An Effective Deep Learning Classification of Diabetes Based Eye Disease Grades: An Retinal Analysis Approach
- Title
- An Effective Deep Learning Classification of Diabetes Based Eye Disease Grades: An Retinal Analysis Approach
- Creator
- Ajesh F.; Jims A.; Alapatt B.P.; Philip F.M.
- Description
- Diabetic Retinopathy (DR) is a common misdiagnosis of diabetes mellitus, which damages the retina and impairs eyesight. It can lead to vision impairment if it is not caught early. Tragically, DR is an unbreakable cycle, and treatment only serves to reinforce the perception. Early detection of DR and effective treatment can significantly lower the risk of visual loss. In comparison to PC-aided conclusion frameworks, the manual analysis process used by ophthalmologists to diagnose DR retina fundus images takes a lot of time, effort, and money and is prone to error. As of late, profound learning has become quite possibly the most well-known procedure that has accomplished better execution in numerous areas, particularly in clinical picture examination and classification. Thereby, this paper brings an effective deep learning-based diabetes-based retinography in which the following are the stages: a) Data collection from MESSIDOR which contains 1200 images classified into 4 levels and graded from 03 followed by b) Preprocessing using grayscale normalized data. Then followed by c) feature extraction using Discrete Wavelet Transform (DWT), d) feature selection using Particle Swarm Optimization (PSO) and finally given for e) classification using Densenet 169. Experimental states that the proposed model outperforms and effectively classified grades compared to other state-of-art models (accuracy:0.95, sensitivity:0.96, specificity;0.97). 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
- Source
- Lecture Notes in Networks and Systems, Vol-649 LNNS, pp. 234-244.
- Date
- 2023-01-01
- Publisher
- Springer Science and Business Media Deutschland GmbH
- Subject
- Classification; Deep learning; Densenet; Diabetes retinopathy; Feature extraction; Feature selection; Particle Swarm Optimization
- Coverage
- Ajesh F., Department of Computer Science and Engineering, Sree Buddha College of Engineering, Kerala, Alappuzha, India; Jims A., JAIN (Deemed-to-be University), Bangalore, India; Alapatt B.P., CHRIST (Deemed to be University), Delhi-NCR, India; Philip F.M., JAIN (Deemed-to-be University), Bangalore, India
- Rights
- Restricted Access
- Relation
- ISSN: 23673370; ISBN: 978-303127498-5
- Format
- Online
- Language
- English
- Type
- Conference paper
Collection
Citation
Ajesh F.; Jims A.; Alapatt B.P.; Philip F.M., “An Effective Deep Learning Classification of Diabetes Based Eye Disease Grades: An Retinal Analysis Approach,” CHRIST (Deemed To Be University) Institutional Repository, accessed February 24, 2025, https://archives.christuniversity.in/items/show/20012.