Search

Multi-task learning in advanced driver assistance systems aims to endow models with the capacity to jointly handle multiple related tasks, such as object detection, depth estimation, and more. However, existing multi-task learning models largely rely on the extensive number of labelled data. In practice, the process of annotating data for multi-task training proves to be exceedingly costly, yet not always accurate. This study introduces an innovative setting named online multi-task inference learning that updates the multi-task model during inference. And we propose a Strong-Help-Weak (SHW) framework which aims to enhance weaker (or more challenging) tasks by leveraging guidance from closely related stronger (or easier) tasks. Specifically, we first build two benchmarks based on KITTI and BDD with four tasks (object detection, object depth estimation, lane line segmentation, and driving area segmentation). Then, we propose two novel modules inspired by two priors: 1) Detection-guided Depth Inference Learning (DetDis) module that leverages the inverse relationship between object size and distance to refine the predicted object distance; and 2) Area-guided Lane Line Inference Learning (AreaLane) module that utilises inclusion relationship between driving area and lane line to infer more accurate lane line. Both modules are efficient and can provide more reliable supervision for the corresponding weaker tasks (object distance estimation and lane line segmentation), respectively. Extensive experiments on the two benchmarks show that our SHW can obtain consistent improvements on the weaker tasks during the inference stage with low computational costs.

Background: The implication of gut microbiota in the gut-kidney axis affects the pathophysiology of chronic kidney disease (CKD). Gut microbiota composition changes during CKD. We aimed to determine the relative frequency of important gut microbiota members in end-stage renal disease (ERSD) patients before and after renal transplantation compared to healthy subjects.Methods: Fifteen kidney transplant patients and 10 healthy subjects were recruited in this case-control prospective study. Fecal samples were taken sequentially from all patients before kidney transplantation, 1 week, and 1 month after it. The relative frequency of Lactobacillusspp., Bifidobacteriumspp., Akkermansia muciniphila, Bacteroides fragilis, Escherichia coli, and Faecalibacterium pruasnitziiwere determined through quantitative PCR. The obtained data was statistically analyzed by Stata software (Stata Corporation, USA).Results: The mean log number of all bacteria was significantly higher in healthy individuals than kidney transplant recipients (p< 0.001) except for Lactobacillus where the mean levels were almost identical in the two groups (p= 0.67). Moreover, 20% (3) of patients developed a urinary tract infection. Besides, 2 (13.33%) patients were diagnosed with delayed graft function. There were no statistically significant differences regarding changing trends in bacteria log number of Akkermansia muciniphila(p= 0.12), Bacteroid fragilis(p= 0.75), Bifidobacterium(p= 0.99), Escherichia coli(p= 0.5), Faecalibacterium(p= 0.98), and Lactobacilli(p= 0.93) between patients with and without delayed graft function (DGF).Conclusion: Gut microbiota composition in patients with ESRD was significantly different from those without it. However, the microbiota profile did not significantly differ in patients with and without DGF.

BackgroundRoad traffic crashes and associated injuries and mortalities are one of the big public health challenges, especially in low/middle-income countries. Road safety lead agency (RSLA) is a vital factor in the sustainable prevention and promotes road safety. In the recent decades, various policy interventions have been conducted for road safety in Iran. This study aimed at exploring the challenges of RSLA from the perspectives of stakeholders at various levels.MethodsA qualitative study was conducted. In-depth interviews and document reviews were used for data collection. To conduct interviews, different stakeholders at various levels such as policy-makers, senior national authorities, researchers and faculty members were selected. Data collection was conducted between November 2019 and June 2020. Thematic content analysis approach was used for data analysis.ResultsThe experts’ perspectives were analysed and then categorised under five main themes including the role and position of the lead agency, the role and duties of the actors and players, translating policy into practice, intrasectoral and intersectoral cooperation and coordination and evidence production and application and a total of 22 subthemes were identified. The current structure of the RSLA is one of the main challenges emphasised by the research participants.ConclusionTo achieve significant improvements in road safety at the national level, a strong management system and leadership body is a critical issue. Organisational reform to establish an effective unique lead agency is proposed to cope with RSLA challenges.

•Urological cancers include prostate, bladder and renal tumors that affect many individuals around the world.•Dysregulation of Wnt has been associated with tumorigenesis.•Wnt upregulation In PCa enhances carcinogenesis and can mediate proliferation nd metastasis.•Wnt upregulation in bladder cancer (BC) can enhance malignancy and mediate chemoresistance.•Overexpression of Wnt has been correlated with increase renal cancer progression.

Scene understanding plays a crucial role in autonomous driving by utilizing sensory data for contextual information extraction and decision making. Beyond modeling advances, the enabler for vehicles to become aware of their surroundings is the availability of visual sensory data, which expand the vehicular perception and realizes vehicular contextual awareness in real-world environments. Research directions for scene understanding pursued by related studies include person/vehicle detection and segmentation, their transition analysis, lane change, and turns detection, among many others. Unfortunately, these tasks seem insufficient to completely develop fully-autonomous vehicles i.e., achieving level-5 autonomy, travelling just like human-controlled cars. This latter statement is among the conclusions drawn from this review paper: scene understanding for autonomous driving cars using vision sensors still requires significant improvements. With this motivation, this survey defines, analyzes, and reviews the current achievements of the scene understanding research area that mostly rely on computationally complex deep learning models. Furthermore, it covers the generic scene understanding pipeline, investigates the performance reported by the state-of-the-art, informs about the time complexity analysis of avant garde modeling choices, and highlights major triumphs and noted limitations encountered by current research efforts. The survey also includes a comprehensive discussion on the available datasets, and the challenges that, even if lately confronted by researchers, still remain open to date. Finally, our work outlines future research directions to welcome researchers and practitioners to this exciting domain.

BackgroundCohort studies play essential roles in assessing causality, appropriate interventions. The study, Post-crash Prospective Epidemiological Research Studies in IrAN Traffic Safety and Health Cohort, aims to investigate the common health consequences of road traffic injuries (RTIs) postcrash through multiple follow-ups.MethodsThis protocol study was designed to analyse human, vehicle and environmental factors as exposures relating to postcrash outcomes (injury, disability, death, property damage, quality of life, etc). Population sources include registered injured people and followed up healthy people in precrash cohort experienced RTIs. It includes four first-year follow-ups, 1 month (phone-based), 3 months (in-person, video/phone call), 6 and 12 months (phone-based) after crash. Then, 24-month and 36-month follow-ups will be conducted triennially. Various questionnaires such as Post-traumatic Stress Disorder Questionnaire, Patient Health Questionnaire, WHO Disability Assessment Schedules, Cost-related Information, etc are completed. Counselling with a psychiatrist and a medical visit by a practitioner are provided accompanied by extra tools (simulator-based driving assessment, and psychophysiological tests). Through preliminary recruitment plan, 5807, 2905, 2247 and 1051 subjects have been enrolled, respectively at the baseline, first, second and third follow-ups by now. At baseline, cars and motorcycles accounted for over 30% and 25% of RTIs. At first follow-up, 27% of participants were pedestrians engaged mostly in car crashes. Around a fourth of injuries were single injuries. Car occupants were injured in 40% of collisions.DiscussionThe study provides an opportunity to investigate physical-psychosocial outcomes of RTIs, predictors and patterns at follow-up phases postinjury through longitudinal assessments, to provide advocates for evidence-based safety national policy-making.

Iran holds the world's second-largest proven oil reserves and fourth-largest gas reserves. However, the heavy reliance on highly subsidized fossil fuel resources, coupled with growing demand, has led to their depletion and increased greenhouse gas (GHG) emissions. This research introduces a cost-effective hybrid system, based on integrated strategic resource planning, to analyze the development of renewable energy sources. The goal is to minimize the overall cost of this hybrid power generation system on a national level from 2020 to 2050. This involves accounting for evolving power plant technologies, fluctuating fossil fuel prices, environmental concerns, demand-side management, and neural network models. The proposed model aids policymakers in medium and long-term planning and assesses the impact of renewable technology adoption and distributed generation on energy consumption and GHG emissions in Iran's power plant sector. The analysis across different scenarios reveals the most efficient way to utilize the full capacity of each power plant for installation and production. Though cheap reserves of hydrocarbon fuels have shaded the use of renewable sources and efficient power plants in Iran, the share of total renewable energy sources in the production cycle will rise and the emission of greenhouse gases will be diminished.

Background: Neurodegenerative diseases are a progressive loss of structure and/or function of neurons. Weak therapeutic response and progressive nature of the diseases, as well as a wide range of side effects caused by conventional therapeutic approaches make patients seek for complementary and alternative medicine. Objective: The aim of the present paper is to discuss the neuropharmacological basis of medicinal plants and their principle phytochemicals which have been used in traditional Persian medicine for different types of neurodegenerative diseases. Results: Medicinal plants introduced in traditional Persian medicine perform beneficial effects in neurodegenerative diseases via various cellular and molecular mechanisms including suppression of apoptosis mediated by an increase in the expression of anti-apoptotic agents (e.g. Bcl-2) as well as a decrease in the expression and activity of proapoptotic proteins (e.g. Bax, caspase 3 and 9). Alleviating inflammatory responses and suppressing the expression and function of pro-inflammatory cytokines like Tumor necrosis factor α and interleukins, as well as improvement in antioxidative performance mediated by superoxide dismutase and catalase, are among other neuroprotective mechanisms of traditional medicinal plants. Modulation of transcription, transduction, intracellular signaling pathways including ERK, p38, and MAPK, with upstream regulatory activity on inflammatory cascades, apoptosis and oxidative stress associated pathways, play an essential role in the preventive and therapeutic potential of the plants in neurodegenerative diseases. Conclusion: Medicinal plants used in traditional Persian medicine along with their related phytochemicals by affecting various neuropharmacological pathways can be considered as future drugs or adjuvant therapies with conventional pharmacotherapeutics; though, further clinical studies are necessary for the confirmation of their safety and efficacy.

Avoiding unnecessary travel and reducing the number of essential travels are among the effective strategies for preventing the spread of COVID-19 during the pandemic. Given that it is impossible to avoid essential travel, health protocols should be observed to prevent disease transmission. The extent to which health protocols are observed during the trip should be accurately assessed by a valid questionnaire. Therefore, this study aims to develop and validate a questionnaire to assess compliance with COVID-19 prevention protocols during travel.

Artistic rendering of human portraits is different and more challenging than that of landscapes or flowers. Issues are eye, nose, and mouth regions (i.e., facial features) where we need to represent their natural emotions. Shades or darkness around eyes, or shininess at nose tips may negatively impact the rendering result if not properly dealt with. Similarly, a lighter colour around the mouth region caused by lighting might produce some disturbing artefacts. The proposed computerised method attempts to be adaptive to those sensitive areas by utilising a face analysis module. First, the program detects main facial segments and features. Then it utilises a blending of various filtering parameters aiming at an adequate final portrait that represents the subject’s original facial expression, while still supporting a non-photorealistic artistic rendering as the perceived impression.

We have investigated the spectral properties and the multimode structure of joint states of photon pair produced by parametric down conversion in periodically poled nonlinear crystals (PPNC) by adding random length to each domain. We have also analysed the effect of additional random length on the output efficiency by considering two different configurations of PPNC and linear crystals, which we call (+, ‐) (PN) and (+, 0,−, 0) (PZNZ) configuration. In this paper we wish to study the effect of length fluctuations of segments which may occur in the preparation of PPNC in the laboratory. We aim at finding out how much error is reasonable in the process. In order to do this, we added random length to each segment and obtained the relevance between the output intensity and length fluctuations of the crystals.

One of the most challenging and non-trivial tasks in robot-based rescue operations is the Hazardous Materials (HAZMAT) sign detection in dangerous operation fields, in order to prevent further unexpected disasters. Each HAZMAT sign has a specific meaning that the rescue robot should detect and interpret to take a safe action, accordingly. Accurate HAZMAT detection and real-time processing are the two most important factors in such robotics applications. Furthermore, the rescue robot should cope with some secondary challenges such as image distortion and restricted CPU and computational resources, embedded in the robot. In this research, we propose a CNN-Based pipeline called DeepHAZMAT for HAZMAT sign detection and segmentation in four steps: (1) Input data volume optimisation before feeding into the CNN network, (2) Application of a YOLO-based structure to collect the required visual information from the hazardous areas, (3) HAZMAT sign segmentation and separation from the background using adaptive GrabCut technique, and (4) Post-processing optimisation using morphological operators and convex hull algorithms. In spite of the utilisation of a very limited CPU and memory resources, the experimental results show the proposed method has successfully maintained a better performance in terms of detection-speed and detection-accuracy, compared to classical and modern state-of-the-art methods.