Search

Background: Deep learning (DL)-assisted detection and segmentation of intracranial hemorrhage stroke in noncontrast computed tomography (NCCT) scans are well-established, but evidence on this topic is lacking. Materials and methods: PubMed and Embase databases were searched from their inception to November 2023 to identify related studies. The primary outcomes included sensitivity, specificity, and the Dice Similarity Coefficient (DSC); while the secondary outcomes were positive predictive value (PPV), negative predictive value (NPV), precision, area under the receiver operating characteristic curve (AUROC), processing time, and volume of bleeding. Random-effect model and bivariate model were used to pooled independent effect size and diagnostic meta-analysis data, respectively. Results: A total of 36 original studies were included in this meta-analysis. Pooled results indicated that DL technologies have a comparable performance in intracranial hemorrhage detection and segmentation with high values of sensitivity (0.89, 95% CI: 0.88-0.90), specificity (0.91, 95% CI: 0.89-0.93), AUROC (0.94, 95% CI: 0.93-0.95), PPV (0.92, 95% CI: 0.91-0.93), NPV (0.94, 95% CI: 0.91-0.96), precision (0.83, 95% CI: 0.77-0.90), DSC (0.84, 95% CI: 0.82-0.87). There is no significant difference between manual labeling and DL technologies in hemorrhage quantification (MD 0.08, 95% CI: -5.45-5.60, P= 0.98), but the latter takes less process time than manual labeling (WMD 2.26, 95% CI: 1.96-2.56, P=0.001). Conclusion: This systematic review has identified a range of DL algorithms that the performance was comparable to experienced clinicians in hemorrhage lesions identification, segmentation, and quantification but with greater efficiency and reduced cost. It is highly emphasized that multicenter randomized controlled clinical trials will be needed to validate the performance of these tools in the future, paving the way for fast and efficient decision-making during clinical procedure in patients with acute hemorrhagic stroke. [ABSTRACT FROM AUTHOR]

Abstract Background To investigate the morphological parameters of the vastus medialis obliquus (VMO) muscle and delineate its importance in the maintenance of patellofemoral joint stability. Methods The magnetic resonance imaging data of seventy-five knees (fifty-four patients) with recurrent lateral patella dislocation (LPD) and seventy-five knees (seventy patients) without recurrent LPD were retrospectively analysed. Five morphological parameters related to the VMO (elevation in the sagittal plane and coronal plane, craniocaudal extent, muscle-fibre angulation, cross-sectional area ratio) and two patella tilt parameters (patella tilt angle, bisect offset ratio) were measured in MR images. The independent-samples t test or chi-square test was used for statistical comparisons. Results The mean ages of the patients in the recurrent LPD group and control group were 22.1 ± 9.9 years and 24.0 ± 6.5 years, respectively. Eighteen out of seventy-five (24%) patients MRI showed VMO injuries. Compared with the control group, the patients with recurrent LPD showed significantly higher sagittal VMO elevation (10.4 ± 2.3 mm vs. 4.1 ± 1.9 mm), coronal VMO elevation (15.9 ± 5.7 mm vs. 3.9 ± 3.7 mm), muscle-fibre angulation (35.4 ± 8.0° vs. 27.9 ± 6.3°), patella tilt angle (25.9 ± 10.7° vs. 9.1 ± 5.2°), and bisect offset ratio values (0.9 ± 0.3 vs. 0.5 ± 0.1) and significantly lower craniocaudal extent (13.7 ± 5.3 mm vs. 16.7 ± 5.1 mm) and cross-sectional area ratio values (0.05 ± 0.02 vs. 0.07 ± 0.02). Conclusions The results showed that abnormalities in the VMO and patella tilt were clearly present in recurrent LPD patients compared with normal people.

Intelligent connected vehicles (ICVs) are usually expected to run a huge number of emerging smart applications (e.g., autonomous driving, safety early warning, natural language processing, etc.) for assisting both the drivers and passengers in vehicular environments. These kinds of applications typically require significant computing power to perform computation-intensive and latency-sensitive tasks generated by the vehicle sensors for low-latency response. However, the limited computation capacity of on-board computers makes it difficult to satisfy the computation requirements of quality-of-experience (QoE)-demanding applications. To tackle the challenge, fog/edge computing is proposed as innovative computing paradigms to extend computing capacity to the network edge to meet the requirements. This Special Section in IEEE Access focuses on innovative technologies and solutions enabled communication and edge computing toward ICVs.

In this article, we present a comprehensive study with an experimental analysis of federated deep learning approaches for cyber security in the Internet of Things (IoT) applications. Specifically, we first provide a review of the federated learning-based security and privacy systems for several types of IoT applications, including, Industrial IoT, Edge Computing, Internet of Drones, Internet of Healthcare Things, Internet of Vehicles, etc. Second, the use of federated learning with blockchain and malware/intrusion detection systems for IoT applications is discussed. Then, we review the vulnerabilities in federated learning-based security and privacy systems. Finally, we provide an experimental analysis of federated deep learning with three deep learning approaches, namely, Recurrent Neural Network (RNN), Convolutional Neural Network (CNN), and Deep Neural Network (DNN). For each deep learning model, we study the performance of centralized and federated learning under three new real IoT traffic datasets, namely, the Bot-IoT dataset, the MQTTset dataset, and the TON_IoT dataset. The goal of this article is to provide important information on federated deep learning approaches with emerging technologies for cyber security. In addition, it demonstrates that federated deep learning approaches outperform the classic/centralized versions of machine learning (non-federated learning) in assuring the privacy of IoT device data and provide the higher accuracy in detecting attacks.

A high resolution dataset is one of the prerequisites for tea chrysanthemum detection with deep learning algorithms. This is crucial for further developing a selective chrysanthemum harvesting robot. However, generating high resolution datasets of the tea chrysanthemum with complex unstructured environments is a challenge. In this context, we propose a novel tea chrysanthemum – generative adversarial network (TC-GAN) that attempts to deal with this challenge. First, we designed a non-linear mapping network for untangling the features of the underlying code. Then, a customized regularization method was used to provide fine-grained control over the image details. Finally, a gradient diversion design with multi-scale feature extraction capability was adopted to optimize the training process. The proposed TC-GAN was compared with 12 state-of-the-art generative adversarial networks, showing that an optimal average precision (AP) of 90.09% was achieved with the generated images (512 × 512) on the developed TC-YOLO object detection model under the NVIDIA Tesla P100 GPU environment. Moreover, the detection model was deployed into the embedded NVIDIA Jetson TX2 platform with 0.1 s inference time, and this edge computing device could be further developed into a perception system for selective chrysanthemum picking robots in the future.

Abstract Background COVID-19 is a highly infectious respiratory disease. No therapeutics have yet been proven effective for treating severe COVID-19. Objectives To determine whether human umbilical cord mesenchymal stem cell infusion may be effective and safe for the treatment of severe COVID-19. Methods Patients with severe COVID-19 were randomly divided into 2 groups: the standard treatment group and the standard treatment plus hUC-MSC infusion group. The incidence of progression from severe to critical illness, 28-day mortality, clinical symptom improvement, time to clinical symptom improvement, hematologic indicators including C-reactive protein, lymphocyte number, and interleukin 6, and imaging changes were observed and compared between the two groups. Measurements and main results The incidence of progression from severe to critical illness and the 28-day mortality rate were 0 in the hUC-MSC treatment group, while 4 patients in the control group deteriorated to critical condition and received invasive ventilation; 3 of them died, and the 28-day mortality rate was 10.34%. In the hUC-MSC treatment group, the time to clinical improvement was shorter than that in the control group. Clinical symptoms of weakness and fatigue, shortness of breath, and low oxygen saturation obviously improved beginning on the third day of stem cell infusion and reached a significant difference on day 7. CRP and IL-6 levels were significantly lower from day 3 of infusion, the time for the lymphocyte count to return to the normal range was significantly faster, and lung inflammation absorption was significantly shorter on CT imaging in the hUC-MSC group than in the control group. Conclusions Intravenous transplantation of hUC-MSCs is a safe and effective method that can be considered a salvage and priority treatment option for severe COVID-19. Trial registration Chinese Clinical Trial Registration; ChiCTR2000031494; Registered on 2 April 2020; http:// www.medresman.org

Abstract Background A comparative analysis of the strengths and weaknesses of three different methods for radiologic evaluation of patellofemoral instability (PFI). Methods Computed tomography (CT) and magnetic resonance imaging (MRI) were performed in 47 patients with or without PFI. The tibial tubercle-trochlear groove (TT-TG) distance was measured by two observers through conventional CT and three-dimensional CT reconstruction (TDR-TT-TG) respectively and the tibial tubercle-posterior cruciate ligament (TT-PCL) distance with MRI. The intraclass correlation coefficient (ICC) was used to evaluate the interobserver reliability. In addition, the differences of three measurements between different patients were compared. The consistency of TT-TG and TDR-TT-TG was analyzed by the Bland-Altman method. Results The ICCs of three measurements were high between two observers; the results were TT-TG (ICC = 0.852), TDR-TT-TG (ICC = 0.864), and TT-PCL (ICC = 0.758). The values of PFI patients were significantly higher than those of non-PFI patients, and the mean TT-TG, TDR-TT-TG, and TT-PCL distance in patients with PFI were 19.0 ± 3.8 mm, 19.0 ± 3.7 mm, and 25.1 ± 3.6 mm, respectively. There was no statistically significant difference between the TT-TG distance and the TDR-TT-TG distance, we found no significant difference. The Bland-Altman analysis showed that the TDR-TT-TG distance was in good agreement with the TT-TG distance. Conclusion All three methods can be used to assess PFI; the TDR-TT-TG measurement method has superior operability and better interobserver consistency. It may be an alternative method to the conventional TT-TG distance measurement.

A series of aryloxyacetic acid derivatives were designed and synthesized as 4-hydoxyphenylpyruvate dioxygenase (HPPD) inhibitors. Preliminary bioassay results reveal that these derivatives are promising Arabidopsis thaliana HPPD (AtHPPD) inhibitors, in particular compounds I12 (Ki = 0.011 µM) and I23 (Ki = 0.012 µM), which exhibit similar activities to that of mesotrione, a commercial HPPD herbicide (Ki = 0.013 µM). Furthermore, the newly synthesized compounds show significant greenhouse herbicidal activities against tested weeds at dosages of 150 g ai/ha. In particular, II4 exhibited high herbicidal activity for pre-emergence treatment that was slightly better than that of mesotrione. In addition, compound II4 was safe for weed control in maize fields at a rate of 150 g ai/ha, and was identified as the most potent candidate for a novel HPPD inhibitor herbicide. The compounds described herein may provide useful guidance for the design of new HPPD inhibiting herbicides and their modification.

This paper presents research challenges on security and privacy issues in the field of green IoT-based agriculture. We start by describing a four-tier green IoT-based agriculture architecture and summarizing the existing surveys that deal with smart agriculture. Then, we provide a classification of threat models against green IoT-based agriculture into five categories, including, attacks against privacy, authentication, confidentiality, availability, and integrity properties. Moreover, we provide a taxonomy and a side-by-side comparison of the state-of-the-art methods toward secure and privacy-preserving technologies for IoT applications and how they will be adapted for green IoT-based agriculture. In addition, we analyze the privacy-oriented blockchain-based solutions as well as consensus algorithms for IoT applications and how they will be adapted for green IoT-based agriculture. Based on the current survey, we highlight open research challenges and discuss possible future research directions in the security and privacy of green IoT-based agriculture.

Serious challenges for to drive agricultural sustainability combined with climate crisis issues have induced an urgent need to decarbonise agriculture. In this paper, we briefly introduce a novel concept of the Photovoltaic Agricultural Internet of Things (PAIoT). This system approach fuses agricultural production with renewable power generation and control via an IoT platform. We discuss PAIoT applications and potential to realize the next generation of smart farming. In addition, we provide a review of key issues on the feasibility of PAIoT and further propose novel techniques to mitigate these key issues.

Fault diagnosis of rolling bearing is of great importance to ensure high reliability and safety in the industrial machinery system. Entropy measures are useful non-linear indicators for time series complexity analysis and have been widely applied in bearing fault diagnosis in the past decade. In this paper, an improved entropy measure is proposed, named Adaptive Multiscale Weighted Permutation Entropy (AMWPE). Then, a new rolling bearing fault diagnosis method is developed based on the AMWPE and multi-class SVM. For comparison, an experimental bearing dataset is analyzed using the AMWPE and conventional entropy measures, and then multi-class SVM is adopted for fault type classification. Further, the robustness of different entropy measures against noise is studied by analyzing noisy signals with various Signal-to-Noise Ratios (SNRs). The experimental results have demonstrated the effectiveness of the proposed method in bearing fault diagnosis under different fault types, severity degrees, and SNR levels.

In modern transportation, traffic congestion has become an urgent problem in large and medium-sized cities. In smart transportation systems, it is an effective solution to design load balancing path planning algorithms that can dynamically adapt to traffic conditions in order to avoid congestion. In this work, a traffic path planning algorithm based on data prediction (TPPDP) is proposed to find the path with the shortest travel time, which is built on a predictive model based on historical traffic data and current traffic information. Furthermore, a path planning algorithm based on data prediction with load balancing (TPPDP-LB) is also proposed, which combines the predicted information and the number of concurrent requests to achieve the path with shortest travel time while maintaining global load balancing. A specific distributed computing framework for TPPDP-LB algorithm is designed to reduce the runtime of the algorithm. The simulation results proved that both TPPDP and TPPDP-LB algorithms have the advantage of shortest travel time, and TPPDP-LB algorithm achieves load balancing of computing. It is also proved that the distributed computing framework designed for TPPDP-LP algorithm can effectively reduce the runtime of system as well as keep the accuracy of algorithm.

With growing attention from both academia and industry, Industrial Internet of Things (IIoT), comprised of a multitude of connected devices, is supposed to monitor, collect, exchange, analyze, and instantly act on information to intelligently change the industrial device’s behavior or the industrial environment (e.g., autonomous reaction to unexpected changes in production by effectively detecting failures and triggering maintenance processes). Toward this goal, the convergence of sensor networks, cloud computing, and big data in IIoT is recently identified as an essential component. Specifically, to capture various data in IIoT, sensor networks are deployed. Meanwhile, to store and process data powerfully in IIoT, cloud computing platforms are developed. Moreover, to glean valuable findings from the massive data in IIoT, big data analytics tools are utilized. However, to integrate sensor networks, cloud computing, and big data in IIoT in a robust way, there are a lot of tough issues to be solved with respect to various aspects (e.g., framework, greenness, security, quality of service, etc.).

Pseudomonas aeruginosa (PA) is an important pathogen that has been proven to colonize and cause infection in the respiratory tract of patients with structural lung diseases and to lead to bronchial fibrosis. The development of pulmonary fibrosis is a complication of PA colonization of the airway, resulting from repeated infection, damage and repair of the epithelium. Bronchial epithelial cell epithelial-mesenchymal transition (EMT) plays a vital role in bronchial fibrosis. To date, research on bronchial epithelial cell EMT caused by PA-secreted virulence factors has not been reported. Here, we found that PA3611 protein stimulation induced bronchial epithelial cell EMT with mesenchymal cell marker upregulation and epithelial cell marker downregulation. Moreover, integrin αvβ6 expression and TGF-β1 secretion were markedly increased, and p38 MAPK phosphorylation and NF-κB p65 subunit phosphorylation were markedly enhanced. Further research revealed that PA3611 promoted EMT via integrin αvβ6-mediated TGF-β1-induced p38/NF-κB pathway activation. The function of PA3611 was also verified in PA-infected rats, and the results showed that ΔPA3611 reduced lung inflammation and EMT. Overall, our results revealed that PA3611 promoted EMT via integrin αvβ6-mediated TGF-β1-induced p38/NF-κB pathway activation, suggesting that PA3611 acts as a crucial virulence factor in bronchial epithelial cell EMT and is a potential target for the clinical treatment of bronchial EMT and fibrosis caused by chronic PA infection.

Soybean yield is a highly complex trait determined by multiple factors such as genotype, environment, and their interactions. The earlier the prediction during the growing season the better. Accurate soybean yield prediction is important for germplasm innovation and planting environment factor improvement. But until now, soybean yield has been determined by weight measurement manually after soybean plant harvest which is time-consuming, has high cost and low precision. This paper proposed a soybean yield in-field prediction method based on bean pods and leaves image recognition using a deep learning algorithm combined with a generalized regression neural network (GRNN). A faster region-convolutional neural network (Faster R-CNN), feature pyramid network (FPN), single shot multibox detector (SSD), and You Only Look Once (YOLOv3) were employed for bean pods recognition in which recognition precision and speed were 86.2, 89.8, 80.1, 87.4%, and 13 frames per second (FPS), 7 FPS, 24 FPS, and 39 FPS, respectively. Therefore, YOLOv3 was selected considering both recognition precision and speed. For enhancing detection performance, YOLOv3 was improved by changing IoU loss function, using the anchor frame clustering algorithm, and utilizing the partial neural network structure with which recognition precision increased to 90.3%. In order to improve soybean yield prediction precision, leaves were identified and counted, moreover, pods were further classified as single, double, treble, four, and five seeds types by improved YOLOv3 because each type seed weight varies. In addition, soybean seed number prediction models of each soybean planter were built using PLSR, BP, and GRNN with the input of different type pod numbers and leaf numbers with which prediction results were 96.24, 96.97, and 97.5%, respectively. Finally, the soybean yield of each planter was obtained by accumulating the weight of all soybean pod types and the average accuracy was up to 97.43%. The results show that it is feasible to predict the soybean yield of plants in situ with high precision by fusing the number of leaves and different type soybean pods recognized by a deep neural network combined with GRNN which can speed up germplasm innovation and planting environmental factor optimization.

Sensors have been used in various agricultural production scenarios due to significant advances in the Agricultural Internet of Things (Ag-IoT), leading to smart agriculture. Intelligent control or monitoring systems rely heavily on trustworthy sensor systems. Nonetheless, sensor failures are likely due to various factors, including key equipment malfunction or human error. A faulty sensor can produce corrupted measurements, resulting in incorrect decisions. Early detection of potential faults is crucial, and fault diagnosis techniques have been proposed. The purpose of sensor fault diagnosis is to detect faulty data in the sensor and recover or isolate the faulty sensors so that the sensor can finally provide correct data to the user. Current fault diagnosis technologies are based mainly on statistical models, artificial intelligence, deep learning, etc. The further development of fault diagnosis technology is also conducive to reducing the loss caused by sensor failures.

Algebraic persistent fault analysis (APFA), which combines algebraic analysis with persistent fault attacks, brings new challenges to the security of lightweight block ciphers and has received widespread attention since its introduction. Threshold Implementation (TI) is one of the most widely used countermeasures for side channel attacks. Inspired by this method, the SKINNY block cipher adopts the S_box decomposition to reduce the number of variables in the set of algebraic equations and the number of Conjunctive Normal Form (CNF) equations in this paper, thus speeding up the algebraic persistent fault analysis and reducing the number of fault ciphertexts. In our study, we firstly establish algebraic equations for full-round faulty encryption, and then analyze the relationship between the number of fault ciphertexts required and the solving time in different scenarios (decomposed S_boxes and original S_box). By comparing the two sets of experimental results, the success rate and the efficiency of the attack are greatly improved by using S_box decomposition. In this paper, We can recover the master key in a minimum of 2000s using 11 pairs of plaintext and fault ciphertext, while the key recovery cannot be done in effective time using the original S_box expression equations. At the same time, we apply S_box decomposition to another kind of algebraic persistent fault analysis, and the experimental results show that using S_box decomposition can effectively reduce the solving time and solving success rate under the same conditions.

Pulmonary drug delivery systems (PDDS) have gained particular attention of pharmaceutical scientists due to their potency in the therapy of both pulmonary diseases and systemic diseases. Metered dose inhalers (MDI) as a delivery approach of PDDS have ample advantages over other approaches. No drying process is associated with MDI, which avoids the chemical and physical instability. MDI are cost-effective, portable and can be self-administered. It was recently reported that MDI have occupied a large proportion of the respiratory drug market since the last century. To facilitate the future studies on MDI, this work was aimed at conducting bibliometric analysis of the publications of MDI in Science Citation Index Expanded database of Web of Science from 2000 to 2020 (2858 in total). The documents were processed by Clarivate Analytic tool equipped by Web of Science, VOSviewer, Statistical Analysis Toolkit for Informetric (SATI) and bibliometric online platform, and the data were visualized. After describing the detailed bibliometric profile of MDI publications, which included publication years, countries/regions, organizations, research areas, publication media, authorship and funding agencies, we assessed the publication tendencies by virtue of analysis of co-citation, bibliographic coupling, keywords and co-occurrence. Based on the results, we put forward three promising topics for future studies of MDI. Taken together, we believe that MDI became a topic under investigation around the world, and will still be important for fundamental and translational researches.

Abstract Semi‐transparent perovskite solar cells (ST‐PSCs) engendered enormous attention for practical applications such as power generation windows. However, it is still challenging to achieve high‐performance, robust and neutral‐colored ST‐PSCs. Herein we demonstrate a moth‐eye‐inspired structure (MEIS) for light‐trapping photons in the wavelength range where the human eye is less perceptive. This biomimetic structure contributes to the improvements in ST‐PSCs performance and visual appearance simultaneously. Consequently, a record high figure‐of‐merit for ST‐PSC, defined as the product of power conversion efficiency and the average visible transmittance, is achieved. Meanwhile, the optical appearance is converted to a desired near‐neutral color after introducing the MEIS. The investigation of ST‐PSCs with long‐term stability is implemented via engineering blend ratio of halides. The modified device exhibited appealing tolerance against moisture and solar irradiation. This work reveals an in‐depth understanding of light trapping along with modifying the visual appearance of solar cells.