Optical Flow-Based Obstacle Detection for Mid-Air Collision Avoidance.

The sky may seem big enough for two flying vehicles to collide, but the facts show that mid-air collisions still occur occasionally and are a significant concern. Pilots learn manual tactics to avoid collisions, such as see-and-avoid, but these rules have limitations. Automated solutions have reduced collisions, but these technologies are not mandatory in all countries or airspaces, and they are expensive. These problems have prompted researchers to continue the search for low-cost solutions. One attractive solution is to use computer vision to detect obstacles in the air due to its reduced cost and weight. A well-trained deep learning solution is appealing because object detection is fast in most cases, but it relies entirely on the training data set. The algorithm chosen for this study is optical flow. The optical flow vectors can help us to separate the motion caused by camera motion from the motion caused by incoming objects without relying on training data. This paper describes the development of an optical flow-based airborne obstacle detection algorithm to avoid mid-air collisions. The approach uses the visual information from a monocular camera and detects the obstacles using morphological filters, optical flow, focus of expansion, and a data clustering algorithm. The proposal was evaluated using realistic vision data obtained with a self-developed simulator. The simulator provides different environments, trajectories, and altitudes of flying objects. The results showed that the optical flow-based algorithm detected all incoming obstacles along their trajectories in the experiments. The results showed an F-score greater than 75% and a good balance between precision and recall.

Effect of photostimulation through red LED light radiation on natural fermentation of table olives: An innovative case study with Negrinha the Freixo variety.

In this study, for the first time, red LED light radiation was applied to the fermentation process of table olives using the Negrinha de Freixo variety. Photostimulation using LED light emission (630 ± 10 nm) is proposed to shorten and speed up this stage and reduce time to market. Several physical-chemical characteristics and microorganisms (total microbial count of mesophilic aerobic, molds, yeasts, and lactic acid bacteria) and their sequence during fermentation were monitored. The fermentation occurred for 122 days, with two irradiation periods for red LED light. The nutritional composition and sensory analysis were performed at the end of the process. Fermentation under red LED light increased the viable yeast and lactic acid bacteria (LAB) cell counts and decreased the total phenolics in olives. Even though significant differences were observed in some color parameters, the hue values were of the same order of magnitude and similar for both samples. Furthermore, the red LED light did not play a relevant change in the texture profile, preventing the softening of the fruit pulp. Similarly, LED light did not modify the existing type of microflora but increased species abundance, resulting in desirable properties and activities. The species identified were yeasts - Candida boidinii, Pichia membranifaciens, and Saccharomyces cerevisiae, and bacteria - Lactobacillus plantarum and Leuconostoc mesenteroides, being the fermentative process dominated by S. cerevisiae and L. plantarum. At the end of fermentation (122 days), the irradiated olives showed less bitterness and acidity, higher hardness, and lower negative sensory attributes than non-irradiated. Thus, the results of this study indicate that red LED light application can be an innovative technology for table olives production.

Analysis of domain shift in whole prostate gland, zonal and lesions segmentation and detection, using multicentric retrospective data.

Despite being one of the most prevalent forms of cancer, prostate cancer (PCa) shows a significantly high survival rate, provided there is timely detection and treatment. Computational methods can help make this detection process considerably faster and more robust. However, some modern machine-learning approaches require accurate segmentation of the prostate gland and the index lesion. Since performing manual segmentations is a very time-consuming task, and highly prone to inter-observer variability, there is a need to develop robust semi-automatic segmentation models. In this work, we leverage the large and highly diverse ProstateNet dataset, which includes 638 whole gland and 461 lesion segmentation masks, from 3 different scanner manufacturers provided by 14 institutions, in addition to other 3 independent public datasets, to train accurate and robust segmentation models for the whole prostate gland, zones and lesions. We show that models trained on large amounts of diverse data are better at generalizing to data from other institutions and obtained with other manufacturers, outperforming models trained on single-institution single-manufacturer datasets in all segmentation tasks. Furthermore, we show that lesion segmentation models trained on ProstateNet can be reliably used as lesion detection models.

Effect of the cleaning and disinfection methods on the hygienic conditions of fermentation tanks of table olives (Olea europaea L.) Negrinha de Freixo cultivar.

This study aimed to evaluate and identify the microbial community attached to the surfaces of fermenter tanks used in table olive Negrinha de Freixo cultivar processing through molecular analysis and verify if the cleaning/disinfection was done correctly. Four fermentation tanks previously used in table olive processing were sampled at three different inside areas: upper, middle, and lower. Before sampling, four cleaning/disinfection methods were applied to the tanks, including (i) pressurised water; (ii) a disinfectant product used to clean bowls (Vasiloxe); (iii) 10% sodium hydroxide solution (caustic soda liquid); and (iv) a disinfectant product used by the wine industry (Hosbit). For each sample collected, mesophilic aerobic bacteria, yeast and moulds (YMC), lactic acid bacteria (LAB), as well as total coliforms (TC) and Pseudomonas aeruginosa were evaluated. The results showed significant differences between the different cleaning/disinfection methods applied. The fermenter sanitised with only pressurised water showed a greater abundance of microorganisms than the others. Mesophilic aerobic bacteria were the predominant population, with counts ranging between 2.63 and 5.56 log10 CFU/100 cm2, followed by the moulds (3.11-5.03 log10 CFU/100 cm2) and yeasts (2.42-5.12 log10 CFU/100 cm2). High diversity of microbial communities was observed between the different fermenter tanks. The most abundant species belonged to Aureobasidium, Bacillaceae, Cladosporium, and Rhodotorula genera. LAB, TC, and P. aeruginosa were not detected. This study hopes to improve hygienic conditions and increase the quality assurance and safety of the final product.

Application of the FTIR technique as a non-invasive tool to discriminate Portuguese olive oils with Protected Designation of Origin.

Three Portuguese olive oils with PDO ('Azeite do Alentejo Interior', 'Azeites da Beira Interior' and 'Azeite de Trás-os-Montes') were studied considering their physicochemical quality, antioxidant capacity, oxidative stability, total phenols content, gustatory sensory sensations and Fourier transform infrared (FTIR) spectra. All oils fulfilled the legal thresholds of EVOOs and the PDO's specifications. Olive oils from 'Azeite da Beira Interior' and 'Azeite de Trás-os-Montes' showed greater total phenols contents and antioxidant capacities, while 'Azeites da Beira Interior' presented higher oxidative stabilities. Linear discriminant models were developed using FTIR spectra (transmittance and the 1st and 2nd derivatives), allowing the correct identification of the oils' PDO (100 % sensitivity and specificity, repeated K-fold-CV). This study also revealed that multiple linear regression models, based on FTIR transmittance data, could predict the sweet, bitter, and pungent intensities of the PDO oils (R2 ≥0.979±0.016; RMSE≤0.26±0.05, repeated K-fold-CV). This demonstrates the potential of using FTIR as a non-destructive technique for authenticating oils with PDO.

Requests, Issuing, and Hospital Admissions Regarding Mental Health Warrants Before and During the COVID-19 Pandemic in a Region in Portugal.

INTRODUCTION: Mental health warrants exist in most countries and are issued when patients have severe mental illness, refuse treatment, and present a serious risk to themselves or others. We describe the epidemiology of mental health warrant requests received, and warrants issued by a Public Health Unit in a Portuguese region, as well as subsequent hospital admissions before and during the COVID-19 pandemic. METHODS: We used routine administrative data of mental health warrant request entries from a Public Health Unit serving a population of 219 739 individuals and compared the average of monthly requests, issued warrants, and hospital admissions during two separate periods (January 2013 to January 2021 and February 2021 to October 2022) as well as the proportion of warrants issued, hospital admissions among requests, and other patient characteristics. We identified factors associated with hospital admissions among the requests using logistic regression. RESULTS: Monthly average warrant requests, issued warrants and hospital admissions increased after February 2021 (x̄ 2.87 vs 7.09 p < 0.001; x̄ 2.67 vs 6.42 p < 0.001; x̄ 1.55 vs 3.58 p < 0.001). We found no differences by period in the proportion of requests with issued warrants (92.8% vs 90.6% p = 0.42) nor the proportion of requests with subsequent hospital admissions (54.0% vs 49.0% p = 0.33). In the second period, there were differences in the proportion of patients with a previously diagnosed mental health disorder (95.3% vs 90.4% p = 0.049). There were significant differences in the distribution of the origin of requests. Being unemployed (OR:2.5 CI:1.2 - 5.2), not having completed high school (OR:2.01 CI:1.12 - 3.77) and having university education (OR:3.67 CI:1.27 - 10.57) degree were associated with hospital admission. CONCLUSION: Severe mental illness with criteria for mental health warrants may require more resources and different approaches due to a possible increase during and after the COVID-19 pandemic. Community based mental healthcare, incentivized follow-up by primary care and ambulatory treatment may be considered. Further research should evaluate both the national and international trends and associated factors.

A Systematic Review on Deep Learning with CNNs Applied to Surface Defect Detection.

Surface defect detection with machine learning has become an important tool in industries and a large field of study for researchers or workers in recent years. It is necessary to have a simplified source of information that helps us to better focus on one type of surface. In this systematic review, we present a classification for surface defect detection based on convolutional neural networks (CNNs) focused on surface types. Findings: Out of 253 records identified, 59 primary studies were eligible. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, we analyzed the structures of each study and the concepts related to defects and their types on surfaces. The presented review is mainly focused on finding a classification for the types of surfaces most used in industry (metal, building, ceramic, wood, and special). We delve into the specifics of each surface category, offering illustrative examples of their applications within both industrial and laboratory settings. Furthermore, we propose a new taxonomy of machine learning based on the obtained results and collected information. We summarized the studies and extracted the main characteristics such as type of surface, problem types, timeline, type of network, techniques, and datasets. Among the most relevant results of our analysis, we found that the metallic surface is the most used, as it is the one found in 62.71% of the studies, and the most prevalent problem type is classification, accounting for 49.15% of the total. Furthermore, we observe that transfer learning was employed in 83.05% of the studies, while data augmentation was utilized in 59.32%. Our findings also provide insights into the cameras most frequently employed, along with the strategies adopted to address illumination challenges present in certain articles and the approach to creating datasets for real-world applications. The main results presented in this review allow for a quick and efficient search of information for researchers and professionals interested in improving the results of their defect detection projects. Finally, we analyzed the trends that could open new fields of study for future research in the area of surface defect detection.

Vision-Based Flying Obstacle Detection for Avoiding Midair Collisions: A Systematic Review.

This paper presents a systematic review of articles on computer-vision-based flying obstacle detection with a focus on midair collision avoidance. Publications from the beginning until 2022 were searched in Scopus, IEEE, ACM, MDPI, and Web of Science databases. From the initial 647 publications obtained, 85 were finally selected and examined. The results show an increasing interest in this topic, especially in relation to object detection and tracking. Our study hypothesizes that the widespread access to commercial drones, the improvements in single-board computers, and their compatibility with computer vision libraries have contributed to the increase in the number of publications. The review also shows that the proposed algorithms are mainly tested using simulation software and flight simulators, and only 26 papers report testing with physical flying vehicles. This systematic review highlights other gaps to be addressed in future work. Several identified challenges are related to increasing the success rate of threat detection and testing solutions in complex scenarios.

Perceived Quality of Life and Life Satisfaction: Does the Role of Gender, Age, Skills, and Psychological Factors Remain Relevant after the COVID-19 Pandemic?

BACKGROUND: After two years of psychological, physical, social, economic, environmental, and societal challenges, this paper examines the psychological health and well-being of Portuguese students based on their socioemotional skills (SSES), positive youth development (PYD), depression, anxiety, and stress (DASS), as well as the relationship between these variables and their influence on perceived quality of life and life satisfaction. METHODS: This study examined 3235 students from lower to upper secondary, half of whom were female (M = 14.46 ± 1.883 years old). Using SPSS software, descriptive statistics were determined for all variables; mean differences between age and gender were found using ANOVA and the post hoc Scheffe test. Linear regressions with the Enter method were used to study how to predict perceived quality of life and satisfaction with life. RESULTS: Males had scores indicating more SSES|optimism, emotional control, resilience, confidence, sociability, creativity, energy, a sense of belonging to school, and PYD. Girls had better skills for cooperating and relating to teachers but more test anxiety and DASS. Younger adolescents had better psychological health, greater skills, and a better perception of quality of life and life satisfaction when compared to older adolescents. Age, gender, SSES, PYD, and the DASS variables can explain 69% of the variance in life satisfaction, while these variables can explain 60.5% of the variance in perceived quality of life. CONCLUSIONS: These results point to the relevance of SSES for psychological health and well-being, suggesting that interventions should focus on promoting these variables, paying special attention to female gender and age-related challenges.

Public-private partnership in the Portuguese health sector.

Since 2001, in Portugal, constant reforms in hospital management have accompanied the transformations in the management models applied to public administration, intending to ensure a higher quality of services and, simultaneously, a more significant economic efficiency. This study aims to analyse, for the period between 2012 and 2021, the economic and financial results (value-for-money) of the PPP model, compared with the public management hospitals (PMH). It used a mixed research approach based on multiple case studies and archival research. As the main results, it was found that: i) the PPP model, applied to the health sector, appears to be advantageous, not only regarding the economic and financial results but also concerning the quality of service provision; and ii) despite the value-for-money generated by the PPP model, the lower operating costs and the superior performance in comparison with PMH, the government has permanently opted to revert from a PPP model to a PMH model. This study concluded that the hospital management model is instead seen as an instrumentalised political instrument than a management tool that could generate savings for the taxpayers. Several practical implications are presented.

Public-Private Partnerships Model Applied to Hospitals-A Critical Review.

In this paper, a conceptual framework for investigating the PPP model as it relates to hospitals is proposed. When the PPP model is applied to healthcare (hospitals), it is possible to discover the path to success by developing a critical assessment and deriving a clear model. It is concluded that most PPP model implementations in hospitals around the world have produced favorable outcomes, both in terms of the performance of healthcare units and in terms of cost-effectiveness. Additionally, a path-to-success model that applies to hospitals is offered, taking into account six PPP model dimensions: (i) Environment; (ii) Potentiate Benefits; (iii) Constant Measure; (iv) Evaluation; (v) Management; and (vi) Enhance Strengths. The PPP model only applies case by case and under specific requirements that should be met cumulatively to provide additional value to healthcare's quality of service. The right conditions are created, the right benefits are amplified, public concerns are frequently assessed, private contributions are carefully considered, and all pressing challenges are managed by enhancing both public and private strengths. Leading decision- and action-making processes in corporate, governmental, and social sectors is the goal of managing PPP models.

Value of handcrafted and deep radiomic features towards training robust machine learning classifiers for prediction of prostate cancer disease aggressiveness.

There is a growing piece of evidence that artificial intelligence may be helpful in the entire prostate cancer disease continuum. However, building machine learning algorithms robust to inter- and intra-radiologist segmentation variability is still a challenge. With this goal in mind, several model training approaches were compared: removing unstable features according to the intraclass correlation coefficient (ICC); training independently with features extracted from each radiologist's mask; training with the feature average between both radiologists; extracting radiomic features from the intersection or union of masks; and creating a heterogeneous dataset by randomly selecting one of the radiologists' masks for each patient. The classifier trained with this last resampled dataset presented with the lowest generalization error, suggesting that training with heterogeneous data leads to the development of the most robust classifiers. On the contrary, removing features with low ICC resulted in the highest generalization error. The selected radiomics dataset, with the randomly chosen radiologists, was concatenated with deep features extracted from neural networks trained to segment the whole prostate. This new hybrid dataset was then used to train a classifier. The results revealed that, even though the hybrid classifier was less overfitted than the one trained with deep features, it still was unable to outperform the radiomics model.

COVID-19, Wellness and Life Satisfaction in Adolescence: Individual and Contextual Issues.

During and in the aftermath of the COVID-19 pandemic, several works reflected on young people's physical and psychological health. The Dual Factor Model, which we refer to as the quadripartite model, is useful for understanding children's and adolescents' psychological health and differentiating them regarding their attitude toward the effects of the COVID-19 pandemic. In this investigation, students from the fifth to twelfth year of schooling enrolled in the DGEEC study "Psychological Health and Wellbeing in Portuguese schools" were considered. Four groups were created based on life satisfaction (low or high) and psychological distress (with or without symptoms). The study included 4444 students (M = 13.39 years ± 2.41), of whom 47.8% were male. Of the participants, 27.2% were in the second cycle of primary education, and 72.8% were in lower and upper secondary education. Differences in gender and education level (as a proxy for age) were observed. Additionally, when considering students' perceptions of changes in their lives following the COVID-19 pandemic (stayed the same, became worse, became better), these three groups were compared concerning personal and contextual variables, revealing significant differences at both the individual and contextual levels. Finally, the study discusses the influence of education and health professionals and the need for friendly public policies.

Predicting hospital emergency department visits accurately: A systematic review.

OBJECTIVES: The emergency department (ED) is a very important healthcare entrance point, known for its challenging organisation and management due to demand unpredictability. An accurate forecast system of ED visits is crucial to the implementation of better management strategies that optimise resources utilization, reduce costs and improve public confidence. The aim of this review is to investigate the different factors that affect the ED visits forecasting outcomes, in particular the predictive variables and type of models applied. METHODS: A systematic search was conducted in PubMed, Web of Science and Scopus. The review methodology followed the PRISMA statement guidelines. RESULTS: Seven studies were selected, all exploring predictive models to forecast ED daily visits for general care. MAPE and RMAE were used to measure models' accuracy. All models displayed good accuracy, with errors below 10%. CONCLUSIONS: Model selection and accuracy was found to be particularly sensitive to the ED dimension. While ARIMA-based and other linear models have good performance for short-time forecast, some machine learning methods proved to be more stable when forecasting multiple horizons. The inclusion of exogenous variables was found to be advantageous only in bigger EDs.

A Comparative Study of Automated Deep Learning Segmentation Models for Prostate MRI.

Prostate cancer is one of the most common forms of cancer globally, affecting roughly one in every eight men according to the American Cancer Society. Although the survival rate for prostate cancer is significantly high given the very high incidence rate, there is an urgent need to improve and develop new clinical aid systems to help detect and treat prostate cancer in a timely manner. In this retrospective study, our contributions are twofold: First, we perform a comparative unified study of different commonly used segmentation models for prostate gland and zone (peripheral and transition) segmentation. Second, we present and evaluate an additional research question regarding the effectiveness of using an object detector as a pre-processing step to aid in the segmentation process. We perform a thorough evaluation of the deep learning models on two public datasets, where one is used for cross-validation and the other as an external test set. Overall, the results reveal that the choice of model is relatively inconsequential, as the majority produce non-significantly different scores, apart from nnU-Net which consistently outperforms others, and that the models trained on data cropped by the object detector often generalize better, despite performing worse during cross-validation.

Forage based diet as an alternative to a high concentrate diet for finishing young bulls - Effects on growth performance, greenhouse gas emissions and meat quality.

Two groups of 8 individually housed young crossbred-bulls, in the finishing period, were used to test the effect of a Total Mixed Ration diet with high forage content (54% DM), low starch content (14% DM), supplemented with sunflower seeds (10% DM) (HFS) on growth performance, carcass and meat quality, fatty acid profile and carbon footprint, with reference to a conventional concentrate-based (90% DM) (Control) diet. The experiment lasted 64 days before slaughter. During the experiment, feed intake was monitored daily and live weight every 14 days. Individual CH4 emissions were assessed at 16-days intervals, using a GreenFeed for Large Animal unit. Feed intake and feed conversion ratio were higher for HFS diet, but average daily weight gain and feeding costs were similar for the two diets. Dressing percentage was reduced with HFS diet. The HFS increased redness, yellowness and Chroma of subcutaneous fat, but did not compromise commercial value of the carcasses. Meat colour, shear force, or sensory parameters were not affected by diet. The HFS diet allowed a healthier FA profile, due to the higher proportions of 18:3n-3, t11-18:1 and c9,t11-18:2 and the lower proportion of t10-18:1. The HFS diet did not reduce the carbon footprint in the finishing period of young bulls, due to increased digestive CH4 emissions. The results of this experiment showed that the HFS diet can be an alternative to the conventional diets used in finishing young-bulls. Although it may result in a slight reduction in animal performance, it has a strong impact on reducing dependence on inputs from outside the farm.

Geographical discrimination of olive oils from Cv. 'Galega Vulgar'.

Olive oils from seven Portuguese regions were selected to study the effect of the geographical origin on the oils' composition. Quality parameters, fatty acids, tocopherols, hydroxytyrosol and tyrosol derivatives, and oxidative stability were evaluated. All olive oils could be classified as extra virgin, and the geographical origin significantly affected the oils chemical composition. Principal component analysis further confirmed the significant impact of the geographical origin on the composition and, indirectly, on stability of the oils, showing that the evaluated parameters could be used as markers for geographical origin identification. Alternatively, Fourier transform infrared spectroscopy was applied, allowing to establish a linear discriminant model that correctly identified the geographical origin of the olive oils with a mean sensitivity of 99 ± 3 % (internal validation), confirming the impact of the oil origin on its characteristics. This finding allowed foreseeing the future application of the spectroscopy approach as a green, fast and non-invasive authentication tool.

A Real-Time Automated Defect Detection System for Ceramic Pieces Manufacturing Process Based on Computer Vision with Deep Learning.

Defect detection is a key element of quality control in today's industries, and the process requires the incorporation of automated methods, including image sensors, to detect any potential defects that may occur during the manufacturing process. While there are various methods that can be used for inspecting surfaces, such as those of metal and building materials, there are only a limited number of techniques that are specifically designed to analyze specialized surfaces, such as ceramics, which can potentially reveal distinctive anomalies or characteristics that require a more precise and focused approach. This article describes a study and proposes an extended solution for defect detection on ceramic pieces within an industrial environment, utilizing a computer vision system with deep learning models. The solution includes an image acquisition process and a labeling platform to create training datasets, as well as an image preprocessing technique, to feed a machine learning algorithm based on convolutional neural networks (CNNs) capable of running in real time within a manufacturing environment. The developed solution was implemented and evaluated at a leading Portuguese company that specializes in the manufacturing of tableware and fine stoneware. The collaboration between the research team and the company resulted in the development of an automated and effective system for detecting defects in ceramic pieces, achieving an accuracy of 98.00% and an F1-Score of 97.29%.

An Electronic Nose as a Non-Destructive Analytical Tool to Identify the Geographical Origin of Portuguese Olive Oils from Two Adjacent Regions.

The geographical traceability of extra virgin olive oils (EVOO) is of paramount importance for oil chain actors and consumers. Oils produced in two adjacent Portuguese regions, Côa (36 oils) and Douro (31 oils), were evaluated and fulfilled the European legal thresholds for EVOO categorization. Compared to the Douro region, oils from Côa had higher total phenol contents (505 versus 279 mg GAE/kg) and greater oxidative stabilities (17.5 versus 10.6 h). The majority of Côa oils were fruity-green, bitter, and pungent oils. Conversely, Douro oils exhibited a more intense fruity-ripe and sweet sensation. Accordingly, different volatiles were detected, belonging to eight chemical families, from which aldehydes were the most abundant. Additionally, all oils were evaluated using a lab-made electronic nose, with metal oxide semiconductor sensors. The electrical fingerprints, together with principal component analysis, enabled the unsupervised recognition of the oils' geographical origin, and their successful supervised linear discrimination (sensitivity of 98.5% and specificity of 98.4%; internal validation). The E-nose also quantified the contents of the two main volatile chemical classes (alcohols and aldehydes) and of the total volatiles content, for the studied olive oils split by geographical origin, using multivariate linear regression models (0.981 ≤ R2 ≤ 0.998 and 0.40 ≤ RMSE ≤ 2.79 mg/kg oil; internal validation). The E-nose-MOS was shown to be a fast, green, non-invasive and cost-effective tool for authenticating the geographical origin of the studied olive oils and to estimate the contents of the most abundant chemical classes of volatiles.