Fundus images analysis using deep features for detection of exudates, hemorrhages and microaneurysms
Abstract Background Convolution neural networks have been considered for automatic analysis of fundus images to detect signs of diabetic retinopathy but suffer from low sensitivity. Methods This study has proposed an alternate method using probabilistic output from Convolution neural network to auto...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2018-11-01
|
| Series: | BMC Ophthalmology |
| Subjects: | |
| Online Access: | http://link.springer.com/article/10.1186/s12886-018-0954-4 |
| _version_ | 1828377558907879424 |
|---|---|
| author | Parham Khojasteh Behzad Aliahmad Dinesh K. Kumar |
| author_facet | Parham Khojasteh Behzad Aliahmad Dinesh K. Kumar |
| author_sort | Parham Khojasteh |
| collection | DOAJ |
| description | Abstract Background Convolution neural networks have been considered for automatic analysis of fundus images to detect signs of diabetic retinopathy but suffer from low sensitivity. Methods This study has proposed an alternate method using probabilistic output from Convolution neural network to automatically and simultaneously detect exudates, hemorrhages and microaneurysms. The method was evaluated using two approaches: patch and image-based analysis of the fundus images on two public databases: DIARETDB1 and e-Ophtha. The novelty of the proposed method is that the images were analyzed using probability maps generated by score values of the softmax layer instead of the use of the binary output. Results The sensitivity of the proposed approach was 0.96, 0.84 and 0.85 for detection of exudates, hemorrhages and microaneurysms, respectively when considering patch-based analysis. The results show overall accuracy for DIARETDB1 was 97.3% and 86.6% for e-Ophtha. The error rate for image-based analysis was also significantly reduced when compared with other works. Conclusion The proposed method provides the framework for convolution neural network-based analysis of fundus images to identify exudates, hemorrhages, and microaneurysms. It obtained accuracy and sensitivity which were significantly better than the reported studies and makes it suitable for automatic diabetic retinopathy signs detection. |
| first_indexed | 2024-04-14T08:14:03Z |
| format | Article |
| id | doaj.art-cc39dcdc059d48b984b3ab1fe5f64576 |
| institution | Directory Open Access Journal |
| issn | 1471-2415 |
| language | English |
| last_indexed | 2024-04-14T08:14:03Z |
| publishDate | 2018-11-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Ophthalmology |
| spelling | doaj.art-cc39dcdc059d48b984b3ab1fe5f645762022-12-22T02:04:28ZengBMCBMC Ophthalmology1471-24152018-11-0118111310.1186/s12886-018-0954-4Fundus images analysis using deep features for detection of exudates, hemorrhages and microaneurysmsParham Khojasteh0Behzad Aliahmad1Dinesh K. Kumar2Biosignal Lab, School of Engineering, RMIT UniversityBiosignal Lab, School of Engineering, RMIT UniversityBiosignal Lab, School of Engineering, RMIT UniversityAbstract Background Convolution neural networks have been considered for automatic analysis of fundus images to detect signs of diabetic retinopathy but suffer from low sensitivity. Methods This study has proposed an alternate method using probabilistic output from Convolution neural network to automatically and simultaneously detect exudates, hemorrhages and microaneurysms. The method was evaluated using two approaches: patch and image-based analysis of the fundus images on two public databases: DIARETDB1 and e-Ophtha. The novelty of the proposed method is that the images were analyzed using probability maps generated by score values of the softmax layer instead of the use of the binary output. Results The sensitivity of the proposed approach was 0.96, 0.84 and 0.85 for detection of exudates, hemorrhages and microaneurysms, respectively when considering patch-based analysis. The results show overall accuracy for DIARETDB1 was 97.3% and 86.6% for e-Ophtha. The error rate for image-based analysis was also significantly reduced when compared with other works. Conclusion The proposed method provides the framework for convolution neural network-based analysis of fundus images to identify exudates, hemorrhages, and microaneurysms. It obtained accuracy and sensitivity which were significantly better than the reported studies and makes it suitable for automatic diabetic retinopathy signs detection.http://link.springer.com/article/10.1186/s12886-018-0954-4Fundus image analysisDiabetic retinopathyDeep learningConvolutional neural networksImage processing |
| spellingShingle | Parham Khojasteh Behzad Aliahmad Dinesh K. Kumar Fundus images analysis using deep features for detection of exudates, hemorrhages and microaneurysms BMC Ophthalmology Fundus image analysis Diabetic retinopathy Deep learning Convolutional neural networks Image processing |
| title | Fundus images analysis using deep features for detection of exudates, hemorrhages and microaneurysms |
| title_full | Fundus images analysis using deep features for detection of exudates, hemorrhages and microaneurysms |
| title_fullStr | Fundus images analysis using deep features for detection of exudates, hemorrhages and microaneurysms |
| title_full_unstemmed | Fundus images analysis using deep features for detection of exudates, hemorrhages and microaneurysms |
| title_short | Fundus images analysis using deep features for detection of exudates, hemorrhages and microaneurysms |
| title_sort | fundus images analysis using deep features for detection of exudates hemorrhages and microaneurysms |
| topic | Fundus image analysis Diabetic retinopathy Deep learning Convolutional neural networks Image processing |
| url | http://link.springer.com/article/10.1186/s12886-018-0954-4 |
| work_keys_str_mv | AT parhamkhojasteh fundusimagesanalysisusingdeepfeaturesfordetectionofexudateshemorrhagesandmicroaneurysms AT behzadaliahmad fundusimagesanalysisusingdeepfeaturesfordetectionofexudateshemorrhagesandmicroaneurysms AT dineshkkumar fundusimagesanalysisusingdeepfeaturesfordetectionofexudateshemorrhagesandmicroaneurysms |