机器学习在眼表疾病诊断及角膜手术中的应用进展

doi:10.6040/j.issn.1673-3770.0.2021.329

摘要/Abstract

摘要： 近年来,机器学习及其分支深度学习技术得到了广泛的关注和应用。在眼科学领域,国内外学者对应用经典机器学习和深度学习技术进行眼病的筛查和诊断进行了大量研究。论文综述了常见机器学习算法在眼表疾病和手术中的应用进展,包括圆锥角膜、糖尿病周围神经病变、干眼的诊断和筛查以及角膜手术设计等方面的应用。

关键词: 机器学习, 圆锥角膜, 糖尿病周围神经病变, 干眼, 角膜手术, 深度学习

Abstract: Machine learning and its subdivision deep learning, has sparked considerable interest regarding their applications in medicine, including the screening of ophthalmological diseases and the subsequent treatment design. This article summarizes recent development of machine learning in ocular surface diseases and surgical procedures, including screening of keratoconus, diabetic peripheral neuropathy, dry eye disease, and guiding refractive surgery, intracorneal ring implantation and corneal transplant.

Key words: Machine learning, Keratoconus, Diabetic peripheral neuropathy, Dry eye disease, Corneal surgical procedures, Deep learning

中图分类号:

R772.2

黄天泽,陈迪,李莹. 机器学习在眼表疾病诊断及角膜手术中的应用进展[J]. 山东大学耳鼻喉眼学报, 2021, 35(6): 13-19.

Huang Tianze, Chen Di,LI Ying. Advances of machine learning in the diagnosis of ocular surface diseases and guiding corneal surgical procedures[J]. Journal of Otolaryngology and Ophthalmology of Shandong University, 2021, 35(6): 13-19.

参考文献

[1] Shen D, Wu G, Suk HI. Deep learning in medical image analysis[J]. Annu Rev Biomed Eng, 2017, 19: 221-248. doi:10.1146/annurev-bioeng-071516-044442.
[2] Li ZX, He YF, Keel S, et al. Efficacy of a deep learning system for detecting glaucomatous optic neuropathy based on color fundus photographs[J]. Ophthalmology, 2018, 125(8): 1199-1206. doi:10.1016/j.ophtha.2018.01.023.
[3] Lee J, Kim JS, Lee HJ, et al. Discriminating glaucomatous and compressive optic neuropathy on spectral-domain optical coherence tomography with deep learning classifier[J]. Br J Ophthalmol, 2020, 104(12): 1717-1723. doi:10.1136/bjophthalmol-2019-314330.
[4] Liu Y, Yang J, Zhou Y, et al. Prediction of OCT images of short-term response to anti-VEGF treatment for neovascular age-related macular degeneration using generative adversarial network[J]. Br J Ophthalmol, 2020, 104(12): 1735-1740. doi:10.1136/bjophthalmol-2019-315338.
[5] Liu XY, Jiang JW, Zhang K, et al. Localization and diagnosis framework for pediatric cataracts based on slit-lamp images using deep features of a convolutional neural network[J]. PLoS One, 2017, 12(3): e0168606. doi:10.1371/journal.pone.0168606.
[6] Vujosevic S, Aldington SJ, Silva P, et al. Screening for diabetic retinopathy: new perspectives and challenges[J]. Lancet Diabetes Endocrinol, 2020, 8(4): 337-347. doi:10.1016/S2213-8587(19)30411-5.
[7] Salahouddin T, Petropoulos IN, Ferdousi M, et al. Artificial intelligence-based classification of diabetic peripheral neuropathy from corneal confocal microscopy images[J]. Diabetes Care, 2021, 44(7): e151-e153. doi:10.2337/dc20-2012.
[8] Cui T, Wang Y, Ji SF, et al. Applying machine learning techniques in nomogram prediction and analysis for SMILE treatment[J]. Am J Ophthalmol, 2020, 210: 71-77. doi:10.1016/j.ajo.2019.10.015.
[9] Siddiqui AA, Ladas JG, Lee JK. Artificial intelligence in cornea, refractive, and cataract surgery[J]. Curr Opin Ophthalmol, 2020, 31(4): 253-260. doi:10.1097/ICU.0000000000000673.
[10] Lyra D, Ribeiro G, Torquetti L, et al. Computational models for optimization of the intrastromal corneal ring choice in patients with keratoconus using corneal tomography data[J]. J Refract Surg, 2018, 34(8): 547-550. doi:10.3928/1081597X-20180615-01.
[11] Fariselli C, Vega-Estrada A, Arnalich-Montiel F, et al. Artificial neural network to guide intracorneal ring segments implantation for keratoconus treatment: a pilot study[J]. Eye Vis(Lond), 2020, 7: 20. doi:10.1186/s40662-020-00184-5.
[12] Valdés-Mas MA, Martín-Guerrero JD, Rupérez MJ, et al. A new approach based on Machine Learning for predicting corneal curvature(K1)and astigmatism in patients with keratoconus after intracorneal ring implantation[J]. Comput Methods Programs Biomed, 2014, 116(1): 39-47. doi:10.1016/j.cmpb.2014.04.003.
[13] Shen Y, Wang L, Jian WJ, et al. Big-data and artificial-intelligence-assisted vault prediction and EVO-ICL size selection for myopia correction[J]. Br J Ophthalmol, 2021: bjophthalmol-bjophtha2021-319618. doi:10.1136/bjophthalmol-2021-319618.
[14] LeCun Y, Bengio Y, Hinton G. Deep learning[J]. Nature, 2015, 521(7553): 436-444. doi:10.1038/nature14539.
[15] Maeda N, Klyce SD, Smolek MK. Neural network classification of corneal topography. Preliminary demonstration[J]. Invest Ophthalmol Vis Sci, 1995, 36(7): 1327-1335.
[16] Ruiz Hidalgo I, Rozema JJ, Saad A, et al. Validation of an objective keratoconus detection system implemented in a scheimpflug tomographer and comparison with other methods[J]. Cornea, 2017, 36(6): 689-695. doi:10.1097/ICO.0000000000001194.
[17] Arbelaez MC, Versaci F, Vestri G, et al. Use of a support vector machine for keratoconus and subclinical keratoconus detection by topographic and tomographic data[J]. Ophthalmology, 2012, 119(11): 2231-2238. doi:10.1016/j.ophtha.2012.06.005.
[18] Kovács I, Miháltz K, Kránitz K, et al. Accuracy of machine learning classifiers using bilateral data from a Scheimpflug camera for identifying eyes with preclinical signs of keratoconus[J]. J Cataract Refract Surg, 2016, 42(2): 275-283. doi:10.1016/j.jcrs.2015.09.020.
[19] Ambrósio R, Lopes BT, Faria-Correia F, et al. Integration of scheimpflug-based corneal tomography and biomechanical assessments for enhancing ectasia detection[J]. J Refract Surg Thorofare N J, 2017, 33(7): 434-443. doi:10.3928/1081597X-20170426-02.
[20] Herber R, Pillunat LE, Raiskup F. Development of a classification system based on corneal biomechanical properties using artificial intelligence predicting keratoconus severity[J]. Eye Vis(Lond), 2021, 8(1): 21. doi:10.1186/s40662-021-00244-4.
[21] Issarti I, Consejo A, Jiménez-García M, et al. Computer aided diagnosis for suspect keratoconus detection[J]. Comput Biol Med, 2019, 109: 33-42. doi:10.1016/j.compbiomed.2019.04.024.
[22] Kamiya K, Ayatsuka Y, Kato Y, et al. Keratoconus detection using deep learning of colour-coded maps with anterior segment optical coherence tomography: a diagnostic accuracy study[J]. BMJ Open, 2019, 9(9): e031313. doi:10.1136/bmjopen-2019-031313.
[23] Elsawy A, Eleiwa T, Chase C, et al. Multidisease deep learning neural network for the diagnosis of corneal diseases[J]. Am J Ophthalmol, 2021, 226: 252-261. doi:10.1016/j.ajo.2021.01.018.
[24] Xie Y, Zhao LQ, Yang XN, et al. Screening candidates for refractive surgery with corneal tomographic-based deep learning[J]. JAMA Ophthalmol, 2020, 138(5): 519-526. doi:10.1001/jamaophthalmol.2020.0507.
[25] Daud MM, Zaki WMDW, Hussain A, et al. Keratoconus detection using the fusion features of anterior and lateral segment photographed images[J]. IEEE Access, 2020, 8: 142282-142294. doi:10.1109/ACCESS.2020.3012583.
[26] Mahmoud HAH, Mengash HA. Automated keratoconus detection by 3D corneal images reconstruction[J]. Sensors(Basel), 2021, 21(7): 2326. doi:10.3390/s21072326.
[27] Scarpa F, Grisan E, Ruggeri A. Automatic recognition of corneal nerve structures in images from confocal microscopy[J]. Invest Ophthalmol Vis Sci, 2008, 49(11): 4801-4807. doi:10.1167/iovs.08-2061.
[28] Dabbah MA, Graham J, Petropoulos IN, et al. Automatic analysis of diabetic peripheral neuropathy using multi-scale quantitative morphology of nerve fibres in corneal confocal microscopy imaging[J]. Med Image Anal, 2011, 15(5): 738-747. doi:10.1016/j.media.2011.05.016.
[29] Chen X, Graham J, Dabbah MA, et al. An automatic tool for quantification of nerve fibers in corneal confocal microscopy images[J]. IEEE Trans Biomed Eng, 2017, 64(4): 786-794. doi:10.1109/TBME.2016.2573642.
[30] Dehghani C, Pritchard N, Edwards K, et al. Fully automated, semiautomated, and manual morphometric analysis of corneal subbasal nerve plexus in individuals with and without diabetes[J]. Cornea, 2014, 33(7): 696-702. doi:10.1097/ICO.0000000000000152.
[31] Salahouddin T, Petropoulos IN, Ferdousi M, et al. Artificial intelligence-based classification of diabetic peripheral neuropathy from corneal confocal microscopy images[J]. Diabetes Care, 2021, 44(7): e151-e153. doi:10.2337/dc20-2012.
[32] Williams BM, Borroni D, Liu RJ, et al. An artificial intelligence-based deep learning algorithm for the diagnosis of diabetic neuropathy using corneal confocal microscopy: a development and validation study[J]. Diabetologia, 2020, 63(2): 419-430. doi:10.1007/s00125-019-05023-4.
[33] 亚洲干眼协会中国分会, 海峡两岸医药卫生交流协会眼科学专业委员会眼表与泪液病学组, 中国医师协会眼科医师分会眼表与干眼学组. 中国干眼专家共识:检查和诊断(2020年)[J]. 中华眼科杂志, 2020, 56(10): 741-747.
[34] Yedidya T, Carr P, Hartley R, et al. Enforcing monotonic temporal evolution in dry eye images[M] //Medical Image Computing and Computer-Assisted Intervention-MICCAI 2009. Berlin, Heidelberg: Springer Berlin Heidelberg, 2009: 976-984. doi:10.1007/978-3-642-04271-3_118.
[35] Ramos L, Barreira N, Mosquera A, et al. Analysis of parameters for the automatic computation of the tear film break-up time test based on CCLRU standards[J]. Comput Methods Programs Biomed, 2014, 113(3): 715-724. doi:10.1016/j.cmpb.2013.12.003.
[36] Gumus K, Crockett CH, Rao K, et al. Noninvasive assessment of tear stability with the tear stability analysis system in tear dysfunction patients[J]. Invest Ophthalmol Vis Sci, 2011, 52(1): 456-461. doi:10.1167/iovs.10-5292.
[37] Lan W, Lin L, Yang X, et al. Automatic noninvasive tear breakup time(TBUT)and conventional fluorescent TBUT[J]. Optom Vis Sci, 2014, 91(12): 1412-1418. doi:10.1097/opx.0000000000000418.
[38] Carpente A, Ramos L, Barreira N, et al. On the automation of the tear film non-invasive break-up test[C] //2014 IEEE 27th International Symposium on Computer-Based Medical Systems. May 27-29, 2014, New York, NY, USA. IEEE, 2014: 185-188. doi:10.1109/CBMS.2014.54.
[39] Guillon JP. Non-invasive Tearscope Plus routine for contact lens fitting[J]. Cont Lens Anterior Eye, 1998, 21(Suppl 1): S31-S40. doi:10.1016/s1367-0484(98)80035-0.
[40] Remeseiro B, Bolon-Canedo V, Peteiro-Barral D, et al. A methodology for improving tear film lipid layer classification[J]. IEEE J Biomed Health Inform, 2014, 18(4): 1485-1493. doi:10.1109/JBHI.2013.2294732.
[41] Peteiro-Barral D, Remeseiro B, Méndez R, et al. Evaluation of an automatic dry eye test using MCDM methods and rank correlation[J]. Med Biol Eng Comput, 2017, 55(4): 527-536. doi:10.1007/s11517-016-1534-5.
[42] González-Domínguez J, Remeseiro B, Martín MJ. Parallel definition of tear film maps on distributed-memory clusters for the support of dry eye diagnosis[J]. Comput Methods Programs Biomed, 2017, 139: 51-60. doi:10.1016/j.cmpb.2016.10.027.
[43] Edorh NA, El Maftouhi A, Djerada Z, et al. New model to better diagnose dry eye disease integrating OCT corneal epithelial mapping[J]. Br J Ophthalmol, 2021: bjophthalmol-bjophtha2021-318826. doi:10.1136/bjophthalmol-2021-318826.
[44] Bron AJ, de Paiva CS, Chauhan SK, et al. TFOS DEWS II pathophysiology report[J]. Ocul Surf, 2017, 15(3): 438-510. doi:10.1016/j.jtos.2017.05.011.
[45] Maruoka S, Tabuchi H, Nagasato D, et al. Deep neural network-based method for detecting obstructive meibomian gland dysfunction with in vivo laser confocal microscopy[J]. Cornea, 2020, 39(6): 720-725. doi:10.1097/ICO.0000000000002279.
[46] 周奕文, 于薏, 周亚标, 等. 睑板腺缺失面积的图像深度处理分析研究[J]. 中华眼科杂志, 2020, 56(10): 774-779. doi:10.3760/cma.j.cn112142-20200415-00272. ZHOU Yiwen, YU Yi, ZHOU Yabiao, et al. An advanced imaging method for measuring and assessing meibomian glands based on deep learning[J]. Chin J Ophthalmol, 2020, 56(10): 774-779. doi:10.3760/cma.j.cn112142-20200415-00272.
[47] Lv J, Zhang K, Chen Q, et al. Deep learning-based automated diagnosis of fungal keratitis with in vivo confocal microscopy images[J]. Ann Transl Med, 2020, 8(11): 706. doi:10.21037/atm.2020.03.134.
[48] Gu H, Guo YW, Gu L, et al. Deep learning for identifying corneal diseases from ocular surface slit-lamp photographs[J]. Sci Rep, 2020, 10(1): 17851. doi:10.1038/s41598-020-75027-3.
[49] Hung N, Shih AK, Lin C, et al. Using slit-lamp images for deep learning-based identification of bacterial and fungal keratitis: model development and validation with different convolutional neural networks[J]. Diagnostics(Basel), 2021, 11(7): 1246. doi:10.3390/diagnostics11071246.
[50] Li Z, Jiang J, Chen K, et al. Preventing corneal blindness caused by keratitis using artificial intelligence[J]. Nat Commun, 2021, 12(1): 3738. doi:10.1038/s41467-021-24116-6.
[51] Li ZW, Jiang JW, Chen K, et al. Development of a deep learning-based image quality control system to detect and filter out ineligible slit-lamp images: a multicenter study[J]. Comput Methods Programs Biomed, 2021, 203: 106048. doi:10.1016/j.cmpb.2021.106048.
[52] Achiron A, Gur Z, Aviv U, et al. Predicting refractive surgery outcome: machine learning approach with big data[J]. J Refract Surg, 2017, 33(9): 592-597. doi:10.3928/1081597X-20170616-03.
[53] Vigueras-Guillen JP, van Rooij J, Lemij HG, et al. Convolutional neural network-based regression for biomarker estimation in corneal endothelium microscopy images[J]. Annu Int Conf IEEE Eng Med Biol Soc, 2019, 2019: 876-881. doi:10.1109/EMBC.2019.8857201.
[54] Heslinga FG, Alberti M, Pluim JPW, et al. Quantifying graft detachment after descemet's membrane endothelial keratoplasty with deep convolutional neural networks[J]. Transl Vis Sci Technol, 2020, 9(2): 48. doi:10.1167/tvst.9.2.48.
[55] Tong Y, Lu W, Yu Y, et al. Application of machine learning in ophthalmic imaging modalities[J]. Eye Vis(Lond), 2020, 7: 22. doi:10.1186/s40662-020-00183-6.
[56] Nuzzi R, Boscia G, Marolo P, et al. The impact of artificial intelligence and deep learning in eye diseases: a review[J]. Front Med(Lausanne), 2021, 8: 710329. doi:10.3389/fmed.2021.710329.
[57] Zhang K, Liu XY, Liu F, et al. An interpretable and expandable deep learning diagnostic system for multiple ocular diseases: qualitative study[J]. J Med Internet Res, 2018, 20(11): e11144. doi:10.2196/11144.
[58] 李东芳, 董燕玲, 谢森, 等. 基于深度学习的AS-OCT图像分析系统构建及其在角膜病变辅助诊断中的应用[J]. 中华眼科杂志, 2021, 57(6): 447-453. doi:10.3760/cma.j.cn112142-20200818-00540. LI Dongfang, DONG Yanling, XIE Sen, et al. Deep learning based lesion detection from anterior segment optical coherence tomography images and its application in the diagnosis of keratoconus[J]. Chin J Ophthalmol, 2021, 57(6): 447-453. doi:10.3760/cma.j.cn112142-20200818-00540.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed