Large-scale Detection Research Articles

Using machine learning to detect sarcopenia from electronic health records.

Sarcopenia (low muscle mass and strength) causes dysmobility and loss of independence. Sarcopenia is often not directly coded or described in electronic health records (EHR). The objective was to improve sarcopenia detection using structured data from EHR. Adults undergoing musculoskeletal testing (December 2017-March 2020) were classified as meeting sarcopenia thresholds for 0 (controls), ≥1 (Sarcopenia-1), or ≥2 (Sarcopenia-2) tests. Electronic health record diagnoses, medications, and laboratory testing were extracted from the Indiana Network for Patient Care. Five machine learning models were applied to EHR data for predicting sarcopenia. Of 1304 participants, 1055 were controls, 249 met Sarcopenia-1 and 76 met Sarcopenia-2. Sarcopenic participants were older, with higher fat mass, Charlson Comorbidity Index, and more chronic diseases. All models performed better for Sarcopenia-2 than Sarcopenia-1. The top performing models for Sarcopenia-1 were Logistic Regression [area under the curve (AUC) 71.59 (95% confidence interval [CI], 71.51-71.66)] and Multi-Layer Perceptron [AUC 71.48 (95%CI, 71.00-71.97)]. The top performing models for Sarcopenia-2 were Logistic Regression [AUC 91.44 (95%CI, 91.28-91.60)] and Support Vector Machine [AUC 90.81 (95%CI, 88.41-93.20)]. For the best Logistic Regression Model, important sarcopenia predictors included diabetes mellitus, digestive system complaints, signs and symptoms involving the nervous, musculoskeletal and respiratory systems, metabolic disorders, and kidney or urinary tract disorders. Opioids, corticosteroids, and antihyperlipidemic drugs were also more common among sarcopenic participants. Applying machine learning models, sarcopenia can be predicted from structured data in EHR, which may be developed through future studies to facilitate large-scale early detection and intervention in clinical populations.

A Sidelobe-Aware Small Ship Detection Network for Synthetic Aperture Radar Imagery

Ship detection from synthetic aperture radar (SAR) remote sensing images is essential for monitoring water traffic and marine safety. Numerous methods for ship detection have been developed; however, the detection of small ships presents unique challenges. SAR image characteristics, such as the sidelobe effect and blurred outline induced by the special imaging mechanism, as well as the small ship size, are the primary factors that lower the detection accuracy. This paper provides a sidelobe-aware small ship detection network for synthetic aperture radar imagery. First, considering the sidelobe effect and blurred outline, dual-pooling, i.e., average pooling and max pooling, was utilized to build a feature extraction module that lowered the effects of strong scattering points outside of the ship body and enhanced the ship body information. Second, as the bipartition process of the average pooling and maximum pooling caused some loss of original data information, different feature maps in the network were concatenated to construct a new network structure to compensate for the information lost and enrich the small ship features. Third, because the traditional loss function based on centroid distance and aspect ratio may result in the same loss function value for different prediction box sizes, a novel loss function based on the dual Euclidean distances of the corner point coordinates between the prediction box and the real box was proposed, which could accurately describe various overlapping box situations. Experiments using the Large-Scale SAR Ship Detection Dataset (LS-SSDD), SAR Ship Detection Dataset (SSDD), and AIR-SARShip dataset validated the efficacy and state-of-the-art performance.

Visual Object Detection for Privacy-Preserving Federated Learning

Visual object detection is a computer vision technique based on deep learning. However, due to privacy issues, it is highly challenging to build an object detection model according to the current methods on the centrally stored training dataset. Federated learning is an effective approach to solving the challenge of training data collection by using distributed training. We propose FedVisionBC, a blockchain-based federated learning system for visual object detection that addresses the challenges of single point of failure, model poisoning attacks, and membership inference attacks in traditional federated learning. In the FedVisionBC system, we set up an aggregation node and a verification node instead of a central server to solve the single point of failure problem. We establish a security mechanism that uses encryption techniques, verification nodes, and smart contracts to resist model poisoning attacks. Experimental results show that FedVisionBC can accomplish the object detection task when the percentage of malicious clients is less than 60%. We also propose a new algorithm, ADPFedAvg, to prevent membership inference attacks, which relies on user-level differential privacy technology and the federal average algorithm. Experimental results indicate that ADPFedAvg can achieve a large-scale visual object detection model with differential privacy protection, while only a negligible cost in predictive accuracy.

A Hybrid Approach for Android Malicious Software Classification

Android operating systems have grown in popularity and are currently being increasingly used on smartphone devices. Because of the quickly expanding quantity of Android malware and the potential safeguard of vast amounts of data kept on Android devices, the identification and categorization of Android malware is regarded as a big data challenge. Researchers started using deep learning (DL) techniques for malware detection, according to the most recent state-of-the-art studies. Nevertheless, researchers and practitioners face challenges such as the choice of DL architecture, extracting and processing features and evaluating results. In this paper, several traditional and hybrid machine learning (ML) models were developed and analyzed in detail to classify Android malware. First, some state-of-the-art DL model has been developed like multi-layer perceptron (MLP), deep neural network (DNN), recurrent neural network (RNN), long short-term memory (LSTM), convolutional neural network (CNN) and later several hybrid models have been developed by combining DNNs with support vector machine (SVM). Second, we have focused on maximizing achievement by fine-tuning a variety of configurations to guarantee the best possible combination of the hyper-parameters and attain the maximum statistical metric value. Finally, we have accomplished performance analysis between these hybrid models and a model based on VGG-16-CNN-1D along with SVM has been proposed to enhance the accuracy and efficacy of large-scale Android malware detection which outperforms all the other developed models using Drebin Dataset. Among the DL methods, DNN achieved the highest F1 score 99.07%. After doing a comparison between hybrid models, it was found that DNN-SVM achieved 99.12% while VGG16-CNN-1D-SVM achieved 98.56% precision, 98.78% recall and 99.20% F1 score, respectively.

A flexible perovskite homojunction with metallic ion doping for large-scale and high sensitivity X-ray detection

A large-scale flexible X-ray detector based on a perovskite p–n homojunction is reported in this work. The addition of a Bi3+-doped MAPbI3layer would effectively suppress the dark current and noise while maintaining high sensitivity.

Immersive training of clinical decision making with AI driven virtual patients - a new VR platform called medical tr.AI.ning.

Medical students need to be prepared for various situations in clinical decision-making that cannot be systematically trained with real patients without risking their health or integrity. To target system-related limitations of actor-based training, digital learning methods are increasingly used in medical education, with virtual reality (VR)- training seeming to have high potential. Virtually generated training scenarios allow repetitive training of highly relevant clinical skills within a protected, realistic learning environment. Thanks to Artificial Intelligence (AI), face-to-face interaction with virtual agents is feasible. Combining this technology with VR-simulations offers a new way of situated context-based, first-person training for medical students. The authors' aim is to develop a modular digital training platform for medical education with virtual, interactable agents and to integrate this platform into the medical curriculum. The medical tr.AI.ning platform will provide veridical simulation of clinical scenarios with virtual patients, augmented with highly realistic medical pathologies within a customizable, realistic situational context. Medical tr.AI.ning is scaled to four complementary developmental steps with different scenarios that can be used separately and so each outcome can successively be integrated early within the project. Every step has its own focus (visual, movement, communication, combination) and extends an author toolbox through its modularity. The modules of each step will be specified and designed together with medical didactics experts. To ensure constant improvement of user experience, realism, and medical validity, the authors will perform regular iterative evaluation rounds.Furthermore, integration of medical tr.AI.ning into the medical curriculum will enable long-term and large-scale detection of benefits and limitations of this approach, providing enhanced alternative teaching paradigms for VR technology.

TextDC: Exploring Multidimensional Text Detection via a New Benchmark and Solution

Text detection has been significantly boosted by the development of deep neural networks but most existing methods focus on a single kind of text instance (i.e., overlaid text, layered text, scene text). In this paper, we expand the text detection task from a single dimension to multiple dimensions, thus providing multi-type text descriptions for the scene and content analysis of videos. Specifically, we establish a new task to detect and classify text instances simultaneously, termed TextDC. As far as we know, existing benchmarks cannot meet the requirements of the proposed task. To this end, we collect a large-scale text detection and classification dataset, named Text3C, which is annotated using multilingual labels, location information, and text categories. Together with the collected dataset, we introduce a multi-stage and strict evaluation metric, which penalizes detection approaches for missing text instances, false positive detection, inaccurate location boxes, and error text categories, developing a new benchmark for the proposed TextDC task. In addition, we extend several state-of-the-art detectors by modifying the prediction head to solve the new task. Then, a generalized text detection and classification framework is designed and formulated. Extensive experiments using the updated methods are conducted on the established benchmark to verify the solvability of the proposed task, the challenges of the dataset, and the effectiveness of the solution.

Development of a reverse-transcription droplet digital PCR method for quantitative detection of Cucumber green mottle mosaic virus

Cucumber green mottle mosaic virus (CGMMV) is a re-emerging threat to the production of greenhouse cucumber and other Cucurbitaceae crops worldwide. This seed-borne virus can easily spread from a contaminated seed to seedlings and adjacent plants by mechanical contact between the foliage of diseased and healthy plants, causing extensive yield losses. An accurate method for detecting and quantifying this virus is urgently needed to ensure the safety of the global seed trade. Here, we report the development of a reverse-transcription droplet digital polymerase chain reaction (RT-ddPCR)-based method for specific and high-sensitive detection of CGMMV. By testing three primer–probe sets and optimizing reaction conditions, we showed that the newly developed RT-ddPCR method is highly specific and sensitive, with a detection limit of 1 fg/μL (0.39 copy/μL). The sensitivity of the RT-ddPCR method was compared with that of real-time fluorescence quantitative RT-PCR (RT-qPCR) using a series of plasmid dilutions and total RNAs extracted from infected cucumber seeds, and the detection limit of RT-ddPCR was 10 times higher than RT-qPCR with plasmid dilutions and 100 times higher than RT-qPCR for detecting CGMMV from infected cucumber seeds. The RT-ddPCR method was further assessed for detecting CGMMV from a total of 323 samples of Cucurbitaceae seeds, seedlings, and fruits as compared with the RT-qPCR method. We found that the infection rate of CGMMV on symptomatic fruits was as high as 100%, whereas infection rates were lower for seeds and lowest for seedlings. Notably, the results of two methods in detecting CGMMV from different cucurbit tissues showed the high consistency with Kappa value from 0.84 to 1.0, demonstrating that the newly developed RT-ddPCR method is highly reliable and practically useful for large-scale CGMMV detection and quantification.

An Anchor-Free Detection Algorithm for SAR Ship Targets with Deep Saliency Representation

Target detection in synthetic aperture radar (SAR) images has a wide range of applications in military and civilian fields. However, for engineering applications involving edge deployment, it is difficult to find a suitable balance of accuracy and speed for anchor-based SAR image target detection algorithms. Thus, an anchor-free detection algorithm for SAR ship targets with deep saliency representation, called SRDet, is proposed in this paper to improve SAR ship detection performance against complex backgrounds. First, we design a data enhancement method considering semantic relationships. Second, the state-of-the-art anchor-free target detection framework CenterNet2 is used as a benchmark, and a new feature-enhancing lightweight backbone, called LWBackbone, is designed to reduce the number of model parameters while effectively extracting the salient features of SAR targets. Additionally, a new mixed-domain attention mechanism, called CNAM, is proposed to effectively suppress interference from complex land backgrounds and highlight the target area. Finally, we construct a receptive-field-enhanced detection head module, called RFEHead, to improve the multiscale perception performance of the detection head. Experimental results based on three large-scale SAR target detection datasets, SSDD, HRSID and SAR-ship-dataset, show that our algorithm achieves a better balance between ship target detection accuracy and speed and exhibits excellent generalization performance.

Development of a Large-Scale Roadside Facility Detection Model Based on the Mapillary Dataset.

The detection of road facilities or roadside structures is essential for high-definition (HD) maps and intelligent transportation systems (ITSs). With the rapid development of deep-learning algorithms in recent years, deep-learning-based object detection techniques have provided more accurate and efficient performance, and have become an essential tool for HD map reconstruction and advanced driver-assistance systems (ADASs). Therefore, the performance evaluation and comparison of the latest deep-learning algorithms in this field is indispensable. However, most existing works in this area limit their focus to the detection of individual targets, such as vehicles or pedestrians and traffic signs, from driving view images. In this study, we present a systematic comparison of three recent algorithms for large-scale multi-class road facility detection, namely Mask R-CNN, YOLOx, and YOLOv7, on the Mapillary dataset. The experimental results are evaluated according to the recall, precision, mean F1-score and computational consumption. YOLOv7 outperforms the other two networks in road facility detection, with a precision and recall of 87.57% and 72.60%, respectively. Furthermore, we test the model performance on our custom dataset obtained from the Japanese road environment. The results demonstrate that models trained on the Mapillary dataset exhibit sufficient generalization ability. The comparison presented in this study aids in understanding the strengths and limitations of the latest networks in multiclass object detection on large-scale street-level datasets.

VERI: A Large-scale Open-Source Components Vulnerability Detection in IoT Firmware

IoT device manufacturers integrate open-source components (OSCs) to serve necessary and common functions for facilitating firmware development. However, outdated versions of OSC conceal N-day vulnerabilities and continue to function on IoT devices. The security risks can be predicted once we can identify the OSC versions employed in the firmware. Existing works make attempts at OSC version identification but fail to perform vulnerability detection on a large-scale IoT firmware due to i) unsuitable version identification method for IoT firmware scenario. ii) the lack of a large-scale version-vulnerability relation database. To this end, we propose a system VERI for large-scale vulnerability detection based on lightweight version identification. First, for OSC version identification, VERI leverages symbolic execution with static analysis to identify exact OSC versions even though there are many version-like strings in OSC. Second, VERI employs a deep learning-based method to extract OSC names and vulnerable version ranges from vulnerability descriptions, constructs and maintains an OSC version-vulnerability relation database to serve the vulnerability detection. Finally, VERI polls the relation database to confirm the N-day security risk of the OSC with identified version. The evaluation results show that VERI achieves 96.43% accuracy with high efficiency in OSC version identification. Meanwhile, the deep learning model accurately extracts the OSC names and versions from vulnerability descriptions dataset with 97.19% precision and 96.56% recall. Based on the model, we build a large-scale version-vulnerability relation database. Furthermore, we utilize VERI to conduct a large-scale analysis on 28,890 firmware and find 38,654 vulnerable OSCs with 266,109 N-day vulnerabilities, most of which are with high risks. From the detection results, we find that after the official patch for the vulnerability is released, manufacturers delay an average of 473 days to patch the firmware.

Feature-based algorithm for large-scale rice phenology detection based on satellite images

Knowledge of rice phenology is essential for understanding the agricultural practices and studying its impact on ecosystem services. However, so far, available global-scale rice phenology maps do not provide fine spatiotemporal details of rice phenology in a consistent framework because they rely on the compilation of statistical data. Thus, this paper proposes an algorithm that combines the complementary advantages of Sentinel-1 and Sentinel-2 satellite images to produce large-scale maps that depict rice phenology dynamics. The novelty of this algorithm lies in the correlation with rice phenology features, i.e., rice in water condition and rice color change. The time series of backscattering at Vertical-Horizontal (VH) polarization and Enhanced Vegetation Index (EVI) are proposed to recognize rice planting and heading dates, respectively. For the same time, the Normalized Difference Yellow Index (NDYI) is utilized to detect the rice harvest date for the first time. The proposed algorithm is applied to multiple spatial scales (prefecture, 0.5° gridcell, and site scales) and to multiple rice cropping systems (single, double, and triple croppings) in monsoon Asia. Results reveal that the algorithm is able to accurately detect the rice planting and harvest dates across two rice paddy field distribution maps with moderate-to-high spatial resolution, different validation data, and different rice cropping systems. The bias values of detected planting dates are 2, 0, and 4 days, while that of harvest dates are -2, -5, and -13 days at the prefecture, 0.5° gridcell, and site scales, respectively. These results highlight the potential of this algorithm to generate national, continental, or even global maps of rice phenology dynamics in an efficient manner, which can facilitate research on the impact of rice phenology on rice ecosystem services that echoes environmental and climate change.

LncRNA–disease association prediction method based on the nearest neighbor matrix completion model

State-of-the-art medical studies proved that long noncoding ribonucleic acids (lncRNAs) are closely related to various diseases. However, their large-scale detection in biological experiments is problematic and expensive. To aid screening and improve the efficiency of biological experiments, this study introduced a prediction model based on the nearest neighbor concept for lncRNA–disease association prediction. We used a new similarity algorithm in the model that fused potential associations. The experimental validation of the proposed algorithm proved its superiority over the available Cosine, Pearson, and Jaccard similarity algorithms. Satisfactory results in the comparative leave-one-out cross-validation test (with AUC = 0.96) confirmed its excellent predictive performance. Finally, the proposed model’s reliability was confirmed by performing predictions using a new dataset, yielding AUC = 0.92.

Improving soil organic matter estimation accuracy by combining optimal spectral preprocessing and feature selection methods based on pXRF and vis-NIR data fusion

Rapid and accurate estimation of soil organic matter (SOM) content is of great significance for agricultural production and carbon stock estimation. Visible-near-infrared spectroscopy (Vis-NIR) and portable X-ray fluorescence spectroscopy (pXRF), with the fast, low-cost, and environmental friendly characteristics, can realize the rapid large-scale detection of SOM. However, due to the heterogeneity and complexity of soil, estimation of SOM using a single sensor often leads to high uncertainty in result. Multi-sensor data fusion is one of the approaches to solve this problem, but suitable spectral preprocessing methods and data fusion methods still need further research. In this study, 500 soil samples were collected from two watersheds with large differences in soil parent material, soil type, and climate in the Xinjiang Uygur Autonomous Region of China. The effects of different spectral preprocessing methods for vis-NIR and pXRF spectra on SOM estimation were compared, and the effects of five feature selection methods on single sensor estimation and three different multi-sensor fusion estimations were evaluated. The results showed that vis-NIR spectrum was better than pXRF spectrum for SOM estimation, and the 1.6th-order derivative and baseline correction were the optimal preprocessing method for vis-NIR and pXRF spectra (Beam 3), respectively. Feature selection could improve the accuracy of the single sensor and multi-sensor fusion estimations. The competitive adaptive reweighted sampling (CARS) method was the optimal feature selection method for single sensor estimation and Granger–Ramanathan averaging (GRA) fusion estimation, and the variable importance in projection (VIP) was the optimal feature selection method for direct concatenation and outer-product analysis fusion estimation. In general, the optimal SOM estimation (RMSEp = 3.59 g kg−1, LCCCp = 0.84, RPIQp = 2.67) could be achieved based on CARS feature selection method and GRA fusion. This study proves that the combination of spectral preprocessing and feature selection can effectively improve the accuracy of the single sensor and multi-sensor fusion estimations. Especially, the multi-sensor fusion estimation has higher accuracy than the single sensor estimation. This study will provide a reference for the application of proximal soil sensing.

Validation of a Simple-to-Use, Affordable, Portable Wavefront Aberrometry-Based Auto Refractometer in a Paediatric Population.

InstaRef R20 is a handheld, affordable auto refractometer based on Shack Hartmann aberrometry technology. The study's objective was to compare InstaRef R20's performance for identifying refractive error in a paediatric population to that of standard subjective and objective refraction under both pre- and post-cycloplegic conditions. Refraction was performed using 1) standard clinical procedure consisting of retinoscopy followed by subjective refraction (SR) under pre- and post-cycloplegic conditions and 2) InstaRef R20. Agreement between both methods was evaluated using Bland-Altman analysis. The repeatability of the device based on three measurements in a subgroup of 20 children was assessed. The refractive error was measured in 132 children (mean age 12.31 ± 3 years). The spherical equivalent (M) and cylindrical components (J0 and J45) of the device had clinically acceptable differences (within ±0.50D) and acceptable agreement compared to standard pre- and post-cycloplegic manual retinoscopy and subjective refraction (SR). The device agreed within ± 0.50D of retinoscopy in 67% of eyes for M, 78% for J0 and 80% for J45 and within ± 0.50D of SR in 70% for M and 77% for cylindrical components. InstaRef R20 has an acceptable agreement compared to standard retinoscopy in paediatric population. The measurements from this device can be used as a starting point for subjective acceptance. The device being simple to use, portable, reliable and affordable has the potential for large-scale community-based refractive error detection.

Guest Editorial Introduction to the Special Issue on Advanced Machine Learning Methodologies for Large-Scale Video Object Segmentation and Detection

Video object segmentation and detection are two important tasks toward intelligent video content understanding. Due to their wide applications in real-world vision tasks, such as video surveillance and automatic driving, they have recently attracted great attention in the computer vision and multimedia processing communities. Although numerous deep learning-based approaches have been proposed to solve these problems, implementing effective and efficient video object segmentation and detection is still very challenging for now, and the principles of solutions to address the problems are still understudied. On the one hand, the features learned by the current deep models are not strong enough to capture the rich spatial and temporal information from the input videos. On the other hand, the annotation information in the video data (especially for unconstrained online videos) is usually insufficient, unspecific, or even absent, thus challenging the current mainstream learning schemes.

Farmland Soil Block Identification and Distribution Statistics Based on Deep Learning

Soil block distribution is one of the important indexes to evaluate the tillage performance of agricultural machinery. The traditional manual screening methods have the problems of low efficiency and damaging the original surface of the soil. This study proposes a statistical method of farmland soil block distribution based on deep learning. This method combines the adaptive learning rate and squeeze-and-excitation networks channel attention mechanism based on the original Mask-RCNN and uses the improved model to identify, segment and distribute statistics of the farmland soil blocks. Firstly, the influence of different learning rates and an improved Mask-RCNN algorithm model on training results were analyzed. Secondly, the effectiveness of the model in soil block identification and size measurement was analyzed. Finally, the identified soil blocks were classified accordingly, and the scale problem of soil block distribution after removing edge soil blocks was analyzed. The results show that with the decrease of learning rate, the loss value of model training decreases and the prediction accuracy of model is improved. The average precision value of the improved model increased by 25.29 %, and the recall value increased by 8.92%. The correlation coefficient of the maximum diameter measured by manual measurement and the maximum diameter measured by model algorithm was 0.99, which verifies the feasibility of the algorithm model. The prediction error of the model is the smallest when the camera height is 40 cm. Large-scale detection of soil block size in an experimental field in Hefei, Anhui, with an average confidence of over 97%. At the same time, the soil block is effectively classified according to the set classification standard. This study can provide an effective method for the accurate classification of soil block size and can provide a quantitative basis for the control of farmland cultivation intensity.

Analyses of widely targeted metabolic profiling reveals mechanisms of metabolomic variations during Tibetan sheep (Ovis aries) testis development

In mammals, the testis is the organ with the highest transcriptional activity. After gene transcription, translation, and post-translational protein modification, the transcriptional results are finally presented at the metabolic level. Metabolites not only essential for cell signaling and energy transfer, but also directly influenced by the physiological and pathological changes in tissues and accurately reflect the physiological changes. The fact that the testes are oxygen-deprived organs can explain why Sertoli cells and germ cells may use distinctive metabolic pathways to obtain energy in their different stages of development. Therefore, studying metabolic changes during testis development can better elucidate metabolic profile of the testis, which is essential to revealing characteristic metabolic pathways. The present study applied a widely targeted UPLC-MS/MS-based metabolomics approach with large-scale detection, identification and quantification to investigate the widespread metabolic changes during Tibetan sheep testis development. Firstly, a total of 847 metabolites were detected in the sheep testis, and their changes along with the three testis-development stages were further investigated. The results indicated that those metabolites were clustered into amino acids and their derivatives, carbohydrates and their derivatives, organic acids and their derivatives, benzene and substituted derivatives, alcohols and amines, lipids, nucleotides and their derivatives, bile acids, coenzymes and vitamins, hormones and hormone-related compounds, etc. Among them, the most abundant metabolites in the testis were amino acids and lipid metabolites. The results showed that most of the lipids, carbohydrates with their derivatives, as well as alcohol and amines metabolites were high in sexually immature sheep while organic acids, amino acids and nucleotides showed a continuously increasing trend along with testis development stages. Among them, the content of metabolites with antioxidant effects increased along with testis development, while those related with energy synthesis was downregulated with age. Further correlation analyses of each metabolite–metabolite pair emphasized the cross talk between differential metabolisms across testis development, suggesting a significant correlation between lipids and other metabolites. Finally, based on KEGG pathway analysis, we found that the metabolic pathways in Tibetan sheep testis development were mainly clustered into energy metabolism, gonadal development, and anti-oxidative stress. Reactive oxygen species (ROS) are by-products of normal cellular metabolism and are inevitable during testicular energy metabolism. Thus, the anti-oxidative stress function is a key process in maintaining the normal physiological function of testis. These results contributed to a broader view of the testis metabolome and a comprehensive analysis on metabolomic variation among different testis-development stages, providing a theoretical basis for us to understand the sheep testis metabolic mechanism.

Vertically Oriented Porous PET as Template to Integrated Metal Halide for High-Performance Large-Area and Ultra-Flexible X-Ray Detector.

High-performance X-ray detectors have immense potential in medical and security inspections. However, the current X-ray detectors are limited in flexible, high-spatial-resolution large-scale detection, and integration for imaging. Here, nuclear track-etched porous polyethylene terephthalate (PET) is developed as the template for preparing uniform, large-area (≥105 cm2 ), and flexible metal halide (MH)-based X-ray detectors. Adjustable high-density vertically oriented porous PET with adjustable thickness can provide proper physical support for flexible thick absorption film, thus improving X-ray absorption ability with excellent bending stability. Moreover, vertical channels can block the ion migration, lateral charge diffusion, and water/oxygen attacks, increasing activation energy for ionic transport, charge collection rate of electrodes, and environmental stability. Hence, the related detectors eventually obtain large sensitivity (6722 µC Gyair -1 cm-2 ), low detection limit (1.87 nGyair s-1 ), and high spatial resolution (5.17 lp mm-1 ) compared to the detectors without porous PET template. Meanwhile, the device shows no degradation after storage or working under various thermal attacks. MH-filled-PET is also monolithically integrated on the bottom circuit with different MHs and it is applied to single-pixel mode and fast linear-array imaging in a broad range of X-rays photon energy (20 to 160keV).

Small Aircraft Detection Based on Feature Enhancement and Context Information

In the field of civil aviation safety and air security, research on identification and tracking of small aircraft is very important. Aiming at the problem of low recognition rate of long-distance small air targets, this study proposes a method of GST-YOLOv3 combined with DeepSORT algorithm to detect and track small aircraft. The proposed GST-YOLOv3 uses a ghost module to extract effective features and simplify the network. In feature extraction layer, the spatial pyramid pooling layer is added to fuse local features and global features to enhance the features of small targets. The Transformer module is embedded in the feature fusion layer to focus on global context information. In the multiscale detection layer, a new large-scale feature map detection layer is formed by fusing shallow features and deep features to obtain a smaller receptive field. To avoid frame loss of video detection, the DeepSORT algorithm is used for real-time tracking. In this study, the self-made data set Aplanes and the public data set VisDroneDET are used to verify the effect of the proposed GST-YOLOv3. The computation of the GST-YOLOv3 was reduced by 45.8% compared with YOLOv3; the missed detection rate of airplanes decreased by 2.5%; and the detection precision, recall rate, F1 value, and average precision/0.5 value of small aircraft reached 98.4, 98.7, 98.5, and 98.6%, respectively, which are better than those of similar literature.