Traffic sign recognition is still one of the challenging aspects of intelligent vehicle systems, mainly when processor or memory resources are limited. In this work, real-time traffic sign detection was evaluated using five YOLO model variants—Nano, Small, Medium, Large, and XLarge—across versions 8 to 11. All models were trained and validated with a custom dataset collected in a simulated urban environment designed to replicate FIRA competition tracks. The models were then deployed and tested on a 1:10 scale autonomous vehicle equipped with a mini PC running the detector in real time. Performance was compared using mAP@50–95, F1-score, inference latency, and preprocessing and postprocessing times. The authors also analyzed training behavior, focusing on convergence speed and stopping criteria. The experiments showed that YOLOv10 B achieved the highest performance across varying conditions, while YOLOv8 M provided a better balance between speed and accuracy. These results can help practitioners select appropriate YOLO architectures for embedded traffic sign recognition systems that must operate in real time on resource-constrained autonomous vehicles.

Early and accurate breast cancer detection is critical for patient outcomes. The Curated Breast Imaging Subset of the Digital Database for Screening Mammography (CBIS-DDSM) has been instrumental for computer-aided diagnosis (CAD) systems. However, the lack of a standardized preprocessing pipeline and consistent metadata has limited its utility for reproducible quantitative imaging or radiomics. This paper introduces CBIS-DDSM-R, an open-source, radiomics-ready extension of the original dataset. It provides an automated pipeline for preprocessing mammograms and extracts a standardized set of 93 radiomics features per lesion, adhering to Image Biomarker Standardisation Initiative (IBSI) guidelines using PyRadiomics. The resulting dataset combines clinical and radiomics data into a unified format, offering a robust benchmark for developing and validating reproducible radiomics models for breast cancer characterization.

Este artículo plantea el diseño e implementación de una aplicación móvil que mediante un sistema de visión por computadora permite realizar el conteo de colonias de bacterias en cultivos microbianos disminuyendo significativamente el tiempo de la cuantificación y generando un método estándar de conteo para dispositivos móviles con sistema operativo Android. Así como un simulador de colonias de bacterias que permite generar muestras controlando parámetros de crecimiento de bacterias con el fin de ayudar a probar la eficiencia de la aplicación de conteo de colonias de bacterias. This article proposes the design and implementation of a mobile application using a computer vision system that allows counting bacterial colonies in microbial cultures, decreasing significantly the time of quantification and generating a standard counting method for mobile devices running Android OS. As well as a bacterial colony simulator that allows generating samples controlling bacterial growth parameters in order to help to test the efficiency of the bacterial colony count application.Keywords: Android, Mobile Application, Microbiological Counting, Computer Vision

Lane keeping under degraded visual confidence remains challenging because most existing approaches focus either on improving lane-feature extraction or on estimating vehicle motion, while giving less attention to how unreliable visual measurements should modify the estimator–controller interaction in a physical closed-loop system. This paper presents a confidence-gated Kalman lane-keeping architecture for a 1:10-scale autonomous vehicle. The methodological contribution lies in the direct coupling of visual confidence, state estimation, and steering control: unreliable lane measurements are down-weighted through confidence-dependent measurement noise, while the propagated lane-relative state remains available to the controller. The primary experimental unit is the run, defined as one logged lap under one control configuration; frame-level summaries are retained only as historical reproducibility material. In the run-level comparison, the autonomous vision, inertial measurement unit (IMU)-feedforward, and Kalman filter (KF) group KF_G1—the first inferential KF generation—had lower mean absolute error than the human baseline, while vision and KF_G1 had overlapping run-level confidence intervals for absolute error. KF_G1 shifted the mean signed bias closer to the lane reference than the vision and IMU groups, but with higher run-level spread than vision. KF_G2, the second observational KF generation, is reported only as an observational generation comparison, so no causal claim is made for its estimator–controller correction weight $K_s=0.15$ setting. KF_G3, the single descriptive adverse-condition KF run, is reported without an estimable confidence interval or population-level adverse-illumination inference. A further limitation is that the inertial prediction pathway was inactive in the logged Kalman-filter runs because the inertial coupling coefficient $\alpha =0$. The main contribution is a reproducible run-level evaluation of confidence-gated estimator–controller coupling that distinguishes supported evidence from observational and descriptive cases.

Radiomic feature extraction constitutes a key step in quantitative medicalimaging pipelines, yet the reference implementation PyRadiomics operatesexclusively on the central processing unit, requiring several seconds perscan and limiting throughput in large-scale clinical studies; existingGPU-accelerated alternatives address at most two of the eight standardisedfeature classes and provide no formal numerical validation.This paper introduces \textbf{fastrad}, the first open-source, GPU-nativePython library implementing all eight PyRadiomics-compatible radiomic featureclasses — first-order statistics, shape (3D and 2D), Grey-Level Co-occurrenceMatrix (GLCM), Grey-Level Run-Length Matrix (GLRLM), Grey-Level Size-ZoneMatrix (GLSZM), Grey-Level Dependence Matrix (GLDM), and NeighbourhoodGrey-Tone Difference Matrix (NGTDM) — within a unified PyTorch framework inwhich all computation operates on \texttt{torch.Tensor} objects withtransparent \texttt{auto}, \texttt{cuda}, and \texttt{cpu} device routing.Correctness was validated against the Image Biomarker StandardisationInitiative (IBSI) Phase~1 digital phantom across 105 reference features andagainst PyRadiomics on a clinical CT image from the TCIA Non-Small Cell LungCancer Radiogenomics dataset; runtime benchmarks were conducted on an NVIDIARTX~4070~Ti against PyRadiomics single-thread and 32-thread configurations,and scan-rescan reproducibility was assessed on the RIDER Lung CT dataset viaintraclass correlation coefficients.fastrad achieved 100\% IBSI compliance with maximum absolute relativedeviations at machine-epsilon ($\leq 10^{-13}$\%), and all 105 featuresagreed with PyRadiomics to within $10^{-11}$ on the clinical image, wellbelow the $10^{-4}$ parity threshold.End-to-end GPU extraction completed in $0.116$~s versus $2.90$~s forPyRadiomics, yielding a $\mathbf{25.0\times}$ speedup, with per-class GPUgains from $12.9\times$ (GLRLM) to $49.3\times$ (first-order); fastradsingle-thread CPU exceeded PyRadiomics 32-thread by $2.63\times$ on x86 and$3.56\times$ on Apple Silicon, and full-pipeline peak VRAM was $654.78$~MB.fastrad is openly available on PyPI (\texttt{pip install fastrad}) and GitHubunder the Apache 2.0 licence, providing a drop-in, numerically interoperablereplacement for PyRadiomics in high-throughput and machine learning radiomicsworkflows. fastrad is openly available at\url{https://github.com/helloerikaaa/fastrad} and on PyPI under the Apache 2.0 license.