Artificial Intelligence to Detect Meibomian Gland Dysfunction From in-vivo Laser Confocal Microscopy

被引:12
|
作者
Zhang, Ye-Ye [1 ,2 ]
Zhao, Hui [3 ]
Lin, Jin-Yan [4 ]
Wu, Shi-Nan [5 ]
Liu, Xi-Wang [4 ,6 ]
Zhang, Hong-Dan [6 ]
Shao, Yi [5 ]
Yang, Wei-Feng [2 ,4 ,6 ]
机构
[1] Hainan Univ, Sch Sci, Dept Elect Engn, Haikou, Hainan, Peoples R China
[2] Shantou Univ, Coll Engn, Dept Elect Engn, Shantou, Peoples R China
[3] Shanghai Jiao Tong Univ, Shanghai Peoples Hosp 1, Natl Clin Res Ctr Eye Dis, Dept Ophthalmol, Shanghai, Peoples R China
[4] Shantou Univ, Coll Sci, Res Ctr Adv Opt & Photoelectron, Dept Phys, Shantou, Peoples R China
[5] Nanchang Univ, Affiliated Hosp 1, Jiangxi Ctr Natl Ophthalmol Clin Res Ctr, Dept Ophthalmol, Nanchang, Jiangxi, Peoples R China
[6] Shantou Univ, Coll Sci, Dept Math, Shantou, Peoples R China
关键词
deep learning; meibomian gland dysfunction; convolution neural network; in-vivo confocal microscopy; DenseNet CNN; INTERNATIONAL WORKSHOP; DRY EYE; PREDICTION; DISEASE;
D O I
10.3389/fmed.2021.774344
中图分类号
R5 [内科学];
学科分类号
1002 ; 100201 ;
摘要
Background: In recent years, deep learning has been widely used in a variety of ophthalmic diseases. As a common ophthalmic disease, meibomian gland dysfunction (MGD) has a unique phenotype in in-vivo laser confocal microscope imaging (VLCMI). The purpose of our study was to investigate a deep learning algorithm to differentiate and classify obstructive MGD (OMGD), atrophic MGD (AMGD) and normal groups.Methods: In this study, a multi-layer deep convolution neural network (CNN) was trained using VLCMI from OMGD, AMGD and healthy subjects as verified by medical experts. The automatic differential diagnosis of OMGD, AMGD and healthy people was tested by comparing its image-based identification of each group with the medical expert diagnosis. The CNN was trained and validated with 4,985 and 1,663 VLCMI images, respectively. By using established enhancement techniques, 1,663 untrained VLCMI images were tested.Results: In this study, we included 2,766 healthy control VLCMIs, 2,744 from OMGD and 2,801 from AMGD. Of the three models, differential diagnostic accuracy of the DenseNet169 CNN was highest at over 97%. The sensitivity and specificity of the DenseNet169 model for OMGD were 88.8 and 95.4%, respectively; and for AMGD 89.4 and 98.4%, respectively.Conclusion: This study described a deep learning algorithm to automatically check and classify VLCMI images of MGD. By optimizing the algorithm, the classifier model displayed excellent accuracy. With further development, this model may become an effective tool for the differential diagnosis of MGD.
引用
收藏
页数:8
相关论文
共 50 条
  • [1] The application of in vivo laser confocal microscopy to the diagnosis and evaluation of meibomian gland dysfunction
    Matsumoto, Yukihiro
    Sato, Enrique Adan
    Ibrahim, Osama M. A.
    Dogru, Murat
    Tsubota, Kazuo
    MOLECULAR VISION, 2008, 14 (149-51): : 1263 - 1271
  • [2] The Efficacy, Sensitivity, and Specificity of In Vivo Laser Confocal Microscopy in the Diagnosis of Meibomian Gland Dysfunction
    Ibrahim, Osama M. A.
    Matsumoto, Yukihiro
    Dogru, Murat
    Adan, Enrique Sato
    Wakamatsu, Tais Hitomi
    Goto, Tateki
    Negishi, Kazuno
    Tsubota, Kazuo
    OPHTHALMOLOGY, 2010, 117 (04) : 665 - 672
  • [3] Correction: In vivo confocal microscopy classification in the diagnosis of meibomian gland dysfunction
    Matthieu Randon
    Vittoria Aragno
    Rachid Abbas
    Hong Liang
    Antoine Labbé
    Christophe Baudouin
    Eye, 2019, 33 : 860 - 860
  • [4] In Vivo Confocal Microscopy in Different Types of Dry Eye and Meibomian Gland Dysfunction
    Sim, Ralene
    Yong, Kenneth
    Liu, Yu-Chi
    Tong, Louis
    JOURNAL OF CLINICAL MEDICINE, 2022, 11 (09)
  • [5] In Vivo Confocal Microscopy Evaluation of Meibomian Gland Dysfunction in Atopic-Keratoconjunctivitis Patients
    Ibrahim, Osama M. A.
    Matsumoto, Yukihiro
    Dogru, Murat
    Adan, Enrique Sato
    Wakamatsu, Tais Hitomi
    Shimazaki, Jun
    Fujishima, Hiroshi
    Tsubota, Kazuo
    OPHTHALMOLOGY, 2012, 119 (10) : 1961 - 1968
  • [6] Artificial intelligence in ex vivo confocal laser scanning microscopy
    Hartmann, Daniela
    HAUTARZT, 2021, 72 (12): : 1066 - 1070
  • [7] The evaluation of the treatment response in obstructive meibomian gland disease by in vivo laser confocal microscopy
    Yukihiro Matsumoto
    Yuta Shigeno
    Enrique Adan Sato
    Osama M. A. Ibrahim
    Megumi Saiki
    Kazuno Negishi
    Yoko Ogawa
    Murat Dogru
    Kazuo Tsubota
    Graefe's Archive for Clinical and Experimental Ophthalmology, 2009, 247 : 821 - 829
  • [8] The evaluation of the treatment response in obstructive meibomian gland disease by in vivo laser confocal microscopy
    Matsumoto, Yukihiro
    Shigeno, Yuta
    Sato, Enrique Adan
    Ibrahim, Osama M. A.
    Saiki, Megumi
    Negishi, Kazuno
    Ogawa, Yoko
    Dogru, Murat
    Tsubota, Kazuo
    GRAEFES ARCHIVE FOR CLINICAL AND EXPERIMENTAL OPHTHALMOLOGY, 2009, 247 (06) : 821 - 829
  • [9] Deep Neural Network-Based Method for Detecting Obstructive Meibomian Gland Dysfunction With in Vivo Laser Confocal Microscopy
    Maruoka, Sachiko
    Tabuchi, Hitoshi
    Nagasato, Daisuke
    Masumoto, Hiroki
    Chikama, Taiichiro
    Kawai, Akiko
    Oishi, Naoko
    Maruyama, Toshi
    Kato, Yoshitake
    Hayashi, Takahiko
    Katakami, Chikako
    CORNEA, 2020, 39 (06) : 720 - 725
  • [10] A new classification for meibomian gland diseases with in vivo confocal microscopy
    Randon, M.
    Liang, H.
    Abbas, R.
    Michee, S.
    Denoyer, A.
    Baudouin, C.
    Labbe, A.
    JOURNAL FRANCAIS D OPHTALMOLOGIE, 2016, 39 (03): : 239 - 247