Multi-label classification of fundus images based on graph convolutional network
Abstract Background Diabetic Retinopathy (DR) is the most common and serious microvascular complication in the diabetic population. Using computer-aided diagnosis from the fundus images has become a method of detecting retinal diseases, but the detection of multiple lesions is still a difficult poin...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2021-07-01
|
| Series: | BMC Medical Informatics and Decision Making |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12911-021-01424-x |
| _version_ | 1818909153329938432 |
|---|---|
| author | Yinlin Cheng Mengnan Ma Xingyu Li Yi Zhou |
| author_facet | Yinlin Cheng Mengnan Ma Xingyu Li Yi Zhou |
| author_sort | Yinlin Cheng |
| collection | DOAJ |
| description | Abstract Background Diabetic Retinopathy (DR) is the most common and serious microvascular complication in the diabetic population. Using computer-aided diagnosis from the fundus images has become a method of detecting retinal diseases, but the detection of multiple lesions is still a difficult point in current research. Methods This study proposed a multi-label classification method based on the graph convolutional network (GCN), so as to detect 8 types of fundus lesions in color fundus images. We collected 7459 fundus images (1887 left eyes, 1966 right eyes) from 2282 patients (1283 women, 999 men), and labeled 8 types of lesions, laser scars, drusen, cup disc ratio ( $$C/D>0.6$$ C / D > 0.6 ), hemorrhages, retinal arteriosclerosis, microaneurysms, hard exudates and soft exudates. We constructed a specialized corpus of the related fundus lesions. A multi-label classification algorithm for fundus images was proposed based on the corpus, and the collected data were trained. Results The average overall F1 Score (OF1) and the average per-class F1 Score (CF1) of the model were 0.808 and 0.792 respectively. The area under the ROC curve (AUC) of our proposed model reached 0.986, 0.954, 0.946, 0.957, 0.952, 0.889, 0.937 and 0.926 for detecting laser scars, drusen, cup disc ratio, hemorrhages, retinal arteriosclerosis, microaneurysms, hard exudates and soft exudates, respectively. Conclusions Our results demonstrated that our proposed model can detect a variety of lesions in the color images of the fundus, which lays a foundation for assisting doctors in diagnosis and makes it possible to carry out rapid and efficient large-scale screening of fundus lesions. |
| first_indexed | 2024-12-19T22:22:23Z |
| format | Article |
| id | doaj.art-0058d927845f4434af4cb469a5a2e718 |
| institution | Directory Open Access Journal |
| issn | 1472-6947 |
| language | English |
| last_indexed | 2024-12-19T22:22:23Z |
| publishDate | 2021-07-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Medical Informatics and Decision Making |
| spelling | doaj.art-0058d927845f4434af4cb469a5a2e7182022-12-21T20:03:35ZengBMCBMC Medical Informatics and Decision Making1472-69472021-07-0121S21910.1186/s12911-021-01424-xMulti-label classification of fundus images based on graph convolutional networkYinlin Cheng0Mengnan Ma1Xingyu Li2Yi Zhou3School of Biomedical Engineering, Sun Yat-sen UniversitySchool of Biomedical Engineering, Sun Yat-sen UniversityDepartment of Medical Informatics, Zhongshan School of Medicine, Sun Yat-sen UniversityDepartment of Medical Informatics, Zhongshan School of Medicine, Sun Yat-sen UniversityAbstract Background Diabetic Retinopathy (DR) is the most common and serious microvascular complication in the diabetic population. Using computer-aided diagnosis from the fundus images has become a method of detecting retinal diseases, but the detection of multiple lesions is still a difficult point in current research. Methods This study proposed a multi-label classification method based on the graph convolutional network (GCN), so as to detect 8 types of fundus lesions in color fundus images. We collected 7459 fundus images (1887 left eyes, 1966 right eyes) from 2282 patients (1283 women, 999 men), and labeled 8 types of lesions, laser scars, drusen, cup disc ratio ( $$C/D>0.6$$ C / D > 0.6 ), hemorrhages, retinal arteriosclerosis, microaneurysms, hard exudates and soft exudates. We constructed a specialized corpus of the related fundus lesions. A multi-label classification algorithm for fundus images was proposed based on the corpus, and the collected data were trained. Results The average overall F1 Score (OF1) and the average per-class F1 Score (CF1) of the model were 0.808 and 0.792 respectively. The area under the ROC curve (AUC) of our proposed model reached 0.986, 0.954, 0.946, 0.957, 0.952, 0.889, 0.937 and 0.926 for detecting laser scars, drusen, cup disc ratio, hemorrhages, retinal arteriosclerosis, microaneurysms, hard exudates and soft exudates, respectively. Conclusions Our results demonstrated that our proposed model can detect a variety of lesions in the color images of the fundus, which lays a foundation for assisting doctors in diagnosis and makes it possible to carry out rapid and efficient large-scale screening of fundus lesions.https://doi.org/10.1186/s12911-021-01424-xDiabetic retinopathyFundus imagesGCNMulti-label |
| spellingShingle | Yinlin Cheng Mengnan Ma Xingyu Li Yi Zhou Multi-label classification of fundus images based on graph convolutional network BMC Medical Informatics and Decision Making Diabetic retinopathy Fundus images GCN Multi-label |
| title | Multi-label classification of fundus images based on graph convolutional network |
| title_full | Multi-label classification of fundus images based on graph convolutional network |
| title_fullStr | Multi-label classification of fundus images based on graph convolutional network |
| title_full_unstemmed | Multi-label classification of fundus images based on graph convolutional network |
| title_short | Multi-label classification of fundus images based on graph convolutional network |
| title_sort | multi label classification of fundus images based on graph convolutional network |
| topic | Diabetic retinopathy Fundus images GCN Multi-label |
| url | https://doi.org/10.1186/s12911-021-01424-x |
| work_keys_str_mv | AT yinlincheng multilabelclassificationoffundusimagesbasedongraphconvolutionalnetwork AT mengnanma multilabelclassificationoffundusimagesbasedongraphconvolutionalnetwork AT xingyuli multilabelclassificationoffundusimagesbasedongraphconvolutionalnetwork AT yizhou multilabelclassificationoffundusimagesbasedongraphconvolutionalnetwork |