My research spans statistical modeling, machine learning, and Bayesian methods applied to diverse domains, with a strong current focus on Precision Medicine and Personalized Treatment Effect Prediction. Below are the main fields that structure my work:


Machine Learning and Bayesian Methods for Precision Medicine

I develop advanced statistical and machine learning methods to support personalized healthcare decisions. My work includes Bayesian Additive Regression Trees, meta-heuristic feature selection, and hybrid AI approaches for predicting individual treatment effects (PITE), particularly in clinical trials for alcohol use disorders.


Statistical Learning and Predictive Modeling

I explore regression and tree-based models for complex outcome structures and constrained domains, aiming to improve predictive accuracy and interpretability.


Machine Learning in Remote Sensing and Environmental Monitoring

I apply probabilistic models and machine learning algorithms to predict vegetation and biomass using multispectral and LiDAR UAV data.


Bayesian and Hierarchical Models in Ecology and Climate Research

I develop statistical models for ecological and climate data, including compositional regression and hierarchical approaches to understand species distribution and environmental interactions.


Applied Statistical Modeling in Public Health and Epidemiology

I have worked on Bayesian hierarchical models for disease surveillance and epidemic dynamics, as well as modeling COVID-19 vulnerability and protection indices.


Statistical Methods for Risk and Insurance Models

Earlier research focused on actuarial science, including collective risk models, heavy-tailed distributions, and Bayesian approaches for pricing and risk measures.


Multivariate and Compositional Data Analysis

I have proposed methods for clustering and regression with compositional data, applied to wood properties, marine ecology, and other complex systems.


Bibliometrics and Scientific Visibility

Earlier work includes analyzing scientific impact and visibility through bibliometric methods.