Conference Agenda

This work experimentally evaluates the dynamic behavior of digital and analog Geiger-Müller (GM counters under operational conditions in a facility housing ionizing radiation sources. Although these instruments are widely used in radiation protection, their temporal response to abrupt changes in dose is not always considered in practice. For this study, quantitative criteria were defined to determine the response time (exceeding 90 % of the maximum recorded dose rate) and the recovery time (returning withing 10 % of the reference value). Measurements were obtained through video analysis with a temporal resolution of 1/30 s, and five repetitions were performed for each condition. Time uncertainty was estimated as Type A in accordance with the GUM, and ANOVA and Tukey tests were applied to evaluate differences among indication channels (numerical, audible, and visual). The results show that the characteristic times, on the order of seconds, depend significantly on the instrument type and output channel, being greater for numerical indicators than for alarms. The observed behavior is dominated by signal processing electronics rather than detector dead time, indicating that quantitative reading is not strictly instantaneous in scenarios involving rapid dose rate variations.

Automated medical diagnosis requires predictive models that not only achieve high accuracy but also provide principled and interpretable uncertainty quantification aligned with clinical decision-making. In this work, we develop a comprehensive mathematical framework for binary medical classification grounded in Bayesian inference, integrating clinical decision theory under asymmetric loss, strong asymptotic guarantees, and computational analysis.

We show that the Bayesian predictor minimizes expected clinical risk under realistic loss functions, is asymptotically calibrated, and satisfies the Bernstein–von Mises theorem under regularity conditions. We introduce the Clinical Uncertainty Index (CI), a theoretically grounded metric that decomposes predictive uncertainty into epistemic and aleatory components. The CI is proven to converge to zero with increasing sample size under correct model specification and to act as a diagnostic indicator of model misspecification.

Extending the analysis to approximate inference, we formally demonstrate that mean-field variational approximations systematically underestimate posterior uncertainty, potentially inducing diagnostic overconfidence. These findings highlight the critical role of faithful uncertainty estimation in high-stakes clinical applications.

Overall, this study provides rigorous mathematical foundations for the development of explainable and clinically reliable medical AI systems with explicit probabilistic guarantees.

Mortality prediction in intensive care units represents a fundamental challenge for clinical decision-making. However, developing robust predictive models requires large volumes of data that are typically fragmented across multiple healthcare institutions, each with strict privacy policies that prevent the centralization of sensitive information. This study implements and evaluates a federated learning system based on the Federated Averaging (FedAvg) algorithm to train mortality prediction models without the need to share clinical data among participating institutions. Thru computational simulation, a multi-institutional scenario was reproduced with five virtual hospitals, each with heterogeneous demographic characteristics and data distributions. The results demonstrate that the federated approach achieves an area under the ROC curve (AUC-ROC) of 0.892 ± 0.008, representing only a 3.6% difference compared to the centralized reference model (AUC-ROC = 0.925 ± 0.005), while reducing the volume of transferred data by 98% and fully preserving institutional privacy. Statistical analysis using a paired Student’s t-test confirms that this difference, although statistically significant (p < 0.001), is clinically acceptable. It is concluded that federated learning constitutes a viable alternative for inter-institutional collaboration in clinical research, enabling the development of high-performance predictive models without compromising patient confidentiality.

This systematic literature review (SLR) evaluated the design approaches, technical effectiveness, and existing gaps in the implementation of glove-based sign translator systems that employ Convolutional Neural Networks (CNNs). The PRISMA methodology was used to organize and document the study selection and evaluation process. The initial search of the Scopus and Web of Science databases identified 70 publications; after removing duplicates and applying inclusion and exclusion criteria, the final sample compromised 11 studies with thematic relevance and adequate methodological quality. Based on this selection, the most commonly used CNN architectures were examined and their performance was compared, mainly in terms of accuracy. The findings indicate the integration of these architectures with proper development. Furthermore, signal preprocessing, particularly noise filtering and normalization, improves the stability and consistency of the collected data. Despite these advances, relevant limitations persist, including the lack of hardware and evaluation metric standardization, the limited availability of public datasets for specific sign languages, and the high implementation costs. These factors reduce reproducibility and hinder application in real-world contexts. Consequently, the need for standardized datasets, common evaluation protocols, and low-cost, high-efficiency solutions is emphasized to promote practical deployment and communicative inclusion.

The present study aimed to evaluate the health impact associated with occupational exposure to benzene in fuel dispensers at service stations in the city of Guayaquil. This was a cross-sectional study with a mixed, descriptive-correlational approach, based on the application of internationally validated questionnaires and semi-structured interviews to 103 workers. The results revealed that 53.4% of workers reported at least one respiratory symptom, with the most prevalent being those with a work-related temporal relationship: improvement during weekends (17.5%), nocturnal cough (13.6%), and additional symptoms such as headache and dizziness. 83.5% of workers were exposed to fuel vapors for 6 or more hours daily. Statistical analysis showed no significant associations between individual risk factors (exposure time, PPE use, hygiene practices) and symptoms, suggesting that baseline exposure is elevated for the entire working population regardless of the protective measures implemented. These findings contribute to the generation of local empirical evidence to support the implementation of specific medical surveillance programs, periodic environmental measurements, and more robust engineering control strategies in the Ecuadorian service station sector.

Data quality is a problem in healthcare information systems, especially when the capture of diagnostic information is not standardized. In this article, a hybrid computational framework for clinical data normalization is designed and evaluated, combining deterministic text processing methods with fuzzy similarity algorithms and a human-in-the-loop validation mechanism. The proposed system was implemented and tested on 2,777 outpatient care records from the Villa Nueva Health Center in Honduras between January and December 2024. The proposed multi-level normalization pipeline reduced the cardinality of unique diagnoses by 55.56% (from 90 to 40 categories) and geographic locations by 30.07% (from 153 to 107 categories), with 99.46% of mappings achieved by exact match (Level 1). Diagnostic entropy decreased from 3.06 to 2.55 bits (16.65%), while the Herfindahl–Hirschman index increased by 20.91%, indicating greater uniformity in the frequency distribution. The ranking stability analysis yielded Jaccard indices of 0.667 and Kendall’s τ coefficients of 0.929 for the top 10 diagnoses. Thirteen ambiguous cases (0.47%) were identified that required manual review, demonstrating the feasibility of the semi-automated method. The temporal drift analysis showed an average vocabulary stability of 43% (Jaccard) between successive months. The proposed architecture is replicable, scalable, and exportable as a programd pipeline, providing a practical solution for data governance in epidemiological surveillance systems.