Background: Industry 4.0 technologies have been widely used in the railway industry, focusing mainly on maintenance and control tasks necessary in the railway infrastructure. Given the great potential that these technologies offer, the scientific community has come to use them in varied ways to solve a wide range of problems such as train failures, train station security, rail system control and communication in hard-to-reach areas, among others. For this reason, this paper aims to answer the following research questions: what are the main issues in the railway transport industry, what are the technologic strategies that are currently being used to solve these issues and what are the technologies from industry 4.0 that are used in the railway transport industry to solve the aforementioned issues? Methods: This study adopts a systematic literature review approach. We searched the Science Direct and Web of Science database inception from January 2017 to November 2021. Studies published in conferences or journals written in English or Spanish were included for initial process evaluation. The initial included papers were analyzed by authors and selected based on whether they helped answer the proposed research questions or not. Results: Of the recovered 515 articles, 109 were eligible, from which we could identify three main application domains in the railway industry: monitoring, decision and planification techniques, and communication and security. Regarding industry 4.0 technologies, we identified 9 different technologies applied in reviewed studies: Artificial Intelligence (AI), Internet of Things (IoT), Cloud Computing, Big Data, Cybersecurity, Modelling and Simulation, Smart Decision Support Systems (SDSS), Computer Vision and Virtual Reality (VR). This study is, to our knowledge, one of the first to show how industry 4.0 technologies are currently being used to tackle railway industry problems and current application trends in the scientific community, which is highly useful for the development of future studies and more advanced solutions. Funding: Colombian national organizations Minciencias and the Mining-Energy Planning Unit.

In recent years, the use of deep learning-based models for developing advanced healthcare systems has been growing due to the results they can achieve. However, the majority of the proposed deep learning-models largely use convolutional and pooling operations, causing a loss in valuable data and focusing on local information. In this paper, we propose a deep learning-based approach that uses global and local features which are of importance in the medical image segmentation process. In order to train the architecture, we used extracted three-dimensional (3D) blocks from the full magnetic resonance image resolution, which were sent through a set of successive convolutional neural network (CNN) layers free of pooling operations to extract local information. Later, we sent the resulting feature maps to successive layers of self-attention modules to obtain the global context, whose output was later dispatched to the decoder pipeline composed mostly of upsampling layers. The model was trained using the Mindboggle-101 dataset. The experimental results showed that the self-attention modules allow segmentation with a higher Mean Dice Score of 0.90 ± 0.036 compared with other UNet-based approaches. The average segmentation time was approximately 0.038 s per brain structure. The proposed model allows tackling the brain structure segmentation task properly. Exploiting the global context that the self-attention modules incorporate allows for more precise and faster segmentation. We segmented 37 brain structures and, to the best of our knowledge, it is the largest number of structures under a 3D approach using attention mechanisms.

Noradrenaline (also known as norepinephrine) is released throughout most of the central nervous system by neurons in the locus coeruleus ^1–4 . Here we identified a relationship between the morphologies, gene expression and activity of noradrenergic neurons in locus coeruleus in mice. Axonal projections of individual neurons were extensive but largely confined to subsets of brain regions. Axonal projections and graded gene expression correlated with locations of cell bodies in locus coeruleus. In a behavioural task requiring continuous learning from actions, dorsal locus coeruleus neurons projecting to isocortex were activated when mice switched choices and by reward-prediction-error signals that drive learning. Background activity of neurons in ventral locus coeruleus was higher when mice ignored stimuli indicating potential reward availability. These observations reveal a topographically organized structure and function of a neurotransmitter system and show that it contains learning signals for flexible behaviour.

La resonancia magnetica cerebral es clave para el diagnostico asistido por computador, pero la segmentacion manual es costosa, lenta y dependiente del operador. Se presenta un pipeline para segmentar materia gris (SG) que combina un preprocesamiento minimo (armonizacion de forma, teselado por parches) con un codificador-decodificador 3D que integra autoatencion global y conexiones de salto multiescala. En el conjunto de prueba MRBrainS18, el modelo logra un Dice de 0.650 +/- 0.043, un IoU de 0.483 +/- 0.046, una precision de 0.700 +/- 0.046 y un recall de 0.610 +/- 0.059; las distribuciones por sujeto son compactas, lo que refleja consistencia entre casos. Las metricas de solapamiento muestran ausencia de relacion monotona con la distancia de Hausdorff (r ~ 0.08), lo que resalta la sensibilidad de los bordes aun con buen solapamiento global. El analisis de Bland-Altman evidencia un sesgo volumetrico negativo (RVE = -12.54 % +/- 9.53 %), consistente con una precision mayor que el recall y con predicciones conservadoras en las interfaces tisulares. Las lineas base clasicas (Otsu multiumbral y crecimiento de regiones), evaluadas en un subconjunto representativo, presentan menor rendimiento; el modelo mejora Dice e IoU en un 50 % frente a la mejor linea base.

Light microscopy is routinely used to explore the cellular and molecular structure of tissues, but the scale and complexity of data remains a bottleneck for discovery. We introduce 3D-MAESTRO (3D-Microscopy Automation and Execution with Scalable Tools, Rendering, and Orchestration): an automated image processing workflow for large-scale microscopy data, built for scalable execution across cloud and local computing environments. 3D-MAESTRO orchestrates denoising, stitching of image tiles, atlas registration, and segmentation. Its modular architecture permits the integration and benchmarking of new packages, ensuring that performance evolves as more efficient or accurate algorithms emerge. We introduce a new 3D image template for automated registration of mouse brains cleared with aqueous reagents and an efficient method for detection of fluorescent cells. We apply 3D-MAESTRO to lightsheet images of whole mouse brains in the context of diverse anatomical and functional experiments, illustrating high-throughput and reproducible mapping of microscopic structures across the brain.