1996;103:1260–1270. doi: 10.1016/S0161-6420(96)30512-5.
- Do D.V., Gower E.W., Cassard S.D., Boyer D., Bressler N.M., Bressler S.B., Heier J.S., Jefferys J.L., Singerman L.J., Solomon S.D. Detection of new-onset choroidal neovascularization using optical coherence tomography: The AMD DOC Study. Ophthalmology. 2012;119:771–778. doi: 10.1016/j.ophtha.2011.10.019.
- Gualino V., Tadayoni R., Cohen S.Y., Erginay A., Fajnkuchen F., Haouchine B., Krivosic V., Quentel G., Gaudric A. Optical coherence tomography, fluorescein angiography, and diagnosis of choroidal neovascularization in age-related macular degeneration. Retina. 2019;39:1664–1671. doi: 10.1097/IAE.0000000000002220.
- Padnick-Silver L., Weinberg A.B., Lafranco F.P., Macsai M.S. Pilot study for the detection of early exudative age-related macular degeneration with optical coherence tomography. Retina. 2012;32:1045–1056. doi: 10.1097/IAE.0b03e31823fb82b.
- Wilde C., Patel M., Lakshmanan A., Amankwah R., Dhar-Munshi S., Amoaku W. The diagnostic accuracy of spectral-domain optical coherence tomography for neovascular age-related macular degeneration: A comparison with fundus fluorescein angiography. Eye. 2015;29:602–609. doi: 10.1038/eye.2015.44.
- Moisseiev J., Alhalel A., Masuri R., Treister G. The impact of the Macular Photocoagulation Study results on the treatment of exudative age-related macular degeneration. Arch. Ophthalmol. 1995;113:185–189. doi: 10.1001/archopht.1995.01100020069031.
- Bogunović H., Venhuizenm F., Klimscha S., Apostolopoulos S., Bab-Hadiashar A., Bagci U., Beg M.F., Bekalo L., Chen Q., Ciller C., et al. RETOUCH: The retinal OCT fluid detection and segmentation benchmark and challenge. IEEE Trans. Med. Imaging. 2019;38:1858–1874. doi: 10.1109/TMI.2019.2901398.
- Giani A., Luiselli C., Esmaili D.D., Salvetti P., Cigada M., Miller J.W., Staurenghi G. Spectral-domain optical coherence tomography as an indicator of fluorescein angiography leakage from choroidal neovascularization. Investig. Ophthalmol. Vis. Sci. 2011;52:5579–5586. doi: 10.1167/iovs.10-6617.
Source: PubMed