POPablo Alberto Osorio‐Marulanda

Pablo Alberto Osorio‐Marulanda

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Pablo Alberto Osorio‐Marulanda has a total of 16 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Functional connectivity methods for detecting seizure onset zone in focal epilepsy: an approach through temporal graphs and community detection

    by Pablo A. Osorio Marulanda, Juan D. Martínez-Vargas, Lina M. Sepúlveda-Cano, Alejandro Peña

    Focal epilepsy, a prevalent neurological disease with seizures from localized brain regions, poses significant diagnostic and treatment challenges. Precise seizure onset zone (SOZ) localization is crucial for surgery in drug-refractory cases. However, traditional static or time-averaged functional connectivity (FC) models often fail to capture the dynamic brain processes inherent in epileptic activity. This study introduces a novel methodology for SOZ identification using dynamic graph representations of stereo-electroencephalography (SEEG)-derived FC. We constructed frequency-specific (theta, alpha, beta, gamma) temporal networks using Amplitude Envelope Correlation (AEC) and Phase-Locking Value (PLV), incorporating spatial distance normalization. These networks were thresholded, binarized, and analyzed using an ensemble of community detection algorithms. A key innovation is our novel intra-community density score with size regularization (IDS), designed to prioritize communities indicative of epileptogenic activity. Validation on simulated datasets (Epileptor model) yielded an area under the curve (AUC) of 0.74. On clinical SEEG from 19 focal epilepsy patients, our framework demonstrated significant utility, particularly when analyzing the low gamma band with 6-s epochs using the AEC method. Stratified analyses revealed strong performance in specific subgroups, achieving an AUC of 0.76 in patients with mesial temporal lobe lesions, highlighting the method’s potential for targeted clinical applications. While the average AUC across the entire heterogeneous patient cohort was 0.50, these specific findings underscore the importance of patient-specific factors and tailored analytical approaches. Our results underscore the importance of temporal network analysis in capturing ictal dynamics. This study demonstrates that dynamic, graph-based FC analysis, incorporating advanced community detection and our novel IDS metric, offers a promising strategy to enhance presurgical SOZ evaluation in focal epilepsy.

  • Open Access

    Modeling Patterns of Population Exposure With a Multiscale Framework Utilizing LOTOS‐EUROS Chemical Transport Model, Mobility Surveys and Particulate Toxicity

    by Valeria Solórzano‐Araque, Andrés Yarce Botero, Jhon Edinson Hinestroza‐Ramirez, Fernando Fernandez‐Restrepo, Sebastian Carmona Estrada, Pablo Alberto Osorio‐Marulanda, Lisseth Cruz‐Ruiz, Santiago Isaza‐Cadavid, O. L. Quintero

    The present study proposes a high‐resolution, multi‐component framework to estimate human exposure to air pollution in tropical urban environments, addressing complex topography and limited historical mobility and exposure data. This framework is applied to the city of Medellín in the Aburrá Valley in Colombia. To assess population exposure to particulate matter, numerical simulations, mobility and toxicity data were integrated. The LOTOS‐EUROS Chemical Transport Model, driven by the Weather Research and Forecasting (WRF) model, was used to simulate air quality at a spatial resolution of 1 km 1 km for the year 2019. Mobility dynamics were derived from the Origin‐Destination Survey (ODS), comprising over 180,000 trips across 240 traffic analysis zones. The utilisation of this data set facilitated the estimation of time‐weighted exposure levels for diverse demographic groups during weekdays. The morpho‐chemical properties of the particulate matter samples were analyzed to determine the presence of metals and carbonaceous fractions due to significant health implications. The exposure model is a weighted model that integrates exposure time, and concentration, and cytogenotoxic indicators. The application of cluster analysis to the available data, resulted in the identification of areas of elevated health risk, indicating that the central and southern zones, approximately 60% of the metropolitan population and main highway corridors, exhibited mean particulate matter exposure levels that exceeded 40 μg/m 3 during peak hours, thus surpassing the WHO air quality guidelines. These zones demonstrated the highest levels of exposure, as indicated by cluster significance levels, suggesting more epidemiological studies and public health interventions.

  • Open Access

    Privacy Mechanisms and Evaluation Metrics for Synthetic Data Generation: A Systematic Review

    by Pablo A. Osorio-Marulanda, Gorka Epelde, Mikel Hernandez, Imanol Isasa, Nicolas Moreno Reyes, Andoni Beristain Iraola

    The growth of data publishing, sharing, and mining mechanisms in various fields of industry and science has led to an increase in the flow of data, making it an important asset that needs to be protected and managed effectively. To this end, different mechanisms have been used across different domains, including Privacy Enhancing Technologies like Synthetic Data Generation, which aim to protect user-sensitive data and prevent misuse among different domains. Then, Synthetic data has been used not only to augment datasets and balance classes but also in applications of data analysis paradigms that aim to provide useful insights in terms of utility while preserving the privacy of sensitive data. Still, there is a gap in the conceptual and state-of-the-art understanding of the level of privacy synthetic data generators can provide and how they affect various industries and fields. This systematic review attempts to address how privacy has been assessed and measured in the framework of synthetic data generation, and getting to know which metrics have been used to evaluate those mechanisms. We provide an overview a total of 105 recent studies in this field after a screening process and identify future open research directions. The main findings include a high prevalence of differential privacy as a privacy-preserving technique and privacy budget cost as a trade-off metric, with a high percentage of GAN-based model implementations, and mainly healthcare applications. Our systematic review covers multiple privacy domains and can be understood as a general framework for privacy measurement applied in Synthetic Data Generation.