Mag. Braulio Solano Rojas

Mag. Braulio Solano Rojas

Labor: 
Investigador(a)
Unidad académica: 
Centro de Investigaciones en Tecnologías de la Información y Comunicación (CITIC)
Es estudiante: 
No

Proyectos

Publicaciones

Alzheimer’s Disease Early Detection Using a Low Cost Three-Dimensional Densenet-121 Architecture

Jun
2020

Descripción:

The objective of this work is to detect Alzheimer’s disease using Magnetic Resonance Imaging. For this, we use a three-dimensional densenet-121 architecture. With the use of only freely available tools, we obtain good results: a deep neural network showing metrics of 87% accuracy, 87% sensitivity (micro-average), 88% specificity (micro-average), and 92% AUROC (micro-average) for the task of classifying five different classes (disease stages). The use of tools available for free means that this work can be replicated in developing countries.

Tipo de publicación: Conference Paper

Publicado en: Lecture Notes in Computer Science

A Low-Cost Three-Dimensional DenseNet Neural Network for Alzheimer’s Disease Early Discovery

Feb
2021

Descripción:

Alzheimer’s disease is the most prevalent dementia among the elderly population. Early detection is critical because it can help with future planning for those potentially affected. This paper uses a three-dimensional DenseNet architecture to detect Alzheimer’s disease in magnetic resonance imaging. Our work is restricted to the use of freely available tools. We constructed a deep neural network classifier with metrics of 0.86 mean accuracy, 0.86 mean sensitivity (micro-average), 0.86 mean specificity (micro-average), and 0.91 area under the receiver operating characteristic curve (micro-average) for the task of discriminating between five different disease stages or classes. The use of tools available for free ensures the reproducibility of the study and the applicability of the classification system in developing countries.

Tipo de publicación: Journal Article

Publicado en: Sensors

Micro Evolutionary Particle Swarm Optimization (MEPSO): A new modified metaheuristic

Dic
2023

Descripción:

This work introduces the algorithm Micro Evolutionary Particle Swarm Optimization (MEPSO), which is based on the Particle Swarm Optimization algorithm (PSO). MEPSO’s main idea is to replace PSO’s update of velocity with evolutionary mutations and crossovers that occur probabilistically. MEPSO uses two control hyperparameters of the evolutionary operators, and , to improve convergence ability. These hyperparameters allow the algorithm to adapt to different problems. MEPSO algorithm has a crucial feature: the use of micropopulations. This results in reduced computational complexity, faster convergence rate, prevention of premature convergence, and easy implementation. The proposed approach provides the basis for further study of MEPSO’s suitability for dynamic and multiobjective optimization. We evaluated MEPSO and PSO against 14 chosen benchmark functions. PSO’s results revealed that it had an average success rate of 84% and a median computation time of 49.9 s. MEPSO, on the other hand, had a success rate of 81% and a computation time of 8 s. We found MEPSO to be faster when we performed a right-tail Wilcoxon Signed-Rank statistical test against the Mean Computation Time of 30 runs for each function for each algorithm.

Tipo de publicación: Journal Article

Publicado en: Systems and Soft Computing

Derechos reservados © 2024 Universidad de Costa RIca