Yi Yu, Yiwen Zhou1, Hongmei Zheng2, Yanning Yang3, Miao Tian1, Yabiao Zhou4, Yuejiao Tan4, Lianlian Wu5. 1. Eye Center of Renmin Hospital of Wuhan University, 99 Zhangzhidong Road, Wuhan, 430060, Hubei Province, China. 2. Eye Center of Renmin Hospital of Wuhan University, 99 Zhangzhidong Road, Wuhan, 430060, Hubei Province, China. 13871484442@139.com. 3. Eye Center of Renmin Hospital of Wuhan University, 99 Zhangzhidong Road, Wuhan, 430060, Hubei Province, China. ophyyn@163.com. 4. School of Resources and Environmental Sciences of Wuhan University, Wuhan, China. 5. Department of Gastroenterology, Renmin Hospital of Wuhan University, Wuhan, China.
Abstract
BACKGROUND: Artificial intelligence is developing rapidly, bringing increasing numbers of intelligent products into daily life. However, it has little progress in dry eye, which is a common disease and associated with meibomian gland dysfunction (MGD). Noninvasive infrared meibography, known as an effective diagnostic tool of MGD, allows for objective observation of meibomian glands. Thus, we discuss a deep learning method to measure and assess meibomian glands of meibography. METHODS: We used Mask R-CNN deep learning (DL) framework. A total of 1878 meibography images were collected and manually annotated by two licensed eyelid specialists with two classes: conjunctiva and meibomian glands. The annotated pictures were used to establish a DL model. An independent test dataset that contained 58 images was used to compare the accuracy and efficiency of the deep learning model with specialists. RESULTS: The DL model calculated the ratio of meibomian gland loss with precise values by achieving high accuracy in the identification of conjunctiva (validation loss < 0.35, mAP > 0.976) and meibomian glands (validation loss < 1.0, mAP > 0.92). The comparison between specialists' annotation and the DL model evaluation showed that there is little difference between the gold standard and the model. Each image takes 480 ms for the model to evaluate, almost 21 times faster than specialists. CONCLUSIONS: The DL model can improve the accuracy of meibography image evaluation, help specialists to grade the meibomian glands and save their time to some extent.
BACKGROUND: Artificial intelligence is developing rapidly, bringing increasing numbers of intelligent products into daily life. However, it has little progress in dry eye, which is a common disease and associated with meibomian gland dysfunction (MGD). Noninvasive infrared meibography, known as an effective diagnostic tool of MGD, allows for objective observation of meibomian glands. Thus, we discuss a deep learning method to measure and assess meibomian glands of meibography. METHODS: We used Mask R-CNN deep learning (DL) framework. A total of 1878 meibography images were collected and manually annotated by two licensed eyelid specialists with two classes: conjunctiva and meibomian glands. The annotated pictures were used to establish a DL model. An independent test dataset that contained 58 images was used to compare the accuracy and efficiency of the deep learning model with specialists. RESULTS: The DL model calculated the ratio of meibomian gland loss with precise values by achieving high accuracy in the identification of conjunctiva (validation loss < 0.35, mAP > 0.976) and meibomian glands (validation loss < 1.0, mAP > 0.92). The comparison between specialists' annotation and the DL model evaluation showed that there is little difference between the gold standard and the model. Each image takes 480 ms for the model to evaluate, almost 21 times faster than specialists. CONCLUSIONS: The DL model can improve the accuracy of meibography image evaluation, help specialists to grade the meibomian glands and save their time to some extent.
Authors: J Daniel Nelson; Jun Shimazaki; Jose M Benitez-del-Castillo; Jennifer P Craig; James P McCulley; Seika Den; Gary N Foulks Journal: Invest Ophthalmol Vis Sci Date: 2011-03-30 Impact factor: 4.799
Authors: Ji-Peng Olivia Li; Hanruo Liu; Darren S J Ting; Sohee Jeon; R V Paul Chan; Judy E Kim; Dawn A Sim; Peter B M Thomas; Haotian Lin; Youxin Chen; Taiji Sakomoto; Anat Loewenstein; Dennis S C Lam; Louis R Pasquale; Tien Y Wong; Linda A Lam; Daniel S W Ting Journal: Prog Retin Eye Res Date: 2020-09-06 Impact factor: 21.198