Cotton-wool spots, red-lesions and hard-exudates distinction using CNN enhancement and transfer learning
The automatic retinal disease diagnosis by artificial intelligent is an interesting and challenging topic in the medical field. It requires an appropriate image enhancement technique and a sufficient training dataset for the specific retina conditions. The aim of this study was to design an automati...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Institute of Advanced Engineering and Science
2021
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/96531/1/TanTianSwee2021_CottonWoolSpotsRedLesionsAndHardExudates.pdf |
| _version_ | 1796866242679668736 |
|---|---|
| author | Mohammed, Sinan S. Tan, Tian Swee As’ari, M. A. Wan Hitam, Wan Hazabbah Ngoo, Qi Zhe Foh thye, Matthias Tiong Chia hiik, Kelvin Ling |
| author_facet | Mohammed, Sinan S. Tan, Tian Swee As’ari, M. A. Wan Hitam, Wan Hazabbah Ngoo, Qi Zhe Foh thye, Matthias Tiong Chia hiik, Kelvin Ling |
| author_sort | Mohammed, Sinan S. |
| collection | ePrints |
| description | The automatic retinal disease diagnosis by artificial intelligent is an interesting and challenging topic in the medical field. It requires an appropriate image enhancement technique and a sufficient training dataset for the specific retina conditions. The aim of this study was to design an automatic diagnosis convolutional neural network (CNN) model which does not require a large training dataset to specifically identify diabetic retinopathy symptoms, which are cotton wool, exudates spots, and red lesion in colour fundus pictures. A novel framework comprised image enhancement method by using upgraded contrast limited adaptive histogram equalization (UCLAHE) filter and transferred pre-trained networks was developed to classify the retinal diseases regarding to the symptoms. The performance of the proposed framework was evaluated based on accuracy, sensitivity, and specificity metrics. The collected results have proven the robustness of the proposed framework in offering good accuracy in retina diseases diagnosis. |
| first_indexed | 2024-03-05T21:09:13Z |
| format | Article |
| id | utm.eprints-96531 |
| institution | Universiti Teknologi Malaysia - ePrints |
| language | English |
| last_indexed | 2024-03-05T21:09:13Z |
| publishDate | 2021 |
| publisher | Institute of Advanced Engineering and Science |
| record_format | dspace |
| spelling | utm.eprints-965312022-07-26T08:46:36Z http://eprints.utm.my/96531/ Cotton-wool spots, red-lesions and hard-exudates distinction using CNN enhancement and transfer learning Mohammed, Sinan S. Tan, Tian Swee As’ari, M. A. Wan Hitam, Wan Hazabbah Ngoo, Qi Zhe Foh thye, Matthias Tiong Chia hiik, Kelvin Ling RC Internal medicine The automatic retinal disease diagnosis by artificial intelligent is an interesting and challenging topic in the medical field. It requires an appropriate image enhancement technique and a sufficient training dataset for the specific retina conditions. The aim of this study was to design an automatic diagnosis convolutional neural network (CNN) model which does not require a large training dataset to specifically identify diabetic retinopathy symptoms, which are cotton wool, exudates spots, and red lesion in colour fundus pictures. A novel framework comprised image enhancement method by using upgraded contrast limited adaptive histogram equalization (UCLAHE) filter and transferred pre-trained networks was developed to classify the retinal diseases regarding to the symptoms. The performance of the proposed framework was evaluated based on accuracy, sensitivity, and specificity metrics. The collected results have proven the robustness of the proposed framework in offering good accuracy in retina diseases diagnosis. Institute of Advanced Engineering and Science 2021 Article PeerReviewed application/pdf en http://eprints.utm.my/96531/1/TanTianSwee2021_CottonWoolSpotsRedLesionsAndHardExudates.pdf Mohammed, Sinan S. and Tan, Tian Swee and As’ari, M. A. and Wan Hitam, Wan Hazabbah and Ngoo, Qi Zhe and Foh thye, Matthias Tiong and Chia hiik, Kelvin Ling (2021) Cotton-wool spots, red-lesions and hard-exudates distinction using CNN enhancement and transfer learning. Indonesian Journal of Electrical Engineering and Computer Science, 23 (2). pp. 1170-1179. ISSN 2502-4752 http://dx.doi.org/10.11591/ijeecs.v23.i2.pp1170-1179 DOI:10.11591/ijeecs.v23.i2.pp1170-1179 |
| spellingShingle | RC Internal medicine Mohammed, Sinan S. Tan, Tian Swee As’ari, M. A. Wan Hitam, Wan Hazabbah Ngoo, Qi Zhe Foh thye, Matthias Tiong Chia hiik, Kelvin Ling Cotton-wool spots, red-lesions and hard-exudates distinction using CNN enhancement and transfer learning |
| title | Cotton-wool spots, red-lesions and hard-exudates distinction using CNN enhancement and transfer learning |
| title_full | Cotton-wool spots, red-lesions and hard-exudates distinction using CNN enhancement and transfer learning |
| title_fullStr | Cotton-wool spots, red-lesions and hard-exudates distinction using CNN enhancement and transfer learning |
| title_full_unstemmed | Cotton-wool spots, red-lesions and hard-exudates distinction using CNN enhancement and transfer learning |
| title_short | Cotton-wool spots, red-lesions and hard-exudates distinction using CNN enhancement and transfer learning |
| title_sort | cotton wool spots red lesions and hard exudates distinction using cnn enhancement and transfer learning |
| topic | RC Internal medicine |
| url | http://eprints.utm.my/96531/1/TanTianSwee2021_CottonWoolSpotsRedLesionsAndHardExudates.pdf |
| work_keys_str_mv | AT mohammedsinans cottonwoolspotsredlesionsandhardexudatesdistinctionusingcnnenhancementandtransferlearning AT tantianswee cottonwoolspotsredlesionsandhardexudatesdistinctionusingcnnenhancementandtransferlearning AT asarima cottonwoolspotsredlesionsandhardexudatesdistinctionusingcnnenhancementandtransferlearning AT wanhitamwanhazabbah cottonwoolspotsredlesionsandhardexudatesdistinctionusingcnnenhancementandtransferlearning AT ngooqizhe cottonwoolspotsredlesionsandhardexudatesdistinctionusingcnnenhancementandtransferlearning AT fohthyematthiastiong cottonwoolspotsredlesionsandhardexudatesdistinctionusingcnnenhancementandtransferlearning AT chiahiikkelvinling cottonwoolspotsredlesionsandhardexudatesdistinctionusingcnnenhancementandtransferlearning |