Segmentation of meibomian glands based on deep learning
AIM: To explore the application value of deep learning technology in automatic meibomian glands segmentation. METHODS:Infrared meibomian gland images were collected and 193 of them were picked out for establishing the database. The images were manually labeled by three clinicians. UNet++ network and...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Press of International Journal of Ophthalmology (IJO PRESS)
2022-07-01
|
| Series: | Guoji Yanke Zazhi |
| Subjects: | |
| Online Access: | http://ies.ijo.cn/cn_publish/2022/7/202207025.pdf |
| _version_ | 1818547250398232576 |
|---|---|
| author | Jia-Wen Lin Zhi-Ming Lin Tai-Chen Lai Lin-Ling Guo Jing Zou Li Li |
| author_facet | Jia-Wen Lin Zhi-Ming Lin Tai-Chen Lai Lin-Ling Guo Jing Zou Li Li |
| author_sort | Jia-Wen Lin |
| collection | DOAJ |
| description | AIM: To explore the application value of deep learning technology in automatic meibomian glands segmentation. METHODS:Infrared meibomian gland images were collected and 193 of them were picked out for establishing the database. The images were manually labeled by three clinicians. UNet++ network and automatic data expansion strategy were introduced to construct the automatic meibomian glands segmentation model. The feasibility and effectiveness of the proposed segmentation model were analyzed by precision, sensitivity, specificity, accuracy and intersection over union.RESULTS: Taking manual labeling as the gold standard, the presented method segment the glands effectively and steadily with accuracy, sensitivity and specificity of 94.31%, 82.15% and 96.13% respectively. On the average, only 0.11s was taken for glands segmentation of single image.CONCLUSIONS: In this paper, deep learning technology is introduced to realize automatic segmentation of meibomian glands, achieving high accuracy, good stability and efficiency. It would be quite useful for calculation of gland morphological parameters, the clinical diagnosis and screening of related diseases, improving the diagnostic efficiency. |
| first_indexed | 2024-12-12T08:04:15Z |
| format | Article |
| id | doaj.art-081261208f304b8abbdab0fcea31ad7d |
| institution | Directory Open Access Journal |
| issn | 1672-5123 |
| language | English |
| last_indexed | 2024-12-12T08:04:15Z |
| publishDate | 2022-07-01 |
| publisher | Press of International Journal of Ophthalmology (IJO PRESS) |
| record_format | Article |
| series | Guoji Yanke Zazhi |
| spelling | doaj.art-081261208f304b8abbdab0fcea31ad7d2022-12-22T00:32:01ZengPress of International Journal of Ophthalmology (IJO PRESS)Guoji Yanke Zazhi1672-51232022-07-012271191119410.3980/j.issn.1672-5123.2022.7.25202207025Segmentation of meibomian glands based on deep learningJia-Wen Lin0Zhi-Ming Lin1Tai-Chen Lai2Lin-Ling Guo3Jing Zou4Li Li5College of Computer and Data Science, Fuzhou University, Fuzhou 350108, Fujian Province, China; Fujian Provincial Key Laboratory of Networking Computing and Intelligent Information Processing, Fuzhou University, Fuzhou 350108, Fujian Province, ChinaCollege of Computer and Data Science, Fuzhou University, Fuzhou 350108, Fujian Province, China; Fujian Provincial Key Laboratory of Networking Computing and Intelligent Information Processing, Fuzhou University, Fuzhou 350108, Fujian Province, ChinaSchool of Basic Medical Sciences, Fujian Medical University, Fuzhou 350108, Fujian Province, ChinaSchool of Basic Medical Sciences, Fujian Medical University, Fuzhou 350108, Fujian Province, ChinaSchool of Basic Medical Sciences, Fujian Medical University, Fuzhou 350108, Fujian Province, ChinaDepartment of Ophthalmology, Fujian Provincial Hospital, Fuzhou 350002, Fujian Province, China; Department of Ophthalmology, Fujian Provincial Hospital South Branch, Fuzhou 350002, Fujian Province, ChinaAIM: To explore the application value of deep learning technology in automatic meibomian glands segmentation. METHODS:Infrared meibomian gland images were collected and 193 of them were picked out for establishing the database. The images were manually labeled by three clinicians. UNet++ network and automatic data expansion strategy were introduced to construct the automatic meibomian glands segmentation model. The feasibility and effectiveness of the proposed segmentation model were analyzed by precision, sensitivity, specificity, accuracy and intersection over union.RESULTS: Taking manual labeling as the gold standard, the presented method segment the glands effectively and steadily with accuracy, sensitivity and specificity of 94.31%, 82.15% and 96.13% respectively. On the average, only 0.11s was taken for glands segmentation of single image.CONCLUSIONS: In this paper, deep learning technology is introduced to realize automatic segmentation of meibomian glands, achieving high accuracy, good stability and efficiency. It would be quite useful for calculation of gland morphological parameters, the clinical diagnosis and screening of related diseases, improving the diagnostic efficiency.http://ies.ijo.cn/cn_publish/2022/7/202207025.pdfmeibomian gland dysfunctioninfrared meibomian gland imagesgland segmentationdeep learningunet++ |
| spellingShingle | Jia-Wen Lin Zhi-Ming Lin Tai-Chen Lai Lin-Ling Guo Jing Zou Li Li Segmentation of meibomian glands based on deep learning Guoji Yanke Zazhi meibomian gland dysfunction infrared meibomian gland images gland segmentation deep learning unet++ |
| title | Segmentation of meibomian glands based on deep learning |
| title_full | Segmentation of meibomian glands based on deep learning |
| title_fullStr | Segmentation of meibomian glands based on deep learning |
| title_full_unstemmed | Segmentation of meibomian glands based on deep learning |
| title_short | Segmentation of meibomian glands based on deep learning |
| title_sort | segmentation of meibomian glands based on deep learning |
| topic | meibomian gland dysfunction infrared meibomian gland images gland segmentation deep learning unet++ |
| url | http://ies.ijo.cn/cn_publish/2022/7/202207025.pdf |
| work_keys_str_mv | AT jiawenlin segmentationofmeibomianglandsbasedondeeplearning AT zhiminglin segmentationofmeibomianglandsbasedondeeplearning AT taichenlai segmentationofmeibomianglandsbasedondeeplearning AT linlingguo segmentationofmeibomianglandsbasedondeeplearning AT jingzou segmentationofmeibomianglandsbasedondeeplearning AT lili segmentationofmeibomianglandsbasedondeeplearning |