Automatic screening of tear meniscus from lacrimal duct obstructions using anterior segment optical coherence tomography images by deep learning.
Anterior segment optical coherence tomography image
Area under the curve
Deep learning
Lacrimal duct obstruction
Tear meniscus
Journal
Graefe's archive for clinical and experimental ophthalmology = Albrecht von Graefes Archiv fur klinische und experimentelle Ophthalmologie
ISSN: 1435-702X
Titre abrégé: Graefes Arch Clin Exp Ophthalmol
Pays: Germany
ID NLM: 8205248
Informations de publication
Date de publication:
Jun 2021
Jun 2021
Historique:
received:
25
06
2020
accepted:
06
01
2021
revised:
23
11
2020
pubmed:
13
2
2021
medline:
19
8
2021
entrez:
12
2
2021
Statut:
ppublish
Résumé
We assessed the ability of deep learning (DL) models to distinguish between tear meniscus of lacrimal duct obstruction (LDO) patients and normal subjects using anterior segment optical coherence tomography (ASOCT) images. The study included 117 ASOCT images (19 men and 98 women; mean age, 66.6 ± 13.6 years) from 101 LDO patients and 113 ASOCT images (29 men and 84 women; mean age, 38.3 ± 19.9 years) from 71 normal subjects. We trained to construct 9 single and 502 ensemble DL models with 9 different network structures, and calculated the area under the curve (AUC), sensitivity, and specificity to compare the distinguishing abilities of these single and ensemble DL models. For the highest single DL model (DenseNet169), the AUC, sensitivity, and specificity for distinguishing LDO were 0.778, 64.6%, and 72.1%, respectively. For the highest ensemble DL model (VGG16, ResNet50, DenseNet121, DenseNet169, InceptionResNetV2, InceptionV3, and Xception), the AUC, sensitivity, and specificity for distinguishing LDO were 0.824, 84.8%, and 58.8%, respectively. The heat maps indicated that these DL models placed their focus on the tear meniscus region of the ASOCT images. The combination of DL and ASOCT images could distinguish between tear meniscus of LDO patients and normal subjects with a high level of accuracy. These results suggest that DL might be useful for automatic screening of patients for LDO.
Identifiants
pubmed: 33576859
doi: 10.1007/s00417-021-05078-3
pii: 10.1007/s00417-021-05078-3
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
1569-1577Références
Mishima S, Gasset A, Klyce SD Jr et al (1966) Determination of tear volume and tear flow. Investig Ophthalmol 5:264–276
Maurice DM (1973) The dynamics and drainage of tears. Int Ophthalmol Clin 13:103–116
pubmed: 4724254
Shen M, Li J, Wang J et al (2009) Upper and lower tear menisci in the diagnosis of dry eye. Invest Ophthalmol Vis Sci 50:2722–2726
pubmed: 19218609
Roh JH, Chi MJ (2010) Efficacy of dye disappearance test and tear meniscus height in diagnosis and postoperative assessment of nasolacrimal duct obstruction. Acta Ophthalmol 88:73–77
Thomas R, Thomas S, Braganza A et al (1977) Evaluation of the role of irrigation prior to cataract surgery. Indian J Ophthalmol 45:211–214
Mahesh L (2014) Imaging in lacrimal surgery. In: Isloor S (ed) Lacrimal drainage surgery. Jaypee Brothers, New Delhi, pp 17–22
Galloway JE, Kavie TA, Raflo GT (1984) Digital subtraction macrodacryocystography: a new method of lacrimal system imaging. Ophthalmology 91:956–962
pubmed: 6093019
Sasaki T, Miyashita H, Miyanaga T et al (2012) Dacryoendoscopic observation and incidence of canalicular obstruction or stenosis associated with S-1, anoral anticancer drug. Jpn J Ophthalmol 56:214–218
pubmed: 22411166
Sasaki T, Nagata Y, Sugiyama K (2005) Nasolacrimal duct obstruction classified by dacryoendoscopy and treated with inferior meatal dacryorhinotomy. Part I: Positional diagnosis of primary nasolacrimal duct obstruction with dacryoendoscope. Am J Ophthalmol 140:1065–1069
pubmed: 16376652
Sasaki T, Nagata Y, Sugiyama K (2005) Nasolacrimal duct obstruction classified by dacryoendoscopy and treated with inferior meatal dacryorhinotomy: Part II. Inferior meatal dacryorhinotomy. Am J Ophthalmol 140:1070–1074
pubmed: 16376653
Ibrahim O, Dogru M, Takano Y et al (2010) Application of visante optical coherence tomography tear meniscus height measurement in the diagnosis of dry eye disease. Ophthalmology 117:1923–1929
pubmed: 20605216
Qiu X, Gong L, Sun X et al (2011) Age-related variations of human tear meniscus and diagnosis of dry eye with Fourier-domain anterior segment optical coherence tomography. Cornea 30:543–549
pubmed: 21107246
Czajkowski G, Kaluzny BJ, Laudencka A et al (2012) Tear meniscus measurement by spectral optical coherence tomography. Optom Vis Sci 89:336–342
pubmed: 22282222
Savini G, Goto E, Carbonelli M et al (2009) Agreement between stratus and visante optical coherence tomography systems in tear meniscus measurements. Cornea 28:148–151
pubmed: 19158555
Park DI, Lew H, Lee SY (2012) Tear meniscus measurement in nasolacrimal duct obstruction patients with Fourier-domain optical coherence tomography: novel three-point capture method. Acta Ophthalmol 90:783–787
pubmed: 21726426
Kim SE, Lee SJ, Lee SY et al (2012) Outcomes of 4-snip punctoplasty for severe punctal stenosis: measurement of tear meniscus height by optical coherence tomography. Am J Ophthalmol 153:769–773
pubmed: 22264691
Imamura H, Tabuchi H, Nakakura S et al (2018) Usability and reproducibility of tear meniscus values generated via swept-source optical coherence tomography and the slit lamp with a graticule method. Int Ophthalmol 38:679–686
pubmed: 28393321
Todoroki K, Nakano T, Ishii Y et al (2015) Automatic analyzer for highly polar carboxylic acids based on fluorescence derivatization-liquid chromatography. Biomed Chromatogr 29:445–451
pubmed: 25082081
Nagasato D, Tabuchi H, Masumoto H et al (2019) Automated detection of a nonperfusion area caused by retinal vein occlusion in optical coherence tomography angiography images using deep learning. PLoS One 14:e0223965
pubmed: 31697697
pmcid: 6837754
Masumoto H, Tabuchi H, Nakakura S et al (2019) Accuracy of a deep convolutional neural network in detection of retinitis pigmentosa on ultrawide-field images. PeerJ 7:e6900
pubmed: 31119087
pmcid: 6510218
Nagasawa T, Tabuchi H, Masumoto H et al (2019) Accuracy of ultrawide-field fundus ophthalmoscopy-assisted deep learning for detecting treatment-naïve proliferative diabetic retinopathy. Int Ophthalmol 39:2153–2159
pubmed: 30798455
Ohsugi H, Tabuchi H, Enno H et al (2017) Accuracy of deep learning, a machine-learning technology, using ultra–wide-field fundus ophthalmoscopy for detecting rhegmatogenous retinal detachment. Sci Rep 7:9425
pubmed: 28842613
pmcid: 5573327
Nagasato D, Tabuchi H, Ohsugi H et al (2018) Deep neural network-based method for detecting central retinal vein occlusion using ultrawide-field fundus ophthalmoscopy. J Ophthalmol 2018:1875431
Nagasawa T, Tabuchi H, Masumoto H et al (2018) Accuracy of deep learning, a machine-learning technology, using ultra–widefield fundus ophthalmoscopy for detecting idiopathic macular holes. PeerJ 6:e5696
pubmed: 30370184
pmcid: 6201738
Sonobe T, Tabuchi H, Ohsugi H et al (2019) Comparison between support vector machine and deep learning, machine-learning technologies for detecting epiretinal membrane using 3D-OCT. Int Ophthalmol 39:1871–1877
pubmed: 30218173
Matsuba S, Tabuchi H, Ohsugi H et al (2019) Accuracy of ultra–wide-field fundus ophthalmoscopy-assisted deep learning, a machine-learning technology, for detecting age related macular degeneration. Int Ophthalmol 39:1269–1275
pubmed: 29744763
Masumoto H, Tabuchi H, Nakakura S et al (2018) Deep-learning classifier with an ultrawide-field scanning laser ophthalmoscope detects glaucoma visual field severity. J Glaucoma 27:647–652
pubmed: 29781835
Grewal PS, Oloumi F, Rubin U et al (2018) Deep learning in ophthalmology: a review. Can J Ophthalmol 53:309–313
pubmed: 30119782
De Fauw J, Ledsam JR, Romera-Paredes B et al (2018) Clinically applicable deep learning for diagnosis and referral in retinal disease. Nat Med 24:1342–1350
pubmed: 30104768
Maruoka S, Tabuchi H, Nagasato D et al (2020) Deep neural network-based method for detecting obstructive meibomian gland dysfunction with in vivo laser confocal microscopy. Cornea. 2020. https://doi.org/10.1097/ICO.0000000000002279 [Epub ahead of print]
Sogawa T, Tabuchi H, Nagasato D et al (2020) Accuracy of a deep convolutional neural network in the detection of myopic macular diseases using swept-source optical coherence tomography. PLoS One 15(4):e0227240
Fu H, Baskaran M, Xu Y et al (2019) A deep learning system for automated angle-closure detection in anterior segment optical coherence tomography images. Am J Ophthalmol 203:37–45
pubmed: 30849350
Kamiya K, Ayatsuka Y, Kato Y et al (2019) Keratoconus detection using deep learning of colour-coded maps with anterior segment optical coherence tomography: a diagnostic accuracy study. BMJ Open 9:e031313
pubmed: 31562158
pmcid: 6773416
Mosteller F, Tukey JW (1968) Data analysis, including statistics. In: Lindzey G, Aronson E (eds) Handbook of Social Psychology, Research Methods, vol 2. Addison-Wesley, Reading, pp 80–203
Kohavi R (1995) A study of cross-validation and bootstrap for accuracy estimation and model selection. Proc Int Joint Conf AI 2:1137–1145
Simonyan K, Andrew Z (2014) Very deep convolutional networks for large-scale image recognition. Preprint at https://arxiv.org/pdf/1409.1556.pdf(2014 ). Accessed 1 Sept
He K, Zhang X, Ren S et al (2016) Deep residual learning for image recognition. Presented at the Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 770–778
Huang G, Liu Z, Van Der Maaten L et al (2017) Densely connected convolutional networks. CVPR 1:4700–4708
Szegedy C, Vanhoucke V, Ioffe S et al (2016) Rethinking the inception architecture for computer vision. Presented at the Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2818–2826
Szegedy C, Ioffe S, Vanhoucke V et al (2017) Inception-v4, inception-resnet and the impact of residual connections on learning. AAAI 4:4278–4284
Chollet F (2017) Deep learning with depthwise separable convolutions. Preprint at https://arxiv.org/pdf/1610.02357.pdf . Accessed 4 Apr
Srivastava N, Hinton G, Krizhevsky A et al (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 15:1929–1958
Agrawal P, Girshick R, Malik J (2014) Analyzing the performance of multilayer neural networks for object recognition. In: European Conference on Computer Vision. Springer, Cham, pp 329–344
Qian N (1999) On the momentum term in gradient descent learning algorithms. Neural Netw 12:145–151. https://doi.org/10.1016/S0893-6080(98)00116-6
doi: 10.1016/S0893-6080(98)00116-6
pubmed: 12662723
Nesterov Y (1983) A method for unconstrained convex minimization problem with the rate of convergence O (1/k^2). Proc USSR Acad Sci 269:543–547
Wang, H, Wang Z, Du M et al (2019) Score-CAM: Improved Visual Explanations Via Score-Weighted Class Activation Mapping. arXiv 1910.01279
Welch BL (1947) The generalization of student’s problem when several different population variances are involved. Biometrika 34.1(2):28–35
Fisher RA (1922) On the interpretation of χ2 from contingency tables, and the calculation of P. J R Stat Soc 851922:87–94
Hanley JA, McNeil BJ (1982) The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 143:29–36. https://doi.org/10.1148/radiology.143.1.7063747
doi: 10.1148/radiology.143.1.7063747
pubmed: 7063747
Clopper CJ, Egon SP (1934) The use of confidence or fiducial limits illustrated in the case of the binomial. Biometrika 26:404–413
Deng J, Dong W, Socher R et al (2009) ImageNet: a large-scale hierarchical image database. In: 2009 IEEE conference on computer vision and pattern recognition. IEEE, Miami, pp 248–255
Lee CY, Xie S, Gallagher P et al (2015) Deeply-supervised nets. In: Proceedings of the 18th International Conference on Artificial Intelligence and Statistics (AISTATS). Journal of Machine Learning Research Workshop and Conference Proceedings, San Diego, pp 562–570
Russakovsky O, Deng J, Su H, et al (2015) Imagenet large scale visual recognition challenge. Int J Comput Vision 2015(115):211–252
Hartikainen J, Lehtonen OP, Saari KM (1977) Bacteriology of lacrimal duct obstruction in adults. Br J Ophthalmol 81:37–40
Major JC Jr, Engelbert M, Flynn HW Jr et al (2010) Staphylococcus aureus endophthalmitis: antibiotic susceptibilities, methicillin resistance, and clinical outcomes. Am J Ophthalmol 149:278–283
pubmed: 19926069
Scott IU, Loo RH, Flynn HW Jr et al (2003) Endophthalmitis caused by enterococcus faecalis: antibiotic selection and treatment outcomes. Ophthalmology 110:1573–1577
pubmed: 12917175
Schmidt ME, Smith MA, Levy CS (1993) Endophthalmitis caused by unusual gram-negative bacilli: three case reports and review. Clin Infect Dis 17:686–690
pubmed: 8268350
Kam JK, Cheng NM, Sarossy M et al (2014) Nasolacrimal duct screening to minimize post-cataract surgery endophthalmitis. Clin Exp Ophthalmol 42:447–451
pubmed: 24118663
Lopez PF, Beldavs RA, al-Ghamdi S et al (1993) Pneumococcal endophthalmitis associated with nasolacrimal obstruction. Am J Ophthalmol 116:56–62
pubmed: 8328544
Sidoti PA, Lopez PF, Michon J et al (1995) Delayed-onset pneumococcal endophthalmitis after mitomycin-C trabeculectomy: association with cryptic nasolacrimal obstruction. J Glaucoma 4:11–15
pubmed: 19920631
Lamberts DW, Foster CS, Perry HD (1979) Schirmer test after topical anesthesia and the tear meniscus height in normal eyes. Arch Ophthalmol 97:1082–1085
pubmed: 444137
Doughty MJ, Laiquzzaman M, Oblak E et al (2002) The tear (lacrimal) meniscus height in human eyes: a useful clinical measure or an unusable variable sign? Cont Lens Anterior Eye 25:57–65
pubmed: 16303478