Oscar Perdomo

Background Diaphragm pacing offers an alternative to mechanical ventilation for diaphragm paralysis, yet commercial devices use fixed parameters and lack real-time physiological feedback. We developed and bench-validated an embedded diaphragmatic stimulator that classifies respiratory status and trend from real-time measurements of end-tidal CO2 (PetCO2) and oxygen saturation (SpO2). Methods We engineered a fuzzy-logic–based embedded controller to identify patient condition and trend from real-time PetCO2 and SpO2. The bench setup used an ASL5000 breathing simulator and CapnoStream as the PetCO2 reference; SpO2 was varied with a Vpad-A1 simulator. Multiple clinical scenarios and state transitions were reproduced to reflect real-world contexts. We quantified PetCO2 agreement (Bland–Altman, intraclass correlation coefficient [ICC]) and assessed classification agreement for status (normal/abnormal/critical) and trend (improving/stable/deteriorating) using weighted Cohen’s Kappa. Results Across 379 measurements covering standard and hypocapnic patient profiles, PetCO2 exhibited a mean bias of 0.22 mmHg relative to the reference capnometer, with Bland–Altman limits of agreement −1.7 to 1.25 mmHg and ICC = 0.99. Classification of respiratory status (normal/abnormal/critical) showed perfect agreement with the computational reference (weighted Cohen’s κ = 1.0), and trend classification demonstrated excellent agreement (κ > 0.92). Conclusion The embedded electrical stimulator developed successfully integrates PetCO2 and SpO2 monitoring to identify patient status and trends with high agreement, across physiological situations reproduced in a bench model. This represents a key step toward a closed-loop system capable of modulating breathing patterns based on physiological respiratory inputs.

Introduction: Artificial intelligence (AI) is a branch of technology enabling machines to emulate complex human skills; it can also entail problem‐solving using bioinspired methods. It is used for automating systematic literature reviews (SLR), that is, defining a clinical question, locating relevant literature, preliminary screening, study evaluation, data extraction and analysis. Title and abstract screening is one of the most time‐consuming and error‐prone phases involved in developing a systematic review. While AI promises to expedite this process, adopting it faces challenges due to concerns about compatibility and transparency. This review aims to identify current evidence concerning AI use during preliminary SLR reference screening; it describes characteristics such as the different metrics used for reporting performance and how the different algorithms, pipelines, workflows or web applications are validated. AI resource users' reflections regarding SLR screening automation have also been summarized. Methods: A scoping review was conducted following Joanna Briggs Institute's (JBI) methodology. Its objective was to identify existing evidence regarding the use of AI resources for title and abstract screening automation. Searches were limited to articles published between 2019 and 2026. The review included primary studies reporting the development, assessment, validation, or real‐world use of AI resources for screening automation, as well as systematic reviews and articles reporting experiences or recommendations for their use. Two types of data were extracted: (1) from primary studies—characteristics of AI resources and, where applicable, recommendations for their use; (2) from systematic reviews and experience‐based articles, recommendations for the use of AI resources. Results included frequency descriptions, tables, figures, and a decision flowchart reflecting the number of references and articles retrieved, excluded, or included in the final analysis. Results: A total of 174 unique studies published between 2019 and 2026 were included in this scoping review. These were grouped into web applications (43%), model comparisons (32%), generative models (6%), pre‐trained models (3%) or pipelines/workflows (15%) used for title and abstract screening in systematic literature review (SLR). Most studies came from North America. Evaluating these tools often relied on retrospective comparisons with human reviewers' work (63%), sensitivity ( n = 60), and specificity ( n = 62) being the most reported metric for criterion assessment and Work Saved over Sampling ( n = 28) being the most reported metric for assessing their utility. Considerations concerning AI resource use focused on the need for standardized evaluation metrics, stopping criteria, study design and the data sets used, resource characteristics facilitating usability, best practice and future research areas, with the persistence of the human component in the process ( n = 26) being the most pressing recommendation. Conclusion: The findings indicated substantial heterogeneity regarding the types of AI resources used, considerable variation concerning the metrics used for reporting performance, differences in how such metrics are defined and a clear need for standardizing reporting methods, study designs and related procedures. Although AI technologies will continue to evolve, maintaining a clear and consistent framework for interpreting research on AI resources for automating title and abstract screening can support understanding their level of maturity and facilitate informed decision‐making by users.

Purpose: To develop a deep learning (DL) model for the automatic detection of retinal vein occlusion (RVO) biomarkers in optical coherence tomography angiography (OCT‐A) images using retrospective data. Methods: OCT‐A images from 254 patients were used to train and test an artificial intelligence (AI) model. Scans of the superficial, deep, en face, choriocapillaris, and outer retina to choriocapillaris (ORCC) layers were manually annotated with four biomarkers: perifoveal capillary plexus disruption, nonperfusion areas (NPAs), vascular tortuosity, and cystoid spaces. Deep convolutional neural networks (CNNs) were trained to segment and identify these biomarkers. The detection rate and Jaccard index were the primary outcome measures. Results: The DL model achieved detection rates of 93% for perifoveal capillary plexus disruption, 92% for NPAs, 91% for vascular tortuosity, and 84% for cystoid spaces. The corresponding Jaccard index values were 0.85, 0.77, 0.72, and 0.73, respectively. Conclusion: The proposed DL model achieves outstanding performance in identifying key biomarkers of RVO in OCT‐A images, supporting its potential role in automated disease assessment.

Urban noise levels that exceed the World Health Organization (WHO) recommendations have become a growing concern due to their adverse effects on public health. In Bogotá, Colombia, studies by the District Department of Environment (SDA) indicate that 11.8% of the population is exposed to noise levels above the WHO limits. This research aims to identify and categorize environmental noise sources in real time using an embedded intelligent system. A total of 657 labeled audio clips were collected across eight classes and processed using a 60/20/20 train–validation–test split, ensuring that audio segments from the same continuous recording were not mixed across subsets. The system was implemented on a Raspberry Pi 2W equipped with a UMIK-1 microphone and powered by a 90 W solar panel with a 12 V battery, enabling autonomous operation. The TinyML-based model achieved precision and recall values between 0.92 and 1.00, demonstrating high performance under real urban conditions. Heavy vehicles and motorcycles accounted for the largest proportion of classified samples. Although airplane-related events were less frequent, they reached maximum sound levels of up to 88.4 dB(A), exceeding the applicable local limit of 70 dB(A) by approximately 18 dB(A) rather than by percentage. In conclusion, the results demonstrate that on-device TinyML classification is a feasible and effective strategy for urban noise monitoring. Local inference reduces latency, bandwidth usage, and privacy risks by eliminating the need to transmit raw audio to external servers. This approach provides a scalable and sustainable foundation for noise management in smart cities and supports evidence-based public policies aimed at improving urban well-being. This work presents an introductory and exploratory study on the application of TinyML for acoustic environmental monitoring, aiming to evaluate its feasibility and potential for large-scale implementation.

Introducción: El objetivo primario de este estudio es determinar la composición corporal de las mujeres colombianas y su relación con la densidad mamográfica. Métodos: Estudio descriptivo transversal con componente analítico de 670 pacientes mediante muestreo no probabilístico. Se ajustan modelos de regresión lineal simple para determinar la relación entre indicadores de adiposidad con algunas variables sociodemográficas. Posteriormente una regresión lineal simple y múltiple para valorar la densidad mamográfica con variables clínicas y antropométricas. Finalmente, se construyó un árbol de decisión para identificar las variables explicativas para un modelo final de regresión que pueda estimar la densidad mamográfica, calculando la potencia estadística del modelo para valorar posible utilidad en la práctica clínica. Resultados: El 67,5% de la población tenía algún grado de sobrepeso. Las mujeres de mayor edad presentaron mayor porcentaje de masa grasa (p 0,02) y razón cintura/cadera (p 0,01). El 53,4% de las mujeres tenían una densidad mamográfica menor del 25%. El IMC y la edad presentaron relación lineal inversa con la densidad mamográfica, mientras el consumo de anticonceptivos orales de forma directa. La edad, el IMC, perímetro de cintura y talla del brasier fueron incluidas en el modelo final de estimación de porcentaje de densidad mamográfica. Conclusiones: La mayor proporción de mujeres se encontraron en rango de sobrepeso, al igual que con patrones de densidad mamográfica bajos (<50%). Se pudo proponer la inclusión de variables como la edad, el IMC, la talla del brasier y el perímetro de cintura en un modelo final de estimación de densidad mamográfica.

The current design of reinforcement learning methods requires extensive computational resources. Algorithms such as Deep Q-Network (DQN) have obtained outstanding results in advancing the field. However, the need to tune thousands of parameters and run millions of training episodes remains a significant challenge. This document proposes a comparative analysis between the Q-Learning algorithm, which laid the foundations for Deep Q-Learning, and our proposed method, termed M-Learning. The comparison is conducted using Markov Decision Processes with the delayed reward as a general test bench framework. Firstly, this document provides a full description of the main challenges related to implementing Q-Learning, particularly concerning its multiple parameters. Then, the foundations of our proposed heuristic are presented, including its formulation, and the algorithm is described in detail. The methodology used to compare both algorithms involved training them in the Frozen Lake environment. The experimental results, along with an analysis of the best solutions, demonstrate that our proposal requires fewer episodes and exhibits reduced variability in the outcomes. Specifically, M-Learning trains agents 30.7% faster in the deterministic environment and 61.66% faster in the stochastic environment. Additionally, it achieves greater consistency, reducing the standard deviation of scores by 58.37% and 49.75% in the deterministic and stochastic settings, respectively. The code will be made available in a GitHub repository upon this paper’s publication.

The current design of reinforcement learning methods demands exhaustive computing. Algorithms such as Deep Q-Network achieved outstanding results in the development of the area. However, the need for thousands of parameters and training episodes is still a problem. Thus, this document proposes a comparative analysis of the Q-Learning algorithm (the inception to create Deep Q Learning) and our proposed method termed M-Learning. The comparison among algorithms using Markov decision processes with delayed reward as a general testbench framework. Firstly, a full description of the main problems related to implementing Q-Learning, mainly about its multiple parameters. Then, the foundations of our proposed heuristic with its formulation and the whole algorithm were reported in detail. Finally, the methodology chosen to compare both algorithms was to train the algorithms in the Frozen Lake environment. The experimental results and an analysis of the best solutions found that our proposed algorithm highlights the differences in the number of episodes necessary and their standard variations. The code will be available on a GitHub repository once the paper is published.