Pose Estimation of a Cobot Implemented on a Small AI-Powered Computing System and a Stereo Camera for Precision Evaluation.

The precision of robotic manipulators in the industrial or medical field is very important, especially when it comes to repetitive or exhaustive tasks. Geometric deformations are the most common in this field. For this reason, new robotic vision techniques have been proposed, including 3D methods that made it possible to determine the geometric distances between the parts of a robotic manipulator. The aim of this work is to measure the angular position of a robotic arm with six degrees of freedom. For this purpose, a stereo camera and a convolutional neural network algorithm are used to reduce the degradation of precision caused by geometric errors. This method is not intended to replace encoders, but to enhance accuracy by compensating for degradation through an intelligent visual measurement system. The camera is tested and the accuracy is about one millimeter. The implementation of this method leads to better results than traditional and simple neural network methods.

A dataset on eye movement tracking during the resolution of neuropsychological tests on a screen.

The dataset provides data obtained with eye-tracking while 55 volunteers solved 3 distinct neuropsychological tests on a screen inside a closed room. Among the 55 volunteers, 22 were women and 33 were men, all with ages ranging between 9 and 50, and 5 of whom were diagnosed with Attention Deficit Hyperactivity Disorder (ADHD) [1]. The eye-tracker used for the collection of the data was an EyeTribe, which has a sampling rate of 60 Hz and an average visual angle between 0.5 and 1, which correspond to an on-screen error between 0.5 and 1cm (0.1969 to 0.393 inches aprox) respectively, when the distance to the user is around 60cm (23.62 in) [2], which was the case during the collection of these data. The neuropsychological tests were implemented in a software named NEURO-INNOVA KIDS® [3], which are the following: a domino test adapted from the D-48 intelligence test [4], an adaptation of the MASMI test consisting of unfolded cubes [5], the figures series completion test adapted from [6], and the Poppelreuter figures test [7]. Before each of the tests, a calibration process was performed, ensuring that the visual angle error was less than or equal to 0.5 cm (0.1969 in), which is considered an acceptable calibration. The collective mean duration of the four administered tests amounted to 20 minutes. This dataset exhibits significant promise for potential utilization due to the extensive prevalence of these neuropsychological assessments among healthcare practitioners for evaluating diverse cognitive faculties in individuals. Moreover, it has been empirically established that poor performance on these tests is associated with attention deficits [8].

Non-Invasive Prediction of Choledocholithiasis Using 1D Convolutional Neural Networks and Clinical Data.

This paper introduces a novel one-dimensional convolutional neural network that utilizes clinical data to accurately detect choledocholithiasis, where gallstones obstruct the common bile duct. Swift and precise detection of this condition is critical to preventing severe complications, such as biliary colic, jaundice, and pancreatitis. This cutting-edge model was rigorously compared with other machine learning methods commonly used in similar problems, such as logistic regression, linear discriminant analysis, and a state-of-the-art random forest, using a dataset derived from endoscopic retrograde cholangiopancreatography scans performed at Olive View-University of California, Los Angeles Medical Center. The one-dimensional convolutional neural network model demonstrated exceptional performance, achieving 90.77% accuracy and 92.86% specificity, with an area under the curve of 0.9270. While the paper acknowledges potential areas for improvement, it emphasizes the effectiveness of the one-dimensional convolutional neural network architecture. The results suggest that this one-dimensional convolutional neural network approach could serve as a plausible alternative to endoscopic retrograde cholangiopancreatography, considering its disadvantages, such as the need for specialized equipment and skilled personnel and the risk of postoperative complications. The potential of the one-dimensional convolutional neural network model to significantly advance the clinical diagnosis of this gallstone-related condition is notable, offering a less invasive, potentially safer, and more accessible alternative.

Investigation of Using Hyperspectral Vegetation Indices to Assess Brassica Downy Mildew.

Downy mildew caused by Hyaloperonospora brassicae is a severe disease in Brassica oleracea that significantly reduces crop yield and marketability. This study aims to evaluate different vegetation indices to assess different downy mildew infection levels in the Brassica variety Mildis using hyperspectral data. Artificial inoculation using H. brassicae sporangia suspension was conducted to induce different levels of downy mildew disease. Spectral measurements, spanning 350 nm to 1050 nm, were conducted on the leaves using an environmentally controlled setup, and the reflectance data were acquired and processed. The Successive Projections Algorithm (SPA) and signal sensitivity calculation were used to extract the most informative wavelengths that could be used to develop downy mildew indices (DMI). A total of 37 existing vegetation indices and three proposed DMIs were evaluated to indicate downy mildew (DM) infection levels. The results showed that the classification using a support vector machine achieved accuracies of 71.3%, 80.7%, and 85.3% for distinguishing healthy leaves from DM1 (early infection), DM2 (progressed infection), and DM3 (severe infection) leaves using the proposed downy mildew index. The proposed new downy mildew index potentially enables the development of an automated DM monitoring system and resistance profiling in Brassica breeding lines.

Modeling of Particulate Pollutants Using a Memory-Based Recurrent Neural Network Implemented on an FPGA.

The present work describes the training and subsequent implementation on an FPGA board of an LSTM neural network for the modeling and prediction of the exceedances of criteria pollutants such as nitrogen dioxide (NO2), carbon monoxide (CO), and particulate matter (PM10 and PM2.5). Understanding the behavior of pollutants and assessing air quality in specific geographical regions is crucial. Overexposure to these pollutants can cause harm to both natural ecosystems and living organisms, including humans. Therefore, it is essential to develop a solution that can accurately evaluate pollution levels. One potential approach is to implement a modified LSTM neural network on an FPGA board. This implementation obtained an 11% improvement compared to the original LSTM network, demonstrating that the proposed architecture is able to maintain its functionality despite reducing the number of neurons in its initial layers. It shows the feasibility of integrating a prediction network into a limited system such as an FPGA board, but easily coupled to a different system. Importantly, this implementation does not compromise the prediction accuracy for both 24 h and 72 h time frames, highlighting an opportunity for further enhancement and refinement.

Performance Evaluation of Different Object Detection Models for the Segmentation of Optical Cups and Discs.

Glaucoma is an eye disease that gradually deteriorates vision. Much research focuses on extracting information from the optic disc and optic cup, the structure used for measuring the cup-to-disc ratio. These structures are commonly segmented with deeplearning techniques, primarily using Encoder-Decoder models, which are hard to train and time-consuming. Object detection models using convolutional neural networks can extract features from fundus retinal images with good precision. However, the superiority of one model over another for a specific task is still being determined. The main goal of our approach is to compare object detection model performance to automate segment cups and discs on fundus images. This study brings the novelty of seeing the behavior of different object detection models in the detection and segmentation of the disc and the optical cup (Mask R-CNN, MS R-CNN, CARAFE, Cascade Mask R-CNN, GCNet, SOLO, Point_Rend), evaluated on Retinal Fundus Images for Glaucoma Analysis (REFUGE), and G1020 datasets. Reported metrics were Average Precision (AP), F1-score, IoU, and AUCPR. Several models achieved the highest AP with a perfect 1.000 when the threshold for IoU was set up at 0.50 on REFUGE, and the lowest was Cascade Mask R-CNN with an AP of 0.997. On the G1020 dataset, the best model was Point_Rend with an AP of 0.956, and the worst was SOLO with 0.906. It was concluded that the methods reviewed achieved excellent performance with high precision and recall values, showing efficiency and effectiveness. The problem of how many images are needed was addressed with an initial value of 100, with excellent results. Data augmentation, multi-scale handling, and anchor box size brought improvements. The capability to translate knowledge from one database to another shows promising results too.

Performance evaluation of a recurrent deep neural network optimized by swarm intelligent techniques to model particulate matter.

Atmospheric pollution refers to the presence of substances in the air such as particulate matter (PM) which has a negative impact in population ́s health exposed to it. This makes it a topic of current interest. Since the Metropolitan Zone of the Valley of Mexico's geographic characteristics do not allow proper ventilation and due to its population's density a significant quantity of poor air quality events are registered. This paper proposes a methodology to improve the forecasting of PM10 and PM2.5, in largely populated areas, using a recurrent long-term/short-term memory (LSTM) network optimized by the Ant Colony Optimization (ACO) algorithm. The experimental results show an improved performance in reducing the error by around 13.00% in RMSE and 14.82% in MAE using as reference the averaged results obtained by the LSTM deep neural network. Overall, the current study proposes a methodology to be studied in the future to improve different forecasting techniques in real-life applications where there is no need to respond in real time.Implications: This contribution presents a methodology to deal with the highly non-linear modeling of airborne particulate matter (both PM10 and PM2.5). Most linear approaches to this modeling problem are often not accurate enough when dealing with this type of data. In addition, most machine learning methods require extensive training or have problems when dealing with noise embedded in the time-series data. The proposed methodology deals with this data in three stages: preprocessing, modeling, and optimization. In the preprocessing stage, data is acquired and imputed any missing data. This ensures that the modeling process is robust even when there are errors in the acquired data and is invalid, or the data is missing. In the modeling stage, a recurrent deep neural network called LSTM (Long-Short Term Memory) is used, which shows that regardless of the monitoring station and the geographical characteristics of the site, the resulting model shows accurate and robust results. Furthermore, the optimization stage deals with enhancing the capability of the data modeling by using swarm intelligence algorithms (Ant Colony Optimization, in this case). The results presented in this study were compared with other works that presented traditional algorithms, such as multi-layer perceptron, traditional deep neural networks, and common spatiotemporal models, which show the feasibility of the methodology presented in this contribution. Lastly, the advantages of using this methodology are highlighted.

Implementation of ANN-Based Auto-Adjustable for a Pneumatic Servo System Embedded on FPGA.

Artificial intelligence techniques for pneumatic robot manipulators have become of deep interest in industrial applications, such as non-high voltage environments, clean operations, and high power-to-weight ratio tasks. The principal advantages of this type of actuator are the implementation of clean energies, low cost, and easy maintenance. The disadvantages of working with pneumatic actuators are that they have non-linear characteristics. This paper proposes an intelligent controller embedded in a programmable logic device to minimize the non-linearities of the air behavior into a 3-degrees-of-freedom robot with pneumatic actuators. In this case, the device is suitable due to several electric valves, direct current motors signals, automatic controllers, and several neural networks. For every degree of freedom, three neurons adjust the gains for each controller. The learning process is constantly tuning the gain value to reach the minimum of the mean square error. Results plot a more appropriate behavior for a transitive time when the neurons work with the automatic controllers with a minimum mean error of ±1.2 mm.

Immune Profiles Identification by Vaccinomics After MVA Immunization in Randomized Clinical Study.

Background: Our previous work has demonstrated the benefits of transcutaneous immunization in targeting Langerhans cells and preferentially inducing CD8 T-cell responses. Methods: In this randomized phase Ib clinical trial including 20 HIV uninfected volunteers, we compared the safety and immunogenicity of the MVA recombinant vaccine expressing HIV-B antigen (MVA-B) by transcutaneous and intramuscular routes. We hypothesized that the quality of innate and adaptive immunity differs according to the route of immunization and explored the quality of the vector vaccine-induced immune responses. We also investigated the early blood transcriptome and serum cytokine levels to identify innate events correlated with the strength and quality of adaptive immunity. Results: We demonstrate that MVA-B vaccine is safe by both routes, but that the quality and intensity of both innate and adaptive immunity differ significantly. Transcutaneous vaccination promoted CD8 responses in the absence of antibodies and slightly affected gene expression, involving mainly genes associated with metabolic pathways. Intramuscular vaccination, on the other hand, drove robust changes in the expression of genes involved in IL-6 and interferon signalling pathways, mainly those associated with humoral responses, and also some levels of CD8 response. Conclusion: Thus, vaccine delivery route perturbs early innate responses that shape the quality of adaptive immunity. Clinical Trial Registration: http://ClinicalTrials.gov, identifier PER-073-13.

Expression of CD11c Is Associated with Unconventional Activated T Cell Subsets with High Migratory Potential.

CD11c is an α integrin classically employed to define myeloid dendritic cells. Although there is little information about CD11c expression on human T cells, mouse models have shown an association of CD11c expression with functionally relevant T cell subsets. In the context of genital tract infection, we have previously observed increased expression of CD11c in circulating T cells from mice and women. Microarray analyses of activated effector T cells expressing CD11c derived from naïve mice demonstrated enrichment for natural killer (NK) associated genes. Here we find that murine CD11c+ T cells analyzed by flow cytometry display markers associated with non-conventional T cell subsets, including γδ T cells and invariant natural killer T (iNKT) cells. However, in women, only γδ T cells and CD8+ T cells were enriched within the CD11c fraction of blood and cervical tissue. These CD11c+ cells were highly activated and had greater interferon (IFN)-γ secretory capacity than CD11c- T cells. Furthermore, circulating CD11c+ T cells were associated with the expression of multiple adhesion molecules in women, suggesting that these cells have high tissue homing potential. These data suggest that CD11c expression distinguishes a population of circulating T cells during bacterial infection with innate capacity and mucosal homing potential.

Type 1 Diabetes Prevention in NOD Mice by Targeting DPPIV/CD26 Is Associated with Changes in CD8⁺T Effector Memory Subset.

CD26 is a T cell activation marker consisting in a type II transmembrane glycoprotein with dipeptidyl peptidase IV (DPPIV) activity in its extracellular domain. It has been described that DPPIV inhibition delays the onset of type 1 diabetes and reverses the disease in non-obese diabetic (NOD) mice. The aim of the present study was to assess the effect of MK626, a DPPIV inhibitor, in type 1 diabetes incidence and in T lymphocyte subsets at central and peripheral compartments. Pre-diabetic NOD mice were treated with MK626. Diabetes incidence, insulitis score, and phenotyping of T lymphocytes in the thymus, spleen and pancreatic lymph nodes were determined after 4 and 6 weeks of treatment, as well as alterations in the expression of genes encoding ß-cell autoantigens in the islets. The effect of MK626 was also assessed in two in vitro assays to determine proliferative and immunosuppressive effects. Results show that MK626 treatment reduces type 1 diabetes incidence and after 6 weeks of treatment reduces insulitis. No differences were observed in the percentage of T lymphocyte subsets from central and peripheral compartments between treated and control mice. MK626 increased the expression of CD26 in CD8+ T effector memory (TEM) from spleen and pancreatic lymph nodes and in CD8+ T cells from islet infiltration. CD8+TEM cells showed an increased proliferation rate and cytokine secretion in the presence of MK626. Moreover, the combination of CD8+ TEM cells and MK626 induces an immunosuppressive response. In conclusion, treatment with the DPPIV inhibitor MK626 prevents experimental type 1 diabetes in association to increase expression of CD26 in the CD8+ TEM lymphocyte subset. In vitro assays suggest an immunoregulatory role of CD8+ TEM cells that may be involved in the protection against autoimmunity to ß pancreatic islets associated to DPPIV inhibitor treatment.

Alignment of the measurement scale mark during immersion hydrometer calibration using an image processing system.

The present work presents an improved method to align the measurement scale mark in an immersion hydrometer calibration system of CENAM, the National Metrology Institute (NMI) of Mexico, The proposed method uses a vision system to align the scale mark of the hydrometer to the surface of the liquid where it is immersed by implementing image processing algorithms. This approach reduces the variability in the apparent mass determination during the hydrostatic weighing in the calibration process, therefore decreasing the relative uncertainty of calibration.

Morbilidad materno-perinatal de la rotura prematura de membranas en el embarazo pretérmino en el Hospital Nacional Dos de Mayo. 2011-2012 / Maternal and perinatal morbidity of premature rupture of membranes in preterm pregnancy at the National Hospital Dos de Mayo. 2011-2012

Antecedentes: La morbilidad asociada a la ruptura prematura de membranas puede ser mayor en neonatos pretérmino. Objetivo: Determinar la morbilidad materno-perinatal de la Rotura Prematura de Membranas Pretérmino en el Hospital Nacional Dos de Mayo en el periodo 2011-2012. Métodos: Revisión documentaria de las historias clínicas de las madres y neonatos prematuros que cumplieron criterios de selección, encontrando 170 casos en el periodo de estudio. Se muestran los resultados mediante estadística descriptiva. Resultados: La edad de las madres de neonatos prematuros con RPM fue 25,52 años. La forma de terminación del parto en 34,12 por ciento fue la vía vaginal, y en 65,88 por ciento por cesárea. El tiempo de latencia desde la RPM al parto en 9,41 por ciento fue dentro de las 6 primeras horas, en 21,18 por ciento dentro de las primeras 12 horas, y en 69,41 por ciento luego de las 24horas. Se emplearon corticoides para maduración pulmonar en 46,47 por ciento de casos, todos ellos con dexametasona, y en 27,65 por ciento se completó cuatro cursos. El 47,65 por ciento de neonatos fueron varones y 52,35 por ciento mujeres. En 82,35 por ciento de casos el líquido amniótico fue claro, 15,88 por ciento fue verde claro y en 1,76 por ciento fue purulento. No se usó antibióticos en 10 por ciento de casos, se usó sólo un antibiótico en 48,24 por ciento, dos antibióticos en 24,71 por ciento y tres antibióticos en 17,06 por ciento; el antibiótico más empleado fue la cefazolina. En 45,29 por ciento de madres se presentó endometritis; la duración de la hospitalización en las madres fue 5,73 días. El 14,71 por ciento de neonatos pretérmino fue llevado a alojamiento conjunto con la madre, 29,41 por ciento llegó a cuidados intermedios y 55,88 por ciento pasó a UCI; la indicación de hospitalización en UCI fue la sepsis en 55,79 por ciento, insuficiencia respiratoria en 20 por ciento, prematuridad extrema en 12,63 por ciento. Un 22,35 por ciento de neonatos no requirieron de apoyo...

Background: The morbidity associated with premature rupture of membranes may be higher in preterm infants. Objective: To determine maternal and perinatal morbidity of Premature Rupture of Membranes at the National Hospital Dos de Mayo in the period 2011-2012. Methods: Review of medical records documentary of mothers and preterm infants who met selection criteria, finding 170 cases in the study period. Results are shown using descriptive statistics. Results: The age of the mothers of preterm infants with RPM was 25.52 years. The manner of termination of labor was vaginal in 34.12 per cent, and by cesarean in 65.88 per cent. The latency from the RPM to birth was 9.41 per cent within the first 6 hours, in 21.18 per cent within the first 12 hours, and 69.41 per cent after 24 hours. Corticosteroids were used for 11Ing maturation in 46.47 per cent of cases, all with dexamethasone, and 27.65 per cent completed four courses. The 47.65 per cent of infants were male and 52.35 per cent female. In 82.35 per cent of cases the amniotic fluid was clear, light green in 15.88 per cent and in 1.76 per cent was purulent. No antibiotics were used in 10 per cent of cases, only one antibiotic was used in 48.24 per cent, two antibiotics in 24.71 per cent and three in 17.06 per cent, the most used antibiotic was cefazolin. In 45.29 per cent of mothers showed endometritis, the duration of hospitalization in mothers was 5.73 days. The 14.71 per cent of preterm infants was admitted with mother, intermediate care in 29.41 per cent and 55.88 per cent went to ICU, the indication for hospitalization in ICU was in 55.79 per cent sepsis, respiratory failure in 20 per cent, and in 12.63 per cent extreme prematurity. A 22.35 per cent of neonates did not required ventilatory support, 39.41 per cent required mechanical ventilation and 38.24 per cent used CPAP. In only 5.29 per cent of children there were no complications, a 91.18 per cent of cases had metabolic disorders, 90 per cent developed...