eprintid: 17450 rev_number: 9 eprint_status: archive userid: 2 dir: disk0/00/01/74/50 datestamp: 2025-03-31 23:30:12 lastmod: 2025-03-31 23:30:13 status_changed: 2025-03-31 23:30:12 type: article metadata_visibility: show creators_name: Ejaz, Sara creators_name: Zia, Hafiz U creators_name: Majeed, Fiaz creators_name: Shafique, Umair creators_name: Carvajal-Altamiranda, Stefanía creators_name: Lipari, Vivian creators_name: Ashraf, Imran creators_id: creators_id: creators_id: creators_id: creators_id: stefania.carvajal@uneatlantico.es creators_id: vivian.lipari@uneatlantico.es creators_id: title: Fundus image classification using feature concatenation for early diagnosis of retinal disease ispublished: pub subjects: uneat_bm subjects: uneat_eng divisions: uneatlantico_produccion_cientifica divisions: unincol_produccion_cientifica divisions: uninimx_produccion_cientifica divisions: unic_produccion_cientifica divisions: uniromana_produccion_cientifica full_text_status: public keywords: Public health, retinal disease detection, deep learning, feature extraction, convolutional neural networks abstract: Background Deep learning models assist ophthalmologists in early detection of diseases from retinal images and timely treatment. Aim Owing to robust and accurate results from deep learning models, we aim to use convolutional neural network (CNN) to provide a non-invasive method for early detection of eye diseases. Methodology We used a hybridized CNN with deep learning (DL) based on two separate CNN blocks, to identify multiple Optic Disc Cupping, Diabetic Retinopathy, Media Haze, and Healthy images. We used the RFMiD dataset, which contains various categories of fundus images representing different eye diseases. Data augmenting, resizing, coping, and one-hot encoding are used among other preprocessing techniques to improve the performance of the proposed model. Color fundus images have been analyzed by CNNs to extract relevant features. Two CCN models that extract deep features are trained in parallel. To obtain more noticeable features, the gathered features are further fused utilizing the Canonical Correlation Analysis fusion approach. To assess the effectiveness, we employed eight classification algorithms: Gradient boosting, support vector machines, voting ensemble, medium KNN, Naive Bayes, COARSE- KNN, random forest, and fine KNN. Results With the greatest accuracy of 93.39%, the ensemble learning performed better than the other algorithms. Conclusion The accuracy rates suggest that the deep learning model has learned to distinguish between different eye disease categories and healthy images effectively. It contributes to the field of eye disease detection through the analysis of color fundus images by providing a reliable and efficient diagnostic system. date: 2025-03 publication: DIGITAL HEALTH volume: 11 id_number: doi:10.1177/20552076251328120 refereed: TRUE issn: 2055-2076 official_url: http://doi.org/10.1177/20552076251328120 access: open language: en citation: Artículo Materias > Biomedicina <http://repositorio.unini.edu.mx/view/subjects/uneat=5Fbm.html> Materias > Ingeniería <http://repositorio.unini.edu.mx/view/subjects/uneat=5Feng.html> Universidad Europea del Atlántico > Investigación > Producción Científica <http://repositorio.unini.edu.mx/view/divisions/uneatlantico=5Fproduccion=5Fcientifica.html> Fundación Universitaria Internacional de Colombia > Investigación > Producción Científica <http://repositorio.unini.edu.mx/view/divisions/unincol=5Fproduccion=5Fcientifica.html> Universidad Internacional Iberoamericana México > Investigación > Artículos y libros <http://repositorio.unini.edu.mx/view/divisions/uninimx=5Fproduccion=5Fcientifica.html> Universidad Internacional do Cuanza > Investigación > Producción Científica <http://repositorio.unini.edu.mx/view/divisions/unic=5Fproduccion=5Fcientifica.html> Universidad de La Romana > Investigación > Producción Científica <http://repositorio.unini.edu.mx/view/divisions/uniromana=5Fproduccion=5Fcientifica.html> Abierto Inglés Background Deep learning models assist ophthalmologists in early detection of diseases from retinal images and timely treatment. Aim Owing to robust and accurate results from deep learning models, we aim to use convolutional neural network (CNN) to provide a non-invasive method for early detection of eye diseases. Methodology We used a hybridized CNN with deep learning (DL) based on two separate CNN blocks, to identify multiple Optic Disc Cupping, Diabetic Retinopathy, Media Haze, and Healthy images. We used the RFMiD dataset, which contains various categories of fundus images representing different eye diseases. Data augmenting, resizing, coping, and one-hot encoding are used among other preprocessing techniques to improve the performance of the proposed model. Color fundus images have been analyzed by CNNs to extract relevant features. Two CCN models that extract deep features are trained in parallel. To obtain more noticeable features, the gathered features are further fused utilizing the Canonical Correlation Analysis fusion approach. To assess the effectiveness, we employed eight classification algorithms: Gradient boosting, support vector machines, voting ensemble, medium KNN, Naive Bayes, COARSE- KNN, random forest, and fine KNN. Results With the greatest accuracy of 93.39%, the ensemble learning performed better than the other algorithms. Conclusion The accuracy rates suggest that the deep learning model has learned to distinguish between different eye disease categories and healthy images effectively. It contributes to the field of eye disease detection through the analysis of color fundus images by providing a reliable and efficient diagnostic system. metadata Ejaz, Sara; Zia, Hafiz U; Majeed, Fiaz; Shafique, Umair; Carvajal-Altamiranda, Stefanía; Lipari, Vivian y Ashraf, Imran mail SIN ESPECIFICAR, SIN ESPECIFICAR, SIN ESPECIFICAR, SIN ESPECIFICAR, stefania.carvajal@uneatlantico.es, vivian.lipari@uneatlantico.es, SIN ESPECIFICAR <http://repositorio.unini.edu.mx/id/eprint/17450/1/ejaz-et-al-2025-fundus-image-classification-using-feature-concatenation-for-early-diagnosis-of-retinal-disease.pdf> (2025) Fundus image classification using feature concatenation for early diagnosis of retinal disease. DIGITAL HEALTH, 11. ISSN 2055-2076 document_url: http://repositorio.unini.edu.mx/id/eprint/17450/1/ejaz-et-al-2025-fundus-image-classification-using-feature-concatenation-for-early-diagnosis-of-retinal-disease.pdf