Lightweight Deep Learning for Atrial Fibrillation Detection: Efficient Models for Wearable DevicesThe timely and accurate detection of atrial fibrillation (AF) is crucial for an early intervention and for reducing the associated health risks. Wearable technology has emerged as a viable solution for continuous AF monitoring, but deploying accurate AF detection models on resource-constrained devices remains a challenge due to the high computational and memory demands. This study proposes a lightweight and efficient deep learning approach for real-time AF diagnosis in portable devices. We designed a series of convolutional neural network (CNN) models optimized for high accuracy while maintaining a minimal computational footprint. To further enhance efficiency, we explored deep learning compression techniques, including pruning, quantization, and knowledge distillation. Our results demonstrate that the proposed models achieve state-of-the-art accuracy while significantly reducing memory usage and computational complexity, making them suitable for real-time deployment. Additionally, we validated their feasibility by implementing them on a microcontroller, showcasing their practicality for wearable applications. This research paves the way for accessible, low-power, and high-accuracy AF detection in real-world settings, enabling early diagnosis and timely medical intervention without the need for continuous clinical supervision.
Air quality monitoring at Universidad Iberoamericana, Mexico City: Ozone and meteorological data during the rainy and ozone seasons (Mar–Nov 2024)This dataset contains hourly-averaged measurements of ozone and meteorological variables collected between March 1 and November 30, 2024, at Universidad Iberoamericana, Mexico City (latitude: 19.372292, longitude:99.263679). The data capture conditions across both the ozone season (March-June) and the rainy season (May-October) in a high-altitude urban environment. Variables include ozone concentration (ppb), temperature ( °C), relative humidity (%), wind direction (°), wind speed (m/s), atmospheric pressure (hPa), accumulated precipitation (mm), and solar radiation (W/m²). The dataset was obtained using a calibrated low-cost sensor system, which transmitted data to the cloud for subsequent download in CSV format. This open-access dataset is intended to support research on the interactions between meteorological conditions and air quality. It can be reused in studies focused on air quality modelling, atmospheric dynamics, environmental monitoring, or urban climate in megacities with similar characteristics.
Imaginary chaos fightEste artigo descreve o desenvolvimento do jogo "Imaginary Chaos Fight", um PvP 2D inspirado em títulos como Brawl Stars e Super Smash Bros Ultimate, com diversos personagens com mecânicas, visuais, vantagens e desvantagens, o que permite a adoção das mais diversas estratégias. Imaginary chaos fight é um jogo onde 6 lutadores brigam constantemente para provar o seu valor um para os outros com o sonho de um dia ser declarado o mais forte, o único problema é que nenhum deles jamais derrotou todos os outros 5 sem antes perder. Originado de um projeto de extensão focado em desenvolvimento de jogos digitais. Utilizando o GameMaker Studio 2, o jogo foi concebido com mecânicas simples e alto fator de rejogabilidade para jogadores casuais. A criação dos personagens seguiu padrões de design de software, promovendo a eficiência na reutilização de código. A parte gráfica foi desenvolvida com o Aseprite, adotando o estilo pixel art. O jogo encontra-se em fase de desenvolvimento.