Deep Learning for Glaucoma Classification and Grading: A Comprehensive Review on Fundus Imaging Approaches Eugenia Arrieta-Rodriguez, Jose Araque-Gallardo, Natalia Peñaloza Barrios, Oscar Luis Teheran Forero, Maria Claudia Bonfante, Emiro De-La-Hoz-Franco, Margarita Gamarra, José Escorcia-Gutierrez IEEE Access, 2025 Glaucoma is one of the leading causes of blindness worldwide and is characterized by progressive visual field loss due to optic nerve damage. Early detection is fundamental, yet it is often hindered by the asymptomatic nature of the disease in its initial stages. In response to this challenge, advanced techniques such as Deep Learning (DL) and computer vision are emerging as potential tools to revolutionize glaucoma diagnosis. This article aims to systematically evaluate the current state of artificial intelligence approaches for fundus image-based glaucoma detection and to identify trends, challenges, and opportunities. Following the PRISMA methodology, we conducted a comprehensive systematic review examining a total of 63 publications available in Scopus, ScienceDirect, IEEE, and Web of Science databases, available in English and published during the study period between 2020 and 2024. The review revealed key techniques in the critical stages of automated glaucoma detection. Convolutional neural networks dominated recent literature, with ResNet architectures achieving optimal performance (accuracy range: 82.37%-98.48%). For localization and segmentation, U-Net variants, attention-guided networks, and advanced ensemble approaches were prominent. In feature extraction, methods exploiting structural and textural metrics, wavelet-based transformations, and attention mechanisms showed considerable potential. Classification tasks benefited from Convolutional Neural Network (CNN) optimizations, attention-based architectures, hybrid models, and transformer frameworks, demonstrating high accuracy for both binary and multiclass glaucoma detection. Deep learning (DL) approaches have demonstrated significant potential for both binary and multiclass glaucoma classification from color fundus images. Key findings from this study include: i) attention mechanisms and transformer architectures show superior performance in capturing subtle disease features, with accuracies exceeding 95%; ii) hybrid approaches combining multiple techniques achieve better generalization across datasets; iii) curriculum learning strategies improve multiclass severity grading accuracy; and iv) challenges persist in standardizing evaluation metrics and managing variations in data quality. Future research should focus on developing more robust architectures that can handle diverse image qualities and incorporate clinical knowledge into the learning process. Additionally, systematic analysis revealed critical implementation barriers: only 30%-40% of studies included external validation, with significant performance degradation on independent datasets; approximately 60%-70% relied on structure-only approaches, excluding essential visual field correlation for clinical decision-making.
Enhanced Dwarf Mongoose optimization algorithm with deep learning-based attack detection for drones Yazan A. Alsariera, Waleed Fayez Awwad, Abeer D. Algarni, Hela Elmannai, Margarita Gamarra, José Escorcia-Gutierrez Alexandria Engineering Journal, 2024 Security in smart cities is a challenging issue in urban environments as they depend upon interconnected technologies and data for effective services. To address security challenges, smart cities implement robust cybersecurity measures, including network monitoring, encryption, and intrusion detection systems. Detecting and mitigating possible security risks in drone network B5G is a crucial aspect of ensuring reliable and safe drone operation. It is necessary to establish sophisticated and robust attack detection techniques to defend against security threats as the use of drones becomes increasingly widespread and their applications diversify. This is due to the lack of privacy and security consideration in the drone’s system, including an inadequate computation capability and unsecured wireless channels to perform advanced cryptographic algorithms. Intrusion detection systems (IDS) and anomaly detection systems can identify suspicious activities and monitor network traffic, such as anomalous communication patterns or unauthorized access attempts. Therefore, the study presents an enhanced dwarf mongoose optimization algorithm with deep learning-based attack detection (EDMOA-DLAD) in Networks B5G for the purpose of Drones technique. The presented EDMOA-DLAD technique aims to recognize the attacks and classifies them on the drone network B5G. Primarily, the EDMOA-DLAD technique designs a feature selection (FS) approach using EDMOA. To detect attacks, the EDMOA-DLAD technique uses a deep variational autoencoder (DVAE) classifier. Finally, the EDMOA-DLAD technique applies the beetle antenna search (BAS) technique for the optimum hyperparameter part of DVAE model. The outcome of EDMOA-DLAD approach can be verified on benchmark datasets. A wide range of simulations inferred that the EDMOA-DLAD method obtains enhanced performance of 99.79% over other classification techniques.
A distance-based kernel for classification via Support Vector Machines Nazhir Amaya-Tejera, Margarita Gamarra, Jorge I. Vélez, Eduardo Zurek Frontiers in Artificial Intelligence, 2024 Support Vector Machines (SVMs) are a type of supervised machine learning algorithm widely used for classification tasks. In contrast to traditional methods that split the data into separate training and testing sets, here we propose an innovative approach where subsets of the original data are randomly selected to train the model multiple times. This iterative training process aims to identify a representative data subset, leading to improved inferences about the population. Additionally, we introduce a novel distance-based kernel specifically designed for binary-type features based on a similarity matrix that efficiently handles both binary and multi-class classification problems. Computational experiments on publicly available datasets of varying sizes demonstrate that our proposed method significantly outperforms existing approaches in terms of classification accuracy. Furthermore, the distance-based kernel achieves superior performance compared to other well-known kernels from the literature and those used in previous studies on the same datasets. These findings validate the effectiveness of our proposed classification method and distance-based kernel for SVMs. By leveraging random subset selection and a unique kernel design, we achieve notable improvements in classification accuracy. These results have significant implications for diverse classification problems in Machine Learning and data analysis.
Sea turtle foraging algorithm with hybrid deep learning-based intrusion detection for the internet of drones environment José Escorcia-Gutierrez, Margarita Gamarra, Esmeide Leal, Natasha Madera, Carlos Soto, Romany F. Mansour, Meshal Alharbi, Ahmed Alkhayyat, Deepak Gupta Computers and Electrical Engineering, 2023 The Internet of Drones (IoD) allows for coordinated control of airspace for Unmanned Aerial Vehicles (UAVs), also known as drones. The decreasing costs of processors, sensors, and wireless connectivity have made it possible to use UAVs in many variety of military to civilian applications. While most applications utilizing the drones in the IoD have been real-time related, users are now interested in obtaining real-time services from drones that are tailored to a specific fly zone. This study develops a Sea Turtle Foraging Algorithm with Hybrid Deep Learning-based Intrusion Detection (STFA-HDLID) as a algorithm that recognizes and categorizes intrusions in the IoD environment. For this purpose, it is necessary to implement data pre-processing to standardize the input data via min-max normalization. Additionally, the feature selection process is also based on the STFA. Finally, a Deep Belief Network (DBN) with a Sparrow Search Optimization (SSO) algorithm is used for classification. A comprehensive experimental analysis is performed on a benchmark dataset to demonstrate the performance of the STFA-HDLID, which achieves maximum accuracy of 99.51% and 98.85% on the TON_IoT and UNSW-NB15 datasets, respectively, outperforming other techniques.
SE-Coins System: Software for Supporting Gamification-Based Educational Processes Margarita Gamarra, Jan Charris, Asly Cantillo, Jaime Daza, Brandon Llamas, Mauricio Vásquez-Carbonell, Anderson Dominguez, José Escorcia-Gutierrez Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics, 2023
Hybrid Model of Tourism Recommendation Software Development Isabel Arregocés, Jaime Daza, Jan Charris, Asly Cantillo, Juan Amaya, Margarita Gamarra Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics, 2022
A robust application in vessel recognition based on neural classification of acoustic fingerprint International Journal of Artificial Intelligence, 2018
Spectral analysis techniques for acoustic fingerprints recognition Eduardo E. Zurek, A. Margarita R. Gamarra, G. Jose R. Escorcia, Carlos Gutierrez, Henry Bayona, Roxana Perez, Xavier Garcia 2014 19th Symposium on Image Signal Processing and Artificial Vision Stsiva 2014, 2015
Deep Learning for Glaucoma Classification and Grading: A Comprehensive Review on Fundus Imaging Approaches E Arrieta-Rodriguez, J Araque-Gallardo, NP Barrios, OLT Forero, ... IEEE Access , 2025 2025 Citations: 11
Performance Evaluation of Pretrained Convolutional Neural Networks for Diabetic Macular Edema Diagnosis in Retinal Fundus Imaging J Araque-Gallardo, EA Rodríguez, M Gamarra, J Sierra-Carrillo, ... International Conference on Swarm Intelligence, 217-229 , 2025 2025
Early Detection of Acute Myocardial Infarction (AMI) Risk Using Optimized Machine Learning Models V Fontalvo Reniz, CA Leones Rivera, YY Vecino Yepes, ... Mexican Conference on Pattern Recognition, 259-268 , 2025 2025
Analysis of Pre-trained Convolutional Neural Network Models in Diabetic Macular Edema Detection Through Retinal Fundus Images J Araque-Gallardo, EA Rodríguez, M Gamarra, J Sierra-Carrillo, ... Colombian Conference on Computing, 117-131 , 2024 2024 Citations: 2
Augmented Reality for the Preservation, Dissemination, and Promotion of Cultural Heritage: A Systematic Literature Review J Escorcia-Gutierrez, IA Julio, AS Barliza, MV Carbonell, M Gamarra, ... OSF , 2024 2024
Enhanced Dwarf Mongoose optimization algorithm with deep learning-based attack detection for drones YA Alsariera, WF Awwad, AD Algarni, H Elmannai, M Gamarra, ... Alexandria Engineering Journal 93, 59-66 , 2024 2024 Citations: 17
A Deep Learning Approach to Classification Pneumonia in Thorax Images EA Rodríguez, A Naar, M Gamarra IOP Conference Series: Materials Science and Engineering 1299 (1), 012002 , 2024 2024
A distance-based kernel for classification via Support Vector Machines N Amaya-Tejera, M Gamarra, JI Vélez, E Zurek Frontiers in Artificial Intelligence 7, 1287875 , 2024 2024 Citations: 57
LinkU: An Academic and Social Schedule Management Software A Valencia Rua, CE López Gallardo, DE Martínez Medina, ... New Perspectives in Software Engineering, 233-246 , 2024 2024 Citations: 1
Gamification Software to Support the Learning Process of Children with Emphasis on Psychomotor, Psychoanalytic and Attention Deficit Disabilities J Morales, A Fontalvo, S Rodriguez, M Gamarra New Perspectives in Software Engineering, 247-260 , 2024 2024 Citations: 1
Efficient leukocytes detection and classification in microscopic blood images using convolutional neural network coupled with a dual attention network S Khan, M Sajjad, N Abbas, J Escorcia-Gutierrez, M Gamarra, ... Computers in Biology and Medicine, 108146 , 2024 2024 Citations: 65
SE-Coins System: Software for Supporting Gamification-Based Educational Processes M Gamarra, J Charris, A Cantillo, J Daza, B Llamas, ... International Conference on Computer Information Systems and Industrial … , 2023 2023
Grading Diabetic Retinopathy Using Transfer Learning-Based Convolutional Neural Networks J Escorcia-Gutierrez, J Cuello, M Gamarra, P Romero-Aroca, E Caicedo, ... International Conference on Computer Information Systems and Industrial … , 2023 2023 Citations: 6
Artificial intelligence with big data analytics-based brain intracranial hemorrhage e-diagnosis using CT images RF Mansour, J Escorcia-Gutierrez, M Gamarra, VG Díaz, D Gupta, ... Neural Computing and Applications 35 (22), 16037-16049 , 2023 2023 Citations: 96
Sea turtle foraging algorithm with hybrid deep learning-based intrusion detection for the internet of drones environment J Escorcia-Gutierrez, M Gamarra, E Leal, N Madera, C Soto, RF Mansour, ... Computers and Electrical Engineering 108, 108704 , 2023 2023 Citations: 49
Optimal synergic deep learning for COVID-19 classification using chest x-ray images J Escorcia-Gutierrez, M Gamarra, R Soto-Diaz, A Yafoz, RF Mansour Computers, Materials and Continua 75 (3), 5255-5270 , 2023 2023 Citations: 4
Galactic swarm optimization with deep transfer learning driven colorectal cancer classification for image guided intervention J Escorcia-Gutierrez, M Gamarra, PP Ariza-Colpas, GB Roncallo, N Leal, ... Computers and Electrical Engineering 104, 108462 , 2022 2022 Citations: 20
System for data seclusion in image and audio files RF Mansour, J Escorcia-Gutierrez, M Gamarra, A El Amraoui, F Alenezi, ... 2022
SYSTEM FOR DATA SECLUSION IN IMAGE AND AUDIO FILES RF Mansour, J Escorcia-gutierrez, M Gamarra, A El Amraoui, F Alenezi, ... US Patent App. 17/806,563 , 2022 2022
Intelligent sine cosine optimization with deep transfer learning based crops type classification using hyperspectral images J Escorcia-Gutierrez, M Gamarra, M Torres-Torres, N Madera, ... Canadian Journal of Remote Sensing 48 (5), 621-632 , 2022 2022 Citations: 8
MOST CITED SCHOLAR PUBLICATIONS
Intelligent video anomaly detection and classification using faster RCNN with deep reinforcement learning model RF Mansour, J Escorcia-Gutierrez, M Gamarra, JA Villanueva, N Leal Image and Vision Computing 112, 104229 , 2021 2021 Citations: 170
Unsupervised Deep Learning based Variational Autoencoder Model for COVID-19 Diagnosis and Classification RF Mansour, J Escorcia-Gutierrez, M Gamarra, D Gupta, O Castillo, ... Pattern Recognition Letters 151, 267-274 , 2021 2021 Citations: 114
Split and merge watershed: A two-step method for cell segmentation in fluorescence microscopy images M Gamarra, E Zurek, HJ Escalante, L Hurtado, H San-Juan-Vergara Biomedical Signal Processing and Control 53, 101575 , 2019 2019 Citations: 103
Artificial intelligence with big data analytics-based brain intracranial hemorrhage e-diagnosis using CT images RF Mansour, J Escorcia-Gutierrez, M Gamarra, VG Díaz, D Gupta, ... Neural Computing and Applications 35 (22), 16037-16049 , 2023 2023 Citations: 96
A gamification strategy in engineering education—A case study on motivation and engagement M Gamarra, A Dominguez, J Velazquez, H Páez Computer Applications in Engineering Education , 2022 2022 Citations: 84
Intelligent Agricultural Modelling of Soil Nutrients and pH Classification Using Ensemble Deep Learning Techniques J Escorcia-Gutierrez, M Gamarra, R Soto-Diaz, M Pérez, N Madera, ... Agriculture 12 (7), 977 , 2022 2022 Citations: 77
Efficient leukocytes detection and classification in microscopic blood images using convolutional neural network coupled with a dual attention network S Khan, M Sajjad, N Abbas, J Escorcia-Gutierrez, M Gamarra, ... Computers in Biology and Medicine, 108146 , 2024 2024 Citations: 65
Intelligent deep learning-enabled autonomous small ship detection and classification model J Escorcia-Gutierrez, M Gamarra, K Beleño, C Soto, RF Mansour Computers and Electrical Engineering 100, 107871 , 2022 2022 Citations: 62
A distance-based kernel for classification via Support Vector Machines N Amaya-Tejera, M Gamarra, JI Vélez, E Zurek Frontiers in Artificial Intelligence 7, 1287875 , 2024 2024 Citations: 57
An innovative image-processing model for rust detection using Perlin Noise to simulate oxide textures MRG Acosta, JCV Díaz, NS Castro Corrosion Science 88, 141-151 , 2014 2014 Citations: 57
Deep learning with backtracking search optimization based skin lesion diagnosis model CSS Anupama, L Natrayan, E Laxmi Lydia, AR Wahab Sait, ... Computers, Materials and Continua 70 (1), 1297-1313 , 2022 2022 Citations: 54
Sea turtle foraging algorithm with hybrid deep learning-based intrusion detection for the internet of drones environment J Escorcia-Gutierrez, M Gamarra, E Leal, N Madera, C Soto, RF Mansour, ... Computers and Electrical Engineering 108, 108704 , 2023 2023 Citations: 49
Analysis of pre-trained convolutional neural network models in diabetic retinopathy detection through retinal fundus images J Escorcia-Gutierrez, J Cuello, C Barraza, M Gamarra, P Romero-Aroca, ... International Conference on Computer Information Systems and Industrial … , 2022 2022 Citations: 22
Addendum for: A Study of Image Analysis Algorithms for Segmentation, Feature Extraction and Classification of Cells M Gamarra, E Zurek, H San-Juan Journal of Information Systems Engineering & Management 3 (1), 05 , 2018 2018 Citations: 22
Using genetic algorithm feature selection in neural classification systems for image pattern recognition A Gamarra, M Quintero Ingeniería e Investigación 33 (1), 52-58 , 2013 2013 Citations: 21
Galactic swarm optimization with deep transfer learning driven colorectal cancer classification for image guided intervention J Escorcia-Gutierrez, M Gamarra, PP Ariza-Colpas, GB Roncallo, N Leal, ... Computers and Electrical Engineering 104, 108462 , 2022 2022 Citations: 20
A color fusion model based on Markowitz portfolio optimization for optic disc segmentation in retinal images J Escorcia-Gutierrez, J Torrents-Barrena, M Gamarra, P Romero-Aroca, ... Expert Systems with Applications 174, 114697 , 2021 2021 Citations: 19
Diagnosis of leukemia disease based on enhanced virtual neural network K Muthumayil, S Manikandan, S Srinivasan, J Escorcia-Gutierrez, ... Corporación Universidad de la Costa , 2021 2021 Citations: 19
Identification of microbiota biomarkers with orthologous gene annotation for type 2 diabetes YH Zhang, W Guo, T Zeng, SQ Zhang, L Chen, M Gamarra, RF Mansour, ... Frontiers in Microbiology 12, 711244 , 2021 2021 Citations: 18
Enhanced Dwarf Mongoose optimization algorithm with deep learning-based attack detection for drones YA Alsariera, WF Awwad, AD Algarni, H Elmannai, M Gamarra, ... Alexandria Engineering Journal 93, 59-66 , 2024 2024 Citations: 17
RESEARCH OUTPUTS (PATENTS, SOFTWARE, PUBLICATIONS, PRODUCTS)
Software development: FaceFinder, IQControl, INTELLIGENT RETINAL IMAGE SEGMENTATION- IRIS, ASSISTANT FOR LABORATORY INVESTIGATIONS IN CELL IMAGE ANALYSIS- ALICIA.
