Benha University

Artificial intelligence (AI) is currently being utilized in a wide range of sophisticated applications, but the outcomes of many AI models are challenging to comprehend and trust due to their black-box nature. Usually, it is essential to understand the reasoning behind an AI model’s decision-making. Thus, the need for eXplainable AI (XAI) methods for improving trust in AI models has arisen. XAI has become a popular research subject within the AI field in recent years. Existing survey papers have tackled the concepts of XAI, its general terms, and post-hoc explainability methods but there have not been any reviews that have looked at the assessment methods, available tools, XAI datasets, and other related aspects. Therefore, in this comprehensive study, we provide readers with an overview of the current research and trends in this rapidly emerging area with a case study example. The study starts by explaining the background of XAI, common definitions, and summarizing recently proposed techniques in XAI for supervised machine learning. The review divides XAI techniques into four axes using a hierarchical categorization system: (i) data explainability, (ii) model explainability, (iii) post-hoc explainability, and (iv) assessment of explanations. We also introduce available evaluation metrics as well as open-source packages and datasets with future research directions. Then, the significance of explainability in terms of legal demands, user viewpoints, and application orientation is outlined, termed as XAI concerns. This paper advocates for tailoring explanation content to specific user types. An examination of XAI techniques and evaluation was conducted by looking at 410 critical articles, published between January 2016 and October 2022, in reputed journals and using a wide range of research databases as a source of information. The article is aimed at XAI researchers who are interested in making their AI models more trustworthy, as well as towards researchers from other disciplines who are looking for effective XAI methods to complete tasks with confidence while communicating meaning from data.

Automatic license plate recognition (ALPR) is the extraction of vehicle license plate information from an image or a sequence of images. The extracted information can be used with or without a database in many applications, such as electronic payment systems (toll payment, parking fee payment), and freeway and arterial monitoring systems for traffic surveillance. The ALPR uses either a color, black and white, or infrared camera to take images. The quality of the acquired images is a major factor in the success of the ALPR. ALPR as a real-life application has to quickly and successfully process license plates under different environmental conditions, such as indoors, outdoors, day or night time. It should also be generalized to process license plates from different nations, provinces, or states. These plates usually contain different colors, are written in different languages, and use different fonts; some plates may have a single color background and others have background images. The license plates can be partially occluded by dirt, lighting, and towing accessories on the car. In this paper, we present a comprehensive review of the state-of-the-art techniques for ALPR. We categorize different ALPR techniques according to the features they used for each stage, and compare them in terms of pros, cons, recognition accuracy, and processing speed. Future forecasts of ALPR are given at the end.

Polycystic ovary syndrome (PCOS) is the most common endocrine disorder in women. The clinical manifestation of PCOS varies from a mild menstrual disorder to severe disturbance of reproductive and metabolic functions. Management of women with PCOS depends on the symptoms. These could be ovulatory dysfunction-related infertility, menstrual disorders, or androgen-related symptoms. Weight loss improves the endocrine profile and increases the likelihood of ovulation and pregnancy. Normalization of menstrual cycles and ovulation could occur with modest weight loss as little as 5% of the initial weight. The treatment of obesity includes modifications in lifestyle (diet and exercise) and medical and surgical treatment. In PCOS, anovulation relates to low follicle-stimulating hormone concentrations and the arrest of antral follicle growth in the final stages of maturation. This can be treated with medications such as clomiphene citrate, tamoxifen, aromatase inhibitors, metformin, glucocorticoids, or gonadotropins or surgically by laparoscopic ovarian drilling. In vitro fertilization will remain the last option to achieve pregnancy when others fail. Chronic anovulation over a long period of time is also associated with an increased risk of endometrial hyperplasia and carcinoma, which should be seriously investigated and treated. There are androgenic symptoms that will vary from patient to patient, such as hirsutism, acne, and/or alopecia. These are troublesome presentations to the patients and require adequate treatment. Alternative medicine has been emerging as one of the commonly practiced medicines for different health problems, including PCOS. This review underlines the contribution to the treatment of different symptoms.

A number of vertical mining algorithms have been proposed recently for association mining, which have shown to be very effective and usually outperform horizontal approaches. The main advantage of the vertical format is support for fast frequency counting via intersection operations on transaction ids (tids) and automatic pruning of irrelevant data. The main problem with these approaches is when intermediate results of vertical tid lists become too large for memory, thus affecting the algorithm scalability. In this paper we present a novel vertical data representation called Diffset, that only keeps track of differences in the tids of a candidate pattern from its generating frequent patterns. We show that diffsets drastically cut down the size of memory required to store intermediate results. We show how diffsets, when incorporated into previous vertical mining methods, increase the performance significantly. We also present a new algorithm, using diffsets, for mining maximal patterns. Experimental comparisons, on both dense and sparse databases, show that diffsets deliver order of magnitude performance improvements over the best previous methods. 1

AIMS: Rheumatic heart disease (RHD) accounts for over a million premature deaths annually; however, there is little contemporary information on presentation, complications, and treatment. METHODS AND RESULTS: This prospective registry enrolled 3343 patients (median age 28 years, 66.2% female) presenting with RHD at 25 hospitals in 12 African countries, India, and Yemen between January 2010 and November 2012. The majority (63.9%) had moderate-to-severe multivalvular disease complicated by congestive heart failure (33.4%), pulmonary hypertension (28.8%), atrial fibrillation (AF) (21.8%), stroke (7.1%), infective endocarditis (4%), and major bleeding (2.7%). One-quarter of adults and 5.3% of children had decreased left ventricular (LV) systolic function; 23% of adults and 14.1% of children had dilated LVs. Fifty-five percent (n = 1761) of patients were on secondary antibiotic prophylaxis. Oral anti-coagulants were prescribed in 69.5% (n = 946) of patients with mechanical valves (n = 501), AF (n = 397), and high-risk mitral stenosis in sinus rhythm (n = 48). However, only 28.3% (n = 269) had a therapeutic international normalized ratio. Among 1825 women of childbearing age (12-51 years), only 3.6% (n = 65) were on contraception. The utilization of valvuloplasty and valve surgery was higher in upper-middle compared with lower-income countries. CONCLUSION: Rheumatic heart disease patients were young, predominantly female, and had high prevalence of major cardiovascular complications. There is suboptimal utilization of secondary antibiotic prophylaxis, oral anti-coagulation, and contraception, and variations in the use of percutaneous and surgical interventions by country income level.

Research within the anatomical sciences often relies on human cadaveric tissues. Without the good will of these donors who allow us to use their bodies to push forward our anatomical knowledge, most human anatomical research would come to a standstill. However, many research papers omit an acknowledgement to the donor cadavers or, as no current standardized versions exist, use language that is extremely varied. To remedy this problem, 20 editors-in-chiefs from 17 anatomical journals joined together to put together official recommendations that can be used by authors when acknowledging the donor cadavers used in their studies. The goal of these recommendations is to standardize the writing approach by which donors are acknowledged in anatomical studies that use human cadaveric tissues. Such sections in anatomical papers will not only rightfully thank those who made the donation but might also encourage, motivate, and inspire future individuals to make such gifts for the betterment of the anatomical sciences and patient care.

The coronavirus (COVID-19) pandemic is putting healthcare systems across the world under unprecedented and increasing pressure according to the World Health Organization (WHO). With the advances in computer algorithms and especially Artificial Intelligence, the detection of this type of virus in the early stages will help in fast recovery and help in releasing the pressure off healthcare systems. In this paper, a GAN with deep transfer learning for coronavirus detection in chest X-ray images is presented. The lack of datasets for COVID-19 especially in chest X-rays images is the main motivation of this scientific study. The main idea is to collect all the possible images for COVID-19 that exists until the writing of this research and use the GAN network to generate more images to help in the detection of this virus from the available X-rays images with the highest accuracy possible. The dataset used in this research was collected from different sources and it is available for researchers to download and use it. The number of images in the collected dataset is 307 images for four different types of classes. The classes are the COVID-19, normal, pneumonia bacterial, and pneumonia virus. Three deep transfer models are selected in this research for investigation. The models are the Alexnet, Googlenet, and Restnet18. Those models are selected for investigation through this research as it contains a small number of layers on their architectures, this will result in reducing the complexity, the consumed memory and the execution time for the proposed model. Three case scenarios are tested through the paper, the first scenario includes four classes from the dataset, while the second scenario includes 3 classes and the third scenario includes two classes. All the scenarios include the COVID-19 class as it is the main target of this research to be detected. In the first scenario, the Googlenet is selected to be the main deep transfer model as it achieves 80.6% in testing accuracy. In the second scenario, the Alexnet is selected to be the main deep transfer model as it achieves 85.2% in testing accuracy, while in the third scenario which includes two classes (COVID-19, and normal), Googlenet is selected to be the main deep transfer model as it achieves 100% in testing accuracy and 99.9% in the validation accuracy. All the performance measurement strengthens the obtained results through the research.

Recently, biomedicine and tissue regeneration have emerged as great advances that impacted the spectrum of healthcare. This left the door open for further improvement of their applications to revitalize the impaired tissues. Hence, restoring their functions. The implementation of therapeutic protocols that merge biomimetic scaffolds, bioactive molecules, and cells plays a pivotal role in this track. Smart/stimuli-responsive hydrogels are remarkable three-dimensional (3D) bioscaffolds intended for tissue engineering and other biomedical purposes. They can simulate the physicochemical, mechanical, and biological characters of the innate tissues. Also, they provide the aqueous conditions for cell growth, support 3D conformation, provide mechanical stability for the cells, and serve as potent delivery matrices for bioactive molecules. Many natural and artificial polymers were broadly utilized to design these intelligent platforms with novel advanced characteristics and tailored functionalities that fit such applications. In the present review, we highlighted the different types of smart/stimuli-responsive hydrogels with emphasis on their synthesis scheme. Besides, the mechanisms of their responsiveness to different stimuli were elaborated. Their potential for tissue engineering applications was discussed. Furthermore, their exploitation in other biomedical applications as targeted drug delivery, smart biosensors, actuators, 3D and 4D printing, and 3D cell culture were outlined. In addition, we threw light on smart self-healing hydrogels and their applications in biomedicine. Eventually, we presented their future perceptions in biomedical and tissue regeneration applications. Conclusively, current progress in the design of smart/stimuli-responsive hydrogels enhances their prospective to function as intelligent, and sophisticated systems in different biomedical applications.

Plants play a crucial role in supplying food globally. Various environmental factors lead to plant diseases which results in significant production losses. However, manual detection of plant diseases is a time-consuming and error-prone process. It can be an unreliable method of identifying and preventing the spread of plant diseases. Adopting advanced technologies such as Machine Learning (ML) and Deep Learning (DL) can help to overcome these challenges by enabling early identification of plant diseases. In this paper, the recent advancements in the use of ML and DL techniques for the identification of plant diseases are explored. The research focuses on publications between 2015 and 2022, and the experiments discussed in this study demonstrate the effectiveness of using these techniques in improving the accuracy and efficiency of plant disease detection. This study also addresses the challenges and limitations associated with using ML and DL for plant disease identification, such as issues with data availability, imaging quality, and the differentiation between healthy and diseased plants. The research provides valuable insights for plant disease detection researchers, practitioners, and industry professionals by offering solutions to these challenges and limitations, providing a comprehensive understanding of the current state of research in this field, highlighting the benefits and limitations of these methods, and proposing potential solutions to overcome the challenges of their implementation.

BACKGROUND: For more than three decades, the Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) has provided a framework to quantify health loss due to diseases, injuries, and associated risk factors. This paper presents GBD 2023 findings on disease and injury burden and risk-attributable health loss, offering a global audit of the state of world health to inform public health priorities. This work captures the evolving landscape of health metrics across age groups, sexes, and locations, while reflecting on the remaining post-COVID-19 challenges to achieving our collective global health ambitions. METHODS: The GBD 2023 combined analysis estimated years lived with disability (YLDs), years of life lost (YLLs), and disability-adjusted life-years (DALYs) for 375 diseases and injuries, and risk-attributable burden associated with 88 modifiable risk factors. Of the more than 310 000 total data sources used for all GBD 2023 (about 30% of which were new to this estimation round), more than 120 000 sources were used for estimation of disease and injury burden and 59 000 for risk factor estimation, and included vital registration systems, surveys, disease registries, and published scientific literature. Data were analysed using previously established modelling approaches, such as disease modelling meta-regression version 2.1 (DisMod-MR 2.1) and comparative risk assessment methods. Diseases and injuries were categorised into four levels on the basis of the established GBD cause hierarchy, as were risk factors using the GBD risk hierarchy. Estimates stratified by age, sex, location, and year from 1990 to 2023 were focused on disease-specific time trends over the 2010-23 period and presented as counts (to three significant figures) and age-standardised rates per 100 000 person-years (to one decimal place). For each measure, 95% uncertainty intervals [UIs] were calculated with the 2·5th and 97·5th percentile ordered values from a 250-draw distribution. FINDINGS: Total numbers of global DALYs grew 6·1% (95% UI 4·0-8·1), from 2·64 billion (2·46-2·86) in 2010 to 2·80 billion (2·57-3·08) in 2023, but age-standardised DALY rates, which account for population growth and ageing, decreased by 12·6% (11·0-14·1), revealing large long-term health improvements. Non-communicable diseases (NCDs) contributed 1·45 billion (1·31-1·61) global DALYs in 2010, increasing to 1·80 billion (1·63-2·03) in 2023, alongside a concurrent 4·1% (1·9-6·3) reduction in age-standardised rates. Based on DALY counts, the leading level 3 NCDs in 2023 were ischaemic heart disease (193 million [176-209] DALYs), stroke (157 million [141-172]), and diabetes (90·2 million [75·2-107]), with the largest increases in age-standardised rates since 2010 occurring for anxiety disorders (62·8% [34·0-107·5]), depressive disorders (26·3% [11·6-42·9]), and diabetes (14·9% [7·5-25·6]). Remarkable health gains were made for communicable, maternal, neonatal, and nutritional (CMNN) diseases, with DALYs falling from 874 million (837-917) in 2010 to 681 million (642-736) in 2023, and a 25·8% (22·6-28·7) reduction in age-standardised DALY rates. During the COVID-19 pandemic, DALYs due to CMNN diseases rose but returned to pre-pandemic levels by 2023. From 2010 to 2023, decreases in age-standardised rates for CMNN diseases were led by rate decreases of 49·1% (32·7-61·0) for diarrhoeal diseases, 42·9% (38·0-48·0) for HIV/AIDS, and 42·2% (23·6-56·6) for tuberculosis. Neonatal disorders and lower respiratory infections remained the leading level 3 CMNN causes globally in 2023, although both showed notable rate decreases from 2010, declining by 16·5% (10·6-22·0) and 24·8% (7·4-36·7), respectively. Injury-related age-standardised DALY rates decreased by 15·6% (10·7-19·8) over the same period. Differences in burden due to NCDs, CMNN diseases, and injuries persisted across age, sex, time, and location. Based on our risk analysis, nearly 50% (1·27 billion [1·18-1·38]) of the roughly 2·80 billion total global DALYs in 2023 were attributable to the 88 risk factors analysed in GBD. Globally, the five level 3 risk factors contributing the highest proportion of risk-attributable DALYs were high systolic blood pressure (SBP), particulate matter pollution, high fasting plasma glucose (FPG), smoking, and low birthweight and short gestation-with high SBP accounting for 8·4% (6·9-10·0) of total DALYs. Of the three overarching level 1 GBD risk factor categories-behavioural, metabolic, and environmental and occupational-risk-attributable DALYs rose between 2010 and 2023 only for metabolic risks, increasing by 30·7% (24·8-37·3); however, age-standardised DALY rates attributable to metabolic risks decreased by 6·7% (2·0-11·0) over the same period. For all but three of the 25 leading level 3 risk factors, age-standardised rates dropped between 2010 and 2023-eg, declining by 54·4% (38·7-65·3) for unsafe sanitation, 50·5% (33·3-63·1) for unsafe water source, and 45·2% (25·6-72·0) for no access to handwashing facility, and by 44·9% (37·3-53·5) for child growth failure. The three leading level 3 risk factors for which age-standardised attributable DALY rates rose were high BMI (10·5% [0·1 to 20·9]), drug use (8·4% [2·6 to 15·3]), and high FPG (6·2% [-2·7 to 15·6]; non-significant). INTERPRETATION: Our findings underscore the complex and dynamic nature of global health challenges. Since 2010, there have been large decreases in burden due to CMNN diseases and many environmental and behavioural risk factors, juxtaposed with sizeable increases in DALYs attributable to metabolic risk factors and NCDs in growing and ageing populations. This long-observed consequence of the global epidemiological transition was only temporarily interrupted by the COVID-19 pandemic. The substantially decreasing CMNN disease burden, despite the 2008 global financial crisis and pandemic-related disruptions, is one of the greatest collective public health successes known. However, these achievements are at risk of being reversed due to major cuts to development assistance for health globally, the effects of which will hit low-income countries with high burden the hardest. Without sustained investment in evidence-based interventions and policies, progress could stall or reverse, leading to widespread human costs and geopolitical instability. Moreover, the rising NCD burden necessitates intensified efforts to mitigate exposure to leading risk factors-eg, air pollution, smoking, and metabolic risks, such as high SBP, BMI, and FPG-including policies that promote food security, healthier diets, physical activity, and equitable and expanded access to potential treatments, such as GLP-1 receptor agonists. Decisive, coordinated action is needed to address long-standing yet growing health challenges, including depressive and anxiety disorders. Yet this can be only part of the solution. Our response to the NCD syndemic-the complex interaction of multiple health risks, social determinants, and systemic challenges-will define the future landscape of global health. To ensure human wellbeing, economic stability, and social equity, global action to sustain and advance health gains must prioritise reducing disparities by addressing socioeconomic and demographic determinants, ensuring equitable health-care access, tackling malnutrition, strengthening health systems, and improving vaccination coverage. We live in times of great opportunity. FUNDING: Gates Foundation and Bloomberg Philanthropies.

An increasing number of genetic and metabolic anomalies have been determined to lead to cancer, generally fatal. Cancerous cells may spread to any body part, where they can be life-threatening. Skin cancer is one of the most common types of cancer, and its frequency is increasing worldwide. The main subtypes of skin cancer are squamous and basal cell carcinomas, and melanoma, which is clinically aggressive and responsible for most deaths. Therefore, skin cancer screening is necessary. One of the best methods to accurately and swiftly identify skin cancer is using deep learning (DL). In this research, the deep learning method convolution neural network (CNN) was used to detect the two primary types of tumors, malignant and benign, using the ISIC2018 dataset. This dataset comprises 3533 skin lesions, including benign, malignant, nonmelanocytic, and melanocytic tumors. Using ESRGAN, the photos were first retouched and improved. The photos were augmented, normalized, and resized during the preprocessing step. Skin lesion photos could be classified using a CNN method based on an aggregate of results obtained after many repetitions. Then, multiple transfer learning models, such as Resnet50, InceptionV3, and Inception Resnet, were used for fine-tuning. In addition to experimenting with several models (the designed CNN, Resnet50, InceptionV3, and Inception Resnet), this study's innovation and contribution are the use of ESRGAN as a preprocessing step. Our designed model showed results comparable to the pretrained model. Simulations using the ISIC 2018 skin lesion dataset showed that the suggested strategy was successful. An 83.2% accuracy rate was achieved by the CNN, in comparison to the Resnet50 (83.7%), InceptionV3 (85.8%), and Inception Resnet (84%) models.

Alzheimer's disease (AD) is the most common type of dementia. Its diagnosis and progression detection have been intensively studied. Nevertheless, research studies often have little effect on clinical practice mainly due to the following reasons: (1) Most studies depend mainly on a single modality, especially neuroimaging; (2) diagnosis and progression detection are usually studied separately as two independent problems; and (3) current studies concentrate mainly on optimizing the performance of complex machine learning models, while disregarding their explainability. As a result, physicians struggle to interpret these models, and feel it is hard to trust them. In this paper, we carefully develop an accurate and interpretable AD diagnosis and progression detection model. This model provides physicians with accurate decisions along with a set of explanations for every decision. Specifically, the model integrates 11 modalities of 1048 subjects from the Alzheimer's Disease Neuroimaging Initiative (ADNI) real-world dataset: 294 cognitively normal, 254 stable mild cognitive impairment (MCI), 232 progressive MCI, and 268 AD. It is actually a two-layer model with random forest (RF) as classifier algorithm. In the first layer, the model carries out a multi-class classification for the early diagnosis of AD patients. In the second layer, the model applies binary classification to detect possible MCI-to-AD progression within three years from a baseline diagnosis. The performance of the model is optimized with key markers selected from a large set of biological and clinical measures. Regarding explainability, we provide, for each layer, global and instance-based explanations of the RF classifier by using the SHapley Additive exPlanations (SHAP) feature attribution framework. In addition, we implement 22 explainers based on decision trees and fuzzy rule-based systems to provide complementary justifications for every RF decision in each layer. Furthermore, these explanations are represented in natural language form to help physicians understand the predictions. The designed model achieves a cross-validation accuracy of 93.95% and an F1-score of 93.94% in the first layer, while it achieves a cross-validation accuracy of 87.08% and an F1-Score of 87.09% in the second layer. The resulting system is not only accurate, but also trustworthy, accountable, and medically applicable, thanks to the provided explanations which are broadly consistent with each other and with the AD medical literature. The proposed system can help to enhance the clinical understanding of AD diagnosis and progression processes by providing detailed insights into the effect of different modalities on the disease risk.

The poultry industry contributes significantly to bridging the nutritional gap in many countries because of its meat and eggs products rich in protein and valuable nutrients at a cost less than other animal meat sources. The natural antibiotics alternatives including probiotics, prebiotics, symbiotics, organic acids, essential oils, enzymes, immunostimulants, and phytogenic (phytobiotic) including herbs, botanicals, essential oils, and oleoresins are the most common feed additives that acquire popularity in poultry industry following the ban of antibiotic growth promoters (AGPs). They are commonly used worldwide because of their unique properties and positive impact on poultry production. They can be easily mixed with other feed ingredients, have no tissue residues, improve feed intake, feed gain, feed conversion rate, improve bird immunity, improve digestion, increase nutrients availability as well as absorbability, have antimicrobial effects, do not affect carcass characters, decrease the usage of antibiotics, acts as antioxidants, anti-inflammatory, compete for stress factors and provide healthy organic products for human consumption. Therefore, the current review focuses on a comprehensive description of different natural antibiotic growth promoters' alternatives, the mode of their action, and their impacts on poultry production.

This study scrutinizes the reliability and validity of existing analyses that focus on the impact of various environmental factors on a photovoltaic (PV) system’s performance. For the first time, four environmental factors (the accumulation of dust, water droplets, birds’ droppings, and partial shading conditions) affecting system performance are investigated, simultaneously, in one study. The results obtained from this investigation demonstrate that the accumulation of dust, shading, and bird fouling has a significant effect on PV current and voltage, and consequently, the harvested PV energy. ‘Shading’ had the strongest influence on the efficiency of the PV modules. It was found that increasing the area of shading on a PV module surface by a quarter, half, and three quarters resulted in a power reduction of 33.7%, 45.1%, and 92.6%, respectively. However, results pertaining to the impact of water droplets on the PV panel had an inverse effect, decreasing the temperature of the PV panel, which led to an increase in the potential difference and improved the power output by at least 5.6%. Moreover, dust accumulation reduced the power output by 8.80% and the efficiency by 11.86%, while birds fouling the PV module surface was found to reduce the PV system performance by about 7.4%.

While long non-coding RNA (lncRNA) genes have attracted a lot of attention in the last decade, the focus regarding their mechanisms of action has been primarily on the RNA product of these genes. Recent work on several lncRNAs genes demonstrates that not only is the produced RNA species important, but also that transcription of the lncRNA locus alone can have regulatory functions. Like the functions of lncRNA transcripts, the mechanisms that underlie these genome-based functions are varied. Here we highlight some of these examples and provide an outlook on how the functional mechanisms of a lncRNA gene can be determined.

L., has been utilized for ages in ancient medicine, as well as in cooking and food coloring. Recently, the biological activities of turmeric and curcumin have been thoroughly investigated. The studies mainly focused on their antioxidant, antitumor, anti-inflammatory, neuroprotective, hepatoprotective, and cardioprotective impacts. This review seeks to provide an in-depth, detailed discussion of curcumin usage within the food processing industries and its effect on health support and disease prevention. Curcumin's bioavailability, bio-efficacy, and bio-safety characteristics, as well as its side effects and quality standards, are also discussed. Finally, curcumin's multifaceted uses, food appeal enhancement, agro-industrial techniques counteracting its instability and low bioavailability, nanotechnology and focused drug delivery systems to increase its bioavailability, and prospective clinical use tactics are all discussed.

Triple negative breast cancer (TNBC) is a deadly form of breast cancer due to the development of resistance to chemotherapy affecting over 30% of patients. New therapeutics and companion biomarkers are urgently needed. Recognizing the elevated expression of glucose transporter 1 (GLUT1, encoded by SLC2A1) and associated metabolic dependencies in TNBC, we investigated the vulnerability of TNBC cell lines and patient-derived samples to GLUT1 inhibition. We report that genetic or pharmacological inhibition of GLUT1 with BAY-876 impairs the growth of a subset of TNBC cells displaying high glycolytic and lower oxidative phosphorylation (OXPHOS) rates. Pathway enrichment analysis of gene expression data suggests that the functionality of the E2F pathway may reflect to some extent OXPHOS activity. Furthermore, the protein levels of retinoblastoma tumor suppressor (RB1) strongly correlate with the degree of sensitivity to GLUT1 inhibition in TNBC, where RB1-negative cells are insensitive to GLUT1 inhibition. Collectively, our results highlight a strong and targetable RB1-GLUT1 metabolic axis in TNBC and warrant clinical evaluation of GLUT1 inhibition in TNBC patients stratified according to RB1 protein expression levels.

Unmanned Aerial Vehicles (UAVs) are increasingly being used in many challenging and diversified applications. These applications belong to the civilian and the military fields. To name a few; infrastructure inspection, traffic patrolling, remote sensing, mapping, surveillance, rescuing humans and animals, environment monitoring, and Intelligence, Surveillance, Target Acquisition, and Reconnaissance (ISTAR) operations. However, the use of UAVs in these applications needs a substantial level of autonomy. In other words, UAVs should have the ability to accomplish planned missions in unexpected situations without requiring human intervention. To ensure this level of autonomy, many artificial intelligence algorithms were designed. These algorithms targeted the guidance, navigation, and control (GNC) of UAVs. In this paper, we described the state of the art of one subset of these algorithms: the deep reinforcement learning (DRL) techniques. We made a detailed description of them, and we deduced the current limitations in this area. We noted that most of these DRL methods were designed to ensure stable and smooth UAV navigation by training computer-simulated environments. We realized that further research efforts are needed to address the challenges that restrain their deployment in real-life scenarios.

We investigated the effect of ultrasound-assisted pH shift treatment on the micro-particle, molecular, and spatial structure of rapeseed protein isolates (RPI). Various ultrasonic frequency modes (fixed, and sweep) was used. Protein characterization by the indexes: particle size, zeta potential, sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), scanning electron microscopy (SEM), free sulfhydryl (SH), surface hydrophobicity (Ho), Fourier transform infrared Spectrum (FTIR) and fluorescence intensity was studied to elucidate the changes in solubility and structural attributes of RPI. The results showed that ultrasonic frequency and working modes substantially altered the structure, and modified the solubility of RPI. Ultra + pH mode at fixed frequency of 20 kHz had the best effect on the solubility of RPI. Under the condition of ultra + pH mode, 20 kHz at pH 12.5, solubility, compared to control, increased from 8.90% to 66.84%; and the change in molecular structure of RPI was characterized by smaller particles (from 330.90 to 115.77 nm), high zeta potential (from -17.95 to -14.43 mV, p < 0.05), and increased free sulfhydryl (from 11.63 to 24.50 µmol/g) compared to control. Likewise, surface hydrophobicity increased (from 2053.9 to 2649.4, p < 0.05), whilst ɑ-helix and random coil decreased (p < 0.05), compared to control. The fluorescence spectroscopy and FTIR spectroscopy showed that the secondary and tertiary structure of the RPI were altered. These observations revealed that changes in RPI structure was the direct factor affecting solubility. In conclusion, ultrasound assisted pH shift treatment was proven to be an effective method for the modification of protein, with promising application in food industry.

Diabetic retinopathy (DR) is a disease resulting from diabetes complications, causing non-reversible damage to retina blood vessels. DR is a leading cause of blindness if not detected early. The currently available DR treatments are limited to stopping or delaying the deterioration of sight, highlighting the importance of regular scanning using high-efficiency computer-based systems to diagnose cases early. The current work presented fully automatic diagnosis systems that exceed manual techniques to avoid misdiagnosis, reducing time, effort and cost. The proposed system classifies DR images into five stages-no-DR, mild, moderate, severe and proliferative DR-as well as localizing the affected lesions on retain surface. The system comprises two deep learning-based models. The first model (CNN512) used the whole image as an input to the CNN model to classify it into one of the five DR stages. It achieved an accuracy of 88.6% and 84.1% on the DDR and the APTOS Kaggle 2019 public datasets, respectively, compared to the state-of-the-art results. Simultaneously, the second model used an adopted YOLOv3 model to detect and localize the DR lesions, achieving a 0.216 mAP in lesion localization on the DDR dataset, which improves the current state-of-the-art results. Finally, both of the proposed structures, CNN512 and YOLOv3, were fused to classify DR images and localize DR lesions, obtaining an accuracy of 89% with 89% sensitivity, 97.3 specificity and that exceeds the current state-of-the-art results.