Evaluation of a computer-aided diagnosis system for diabetic retinopathy screening on public data
- PMID: 21527381
- DOI: 10.1167/iovs.10-6633
Evaluation of a computer-aided diagnosis system for diabetic retinopathy screening on public data
Abstract
Purpose: To evaluate the performance of a comprehensive computer-aided diagnosis (CAD) system for diabetic retinopathy (DR) screening, using a publicly available database of retinal images, and to compare its performance with that of human experts.
Methods: A previously developed, comprehensive DR CAD system was applied to 1200 digital color fundus photographs (nonmydriatic camera, single field) of 1200 eyes in the publicly available Messidor dataset (Methods to Evaluate Segmentation and Indexing Techniques in the Field of Retinal Ophthalmology (http://messidor.crihan.fr). The ability of the system to distinguish normal images from those with DR was determined by using receiver operator characteristic (ROC) analysis. Two experts also determined the presence of DR in each of the images.
Results: The system achieved an area under the ROC curve of 0.876 for successfully distinguishing normal images from those with DR with a sensitivity of 92.2% at a specificity of 50%. These compare favorably with the two experts, who achieved sensitivities of 94.5% and 91.2% at a specificity of 50%.
Conclusions: This study shows, for the first time, the performance of a comprehensive DR screening system on an independent, publicly available dataset. The performance of the system on this dataset is comparable with that of human experts.
Similar articles
-
Nonmydriatic ultrawide field retinal imaging compared with dilated standard 7-field 35-mm photography and retinal specialist examination for evaluation of diabetic retinopathy.Am J Ophthalmol. 2012 Sep;154(3):549-559.e2. doi: 10.1016/j.ajo.2012.03.019. Epub 2012 May 23. Am J Ophthalmol. 2012. PMID: 22626617
-
Reliability of screening methods for diabetic retinopathy.Diabet Med. 2009 Aug;26(8):783-90. doi: 10.1111/j.1464-5491.2009.02778.x. Diabet Med. 2009. PMID: 19709148 Review.
-
Information fusion for diabetic retinopathy CAD in digital color fundus photographs.IEEE Trans Med Imaging. 2009 May;28(5):775-85. doi: 10.1109/TMI.2008.2012029. Epub 2009 Jan 13. IEEE Trans Med Imaging. 2009. PMID: 19150786
-
Accuracy of single-field nonmydriatic digital fundus image in screening for diabetic retinopathy.J Med Assoc Thai. 2008 Sep;91(9):1397-403. J Med Assoc Thai. 2008. PMID: 18843870
-
Diabetic retinopathy: the unmet needs for screening and a review of potential solutions.Expert Rev Med Devices. 2006 May;3(3):301-13. doi: 10.1586/17434440.3.3.301. Expert Rev Med Devices. 2006. PMID: 16681452 Review.
Cited by
-
A User-friendly Approach for the Diagnosis of Diabetic Retinopathy Using ChatGPT and Automated Machine Learning.Ophthalmol Sci. 2024 Feb 21;4(4):100495. doi: 10.1016/j.xops.2024.100495. eCollection 2024 Jul-Aug. Ophthalmol Sci. 2024. PMID: 38690313 Free PMC article.
-
Vision transformers: The next frontier for deep learning-based ophthalmic image analysis.Saudi J Ophthalmol. 2023 Jul 14;37(3):173-178. doi: 10.4103/sjopt.sjopt_91_23. eCollection 2023 Jul-Sep. Saudi J Ophthalmol. 2023. PMID: 38074310 Free PMC article.
-
Optic disc detection based on fully convolutional network and weighted matrix recovery model.Med Biol Eng Comput. 2023 Dec;61(12):3319-3333. doi: 10.1007/s11517-023-02891-2. Epub 2023 Sep 5. Med Biol Eng Comput. 2023. PMID: 37668892
-
Fine-grained attention & knowledge-based collaborative network for diabetic retinopathy grading.Heliyon. 2023 Jun 16;9(7):e17217. doi: 10.1016/j.heliyon.2023.e17217. eCollection 2023 Jul. Heliyon. 2023. PMID: 37449186 Free PMC article.
-
Retinal Disease Detection Using Deep Learning Techniques: A Comprehensive Review.J Imaging. 2023 Apr 18;9(4):84. doi: 10.3390/jimaging9040084. J Imaging. 2023. PMID: 37103235 Free PMC article. Review.
Publication types
MeSH terms
LinkOut - more resources
Full Text Sources
Other Literature Sources
Medical
Miscellaneous