Meibomian Gland Density: An Effective Evaluation Index of Meibomian Gland Dysfunction Based on Deep Learning and Transfer Learning
We aimed to establish an artificial intelligence (AI) system based on deep learning and transfer learning for meibomian gland (MG) segmentation and evaluate the efficacy of MG density in the diagnosis of MG dysfunction (MGD). First, 85 eyes of 85 subjects were enrolled for AI system-based evaluation...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-04-01
|
| Series: | Journal of Clinical Medicine |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2077-0383/11/9/2396 |
| _version_ | 1797504181929508864 |
|---|---|
| author | Zuhui Zhang Xiaolei Lin Xinxin Yu Yana Fu Xiaoyu Chen Weihua Yang Qi Dai |
| author_facet | Zuhui Zhang Xiaolei Lin Xinxin Yu Yana Fu Xiaoyu Chen Weihua Yang Qi Dai |
| author_sort | Zuhui Zhang |
| collection | DOAJ |
| description | We aimed to establish an artificial intelligence (AI) system based on deep learning and transfer learning for meibomian gland (MG) segmentation and evaluate the efficacy of MG density in the diagnosis of MG dysfunction (MGD). First, 85 eyes of 85 subjects were enrolled for AI system-based evaluation effectiveness testing. Then, from 2420 randomly selected subjects, 4006 meibography images (1620 upper eyelids and 2386 lower eyelids) graded by three experts according to the meiboscore were analyzed for MG density using the AI system. The updated AI system achieved 92% accuracy (intersection over union, IoU) and 100% repeatability in MG segmentation after 4 h of training. The processing time for each meibography was 100 ms. We discovered a significant and linear correlation between MG density and ocular surface disease index questionnaire (OSDI), tear break-up time (TBUT), lid margin score, meiboscore, and meibum expressibility score (all <i>p</i> < 0.05). The area under the curve (AUC) was 0.900 for MG density in the total eyelids. The sensitivity and specificity were 88% and 81%, respectively, at a cutoff value of 0.275. MG density is an effective index for MGD, particularly supported by the AI system, which could replace the meiboscore, significantly improve the accuracy of meibography analysis, reduce the analysis time and doctors’ workload, and improve the diagnostic efficiency. |
| first_indexed | 2024-03-10T04:01:59Z |
| format | Article |
| id | doaj.art-00188fd28b064d0ea76dbebaad23bab9 |
| institution | Directory Open Access Journal |
| issn | 2077-0383 |
| language | English |
| last_indexed | 2024-03-10T04:01:59Z |
| publishDate | 2022-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Clinical Medicine |
| spelling | doaj.art-00188fd28b064d0ea76dbebaad23bab92023-11-23T08:31:44ZengMDPI AGJournal of Clinical Medicine2077-03832022-04-01119239610.3390/jcm11092396Meibomian Gland Density: An Effective Evaluation Index of Meibomian Gland Dysfunction Based on Deep Learning and Transfer LearningZuhui Zhang0Xiaolei Lin1Xinxin Yu2Yana Fu3Xiaoyu Chen4Weihua Yang5Qi Dai6School of Ophthalmology and Optometry, The Eye Hospital of Wenzhou Medical University, 270 Xueyuanxi Road, Wenzhou 325027, ChinaDepartment of Ophthalmology and Visual Science, Eye, Ear, Nose, and Throat Hospital, Shanghai Medical College, Fudan University, Shanghai 200126, ChinaSchool of Ophthalmology and Optometry, The Eye Hospital of Wenzhou Medical University, 270 Xueyuanxi Road, Wenzhou 325027, ChinaSchool of Ophthalmology and Optometry, The Eye Hospital of Wenzhou Medical University, 270 Xueyuanxi Road, Wenzhou 325027, ChinaSchool of Ophthalmology and Optometry, The Eye Hospital of Wenzhou Medical University, 270 Xueyuanxi Road, Wenzhou 325027, ChinaAffiliated Eye Hospital, Nanjing Medical University, No.138 Hanzhong Road, Nanjing 210029, ChinaSchool of Ophthalmology and Optometry, The Eye Hospital of Wenzhou Medical University, 270 Xueyuanxi Road, Wenzhou 325027, ChinaWe aimed to establish an artificial intelligence (AI) system based on deep learning and transfer learning for meibomian gland (MG) segmentation and evaluate the efficacy of MG density in the diagnosis of MG dysfunction (MGD). First, 85 eyes of 85 subjects were enrolled for AI system-based evaluation effectiveness testing. Then, from 2420 randomly selected subjects, 4006 meibography images (1620 upper eyelids and 2386 lower eyelids) graded by three experts according to the meiboscore were analyzed for MG density using the AI system. The updated AI system achieved 92% accuracy (intersection over union, IoU) and 100% repeatability in MG segmentation after 4 h of training. The processing time for each meibography was 100 ms. We discovered a significant and linear correlation between MG density and ocular surface disease index questionnaire (OSDI), tear break-up time (TBUT), lid margin score, meiboscore, and meibum expressibility score (all <i>p</i> < 0.05). The area under the curve (AUC) was 0.900 for MG density in the total eyelids. The sensitivity and specificity were 88% and 81%, respectively, at a cutoff value of 0.275. MG density is an effective index for MGD, particularly supported by the AI system, which could replace the meiboscore, significantly improve the accuracy of meibography analysis, reduce the analysis time and doctors’ workload, and improve the diagnostic efficiency.https://www.mdpi.com/2077-0383/11/9/2396meibomian gland dysfunctionmeibomian gland densitydeep learningtransfer learningartificial intelligence |
| spellingShingle | Zuhui Zhang Xiaolei Lin Xinxin Yu Yana Fu Xiaoyu Chen Weihua Yang Qi Dai Meibomian Gland Density: An Effective Evaluation Index of Meibomian Gland Dysfunction Based on Deep Learning and Transfer Learning Journal of Clinical Medicine meibomian gland dysfunction meibomian gland density deep learning transfer learning artificial intelligence |
| title | Meibomian Gland Density: An Effective Evaluation Index of Meibomian Gland Dysfunction Based on Deep Learning and Transfer Learning |
| title_full | Meibomian Gland Density: An Effective Evaluation Index of Meibomian Gland Dysfunction Based on Deep Learning and Transfer Learning |
| title_fullStr | Meibomian Gland Density: An Effective Evaluation Index of Meibomian Gland Dysfunction Based on Deep Learning and Transfer Learning |
| title_full_unstemmed | Meibomian Gland Density: An Effective Evaluation Index of Meibomian Gland Dysfunction Based on Deep Learning and Transfer Learning |
| title_short | Meibomian Gland Density: An Effective Evaluation Index of Meibomian Gland Dysfunction Based on Deep Learning and Transfer Learning |
| title_sort | meibomian gland density an effective evaluation index of meibomian gland dysfunction based on deep learning and transfer learning |
| topic | meibomian gland dysfunction meibomian gland density deep learning transfer learning artificial intelligence |
| url | https://www.mdpi.com/2077-0383/11/9/2396 |
| work_keys_str_mv | AT zuhuizhang meibomianglanddensityaneffectiveevaluationindexofmeibomianglanddysfunctionbasedondeeplearningandtransferlearning AT xiaoleilin meibomianglanddensityaneffectiveevaluationindexofmeibomianglanddysfunctionbasedondeeplearningandtransferlearning AT xinxinyu meibomianglanddensityaneffectiveevaluationindexofmeibomianglanddysfunctionbasedondeeplearningandtransferlearning AT yanafu meibomianglanddensityaneffectiveevaluationindexofmeibomianglanddysfunctionbasedondeeplearningandtransferlearning AT xiaoyuchen meibomianglanddensityaneffectiveevaluationindexofmeibomianglanddysfunctionbasedondeeplearningandtransferlearning AT weihuayang meibomianglanddensityaneffectiveevaluationindexofmeibomianglanddysfunctionbasedondeeplearningandtransferlearning AT qidai meibomianglanddensityaneffectiveevaluationindexofmeibomianglanddysfunctionbasedondeeplearningandtransferlearning |