Automated Research Articles

Risk assessment and automatic identification of autistic children based on appearance

The diagnosis of Autism Spectrum Disorder (ASD) is mainly based on some diagnostic scales and evaluations by professional doctors, which may have limitations such as subjectivity, time, and cost. This research introduces a novel assessment and auto-identification approach for autistic children based on the appearance of children, which is a relatively objective, fast, and cost-effective approach. Initially, a custom social interaction scenario was developed, followed by a facial data set (ACFD) that contained 187 children, including 92 ASD and 95 children typically developing (TD). Using computer vision techniques, some appearance features of children including facial appearing time, eye concentration analysis, response time to name calls, and emotional expression ability were extracted. Subsequently, these features were combined and machine learning methods were used for the classification of children. Notably, the Bayes classifier achieved a remarkable accuracy of 94.1%. The experimental results show that the extracted visual appearance features can reflect the typical symptoms of children, and the automatic recognition method can provide an auxiliary diagnosis or data support for doctors.

Automatic detection of grains in partially recrystallized microstructures using deep learning

Precise identification of recrystallizing grains in partially recrystallized microstructures is essential to obtain quantitative information regarding the recrystallization process. Automatic, robust, user-friendly, and unbiased identification methods that do not rely on hard-coded, preselected values would be highly advantageous. In this study, we test convolutional neural network instance segmentation models to achieve automatic segmentation of individual recrystallizing grains in partially recrystallized microstructures. Our training dataset includes micrographs obtained using electron backscattered diffraction from five alloys with different thermal-mechanical histories and more than 100,000 recrystallizing grains. We adapt and train two state of the art deep learning models, namely Mask R-CNN and PointRend. Both models provide instance segmentation results of good quality, enabling quantitative determination of the microstructural parameters. The PointRend model demonstrates better performance for grains with irregular shapes than Mask R-CNN. Compared to conventional methods, the trained deep learning approach is easier to use, more flexible, and applicable to a wide range of materials.

A deep neural network model with spectral correlation function for electrocardiogram classification and diagnosis of atrial fibrillation

Atrial Fibrillation (AF) is a common type of irregular heartbeat, and early detection can significantly improve treatment outcomes and prognoses. Single-lead Electrocardiogram (ECG) devices are under extensive scrutiny for monitoring patients' heart health worldwide. Standardized ECG signal monitoring has demonstrated a significant reduction in mortality rates associated with severe cardiovascular diseases. However, the automatic detection method for AF requires significant improvement. This study presents a novel approach that utilizes the cyclostationary analysis of ECG signals, uncovering a spectral hidden periodicity between the QRS-T (the main wave components representing electrical activity in the heart) complexes of the ECG signal through the Spectral Correlation Function (SCF). To validate the proposed method's performance, the single ECG's SCF coefficients are applied to the Convolutional Recurrent Neural Network (CRNN), which consists of convolutional and long short-term memory (LSTM) layers, on the 2017 PhysioNet challenge dataset. The obtained results demonstrate that the proposed approach efficiently represents ECG signals through SCF coefficients, leading to the accurate detection of AF with an average accuracy of 92.76% and an average F1-score of 89.1%.

Domain adaptation based automatic identification method of vortex induced vibration of long-span bridges without prior information

Machine learning algorithms can sensitively capture the characteristics of vortex induced vibration (VIV) of the girder in long span bridge from the extensive historical data accumulated by structural health monitoring (SHM) system over several years. These algorithms have gradually become a promising method of VIV identification. However, the algorithms proposed by previous researchers require historical VIV data to select the threshold or parameters to identify VIV. Most long-span bridges have not recorded a significant amount of VIV data since VIV is rare, or the bridge were not equipped with SHM system before. This study proposes an adaptive VIV identification method based on domain adaptation methods, which can identify VIV in real-time or in historical monitoring datasets of the target bridge without prior VIV information or parameter settings. The strong generalization ability of the proposed method is verified on the SHM dataset of two long-span suspension bridges in China. It is found that the VIV recognition accuracy of the balanced distribution adaptation (BDA) based VIV identification method is higher than that of other algorithms. In this study, the BDA based algorithm is also applied to the 8 months monitoring datasets of a long span bridge and successfully identifies more than 20 VIV events of the main girder, which has shown the stability and accuracy of the proposed algorithm.

Efficient UAV-Based Automatic Classification of Cassava Fields Using K-Means and Spectral Trend Analysis

High-resolution images captured by Unmanned Aerial Vehicles (UAVs) play a vital role in precision agriculture, particularly in evaluating crop health and detecting weeds. However, the detailed pixel information in these images makes classification a time-consuming and resource-intensive process. Despite these challenges, UAV imagery is increasingly utilized for various agricultural classification tasks. This study introduces an automatic classification method designed to streamline the process, specifically targeting cassava plants, weeds, and soil classification. The approach combines K-means unsupervised classification with spectral trend-based labeling, significantly reducing the need for manual intervention. The method ensures reliable and accurate classification results by leveraging color indices derived from RGB data and applying mean-shift filtering parameters. Key findings reveal that the combination of the blue (B) channel, Visible Atmospherically Resistant Index (VARI), and color index (CI) with filtering parameters, including a spatial radius (sp) = 5 and a color radius (sr) = 10, effectively differentiates soil from vegetation. Notably, using the green (G) channel, excess red (ExR), and excess green (ExG) with filtering parameters (sp = 10, sr = 20) successfully distinguishes cassava from weeds. The classification maps generated by this method achieved high kappa coefficients of 0.96, with accuracy levels comparable to supervised methods like Random Forest classification. This technique offers significant reductions in processing time compared to traditional methods and does not require training data, making it adaptable to different cassava fields captured by various UAV-mounted optical sensors. Ultimately, the proposed classification process minimizes manual intervention by incorporating efficient pre-processing steps into the classification workflow, making it a valuable tool for precision agriculture.

Heterogeneous virus classification using a functional deep learning model based on transmission electron microscopy images

Viruses are submicroscopic agents that can infect other lifeforms and use their hosts’ cells to replicate themselves. Despite having simplistic genetic structures among all living beings, viruses are highly adaptable, resilient, and capable of causing severe complications in their hosts’ bodies. Due to their multiple transmission pathways, high contagion rate, and lethality, viruses pose the biggest biological threat both animal and plant species face. It is often challenging to promptly detect a virus in a host and accurately determine its type using manual examination techniques. However, computer-based automatic diagnosis methods, especially the ones using Transmission Electron Microscopy (TEM) images, have proven effective in instant virus identification. Using TEM images collected from a recent dataset, this article proposes a deep learning-based classification model to identify the virus type within those images. The methodology of this study includes two coherent image processing techniques to reduce the noise present in raw microscopy images and a functional Convolutional Neural Network (CNN) model for classification. Experimental results show that it can differentiate among 14 types of viruses with a maximum of 97.44% classification accuracy and F1-score, which asserts the effectiveness and reliability of the proposed method. Implementing this scheme will impart a fast and dependable virus identification scheme subsidiary to the thorough diagnostic procedures.

Can storage-discharge characteristics of karst matrix system quantified through recession analysis be reliable?

Storage and subsequent release of water in matrix system is a key function of karst catchments that controlling baseflow variation. Hydrograph recession analysis is currently the economic way to assess storage-discharge characteristics broadly at the catchment scale. However, there is large uncertainty in the related quantification due to recession data extraction. This study, by combining recursive digital filters with different automatic extraction methods, slow flow recession segments were extracted for hydrograph recession analysis and the further dynamic matrix storage (DMS) and related recession time (RT) assessment. The diversity and consistence among DMS and RT derived from different recession extraction methods (REMs) were analyzed, using hydrometric data in 20 catchments in the Wujiang river Basin in southwest China. Results indicate that the estimates of DMS and RT remarkedly varied between different REMs, however the order in which they ranked was mostly consistent. Moreover, the relationships between the derived DMS and RT with catchment physical features that potentially control storage and release processes are mostly consistent. Larger DMS are strongly associated with catchments featured as lower soil saturated hydraulic conductivity and smoother hydrographs. Longer RT are mostly related with drier catchments characterized as less variation of elevation and lower soil saturated hydraulic conductivity. This study highlights not only the uncertainty in quantifying storage and accompanied release characteristics, but also the reliability of storage-discharge characteristics quantified through recession analysis in terms of catchment comparison. Considering the diversity among catchments, ensemble of multi-method estimates of DMS and RT can enhance our understanding of the storage-discharge processes in the karst matrix system.

Building Energy Consumption Prediction and Optimization

As the proportion of building energy consumption in global energy consumption is significant and continues to rise, accurate prediction and optimization of building energy consumption are of great significance. This research aims to develop high-precision prediction models and innovative optimization strategies to address the problems of insufficient accuracy of existing prediction methods and poor optimization effects. By collecting and integrating data through multiple channels, a hybrid model architecture combining LSTM, CNN, and the attention mechanism is selected, and an automatic search and optimization method for hyperparameters is adopted. In terms of data processing, multi-source heterogeneous data fusion, data cleaning and enhancement based on spatiotemporal features, and innovative feature engineering are carried out. In the experiments, the dataset is divided considering the dynamic use of buildings and environmental changes, and indicators related to energy costs are introduced for evaluation. The research successfully constructed a model for accurately predicting building energy consumption, proposed effective optimization strategies, reduced energy consumption, and ensured comfort, providing a basis for practical applications through sensitivity analysis and robustness tests.

Automatic optimization combined with genetic algorithm and CFD for temperature uniformity of OLED glass substrate

The heating technology and temperature uniformity of glass substrate (GS) are crucial for the final display quality and efficiency of OLED. In a high-temperature and large heating environment with mutual influences among multiple variables, maintaining temperature difference of the entire GS below 3 °C presents a complex combinatorial optimization problem in industrial manufacturing. To address this challenge, an automatic combined method is established. This method involves parameterized modeling in UG, thermal calculation in CFD, and genetic algorithm optimization in Matlab to determine optimum designs for electric heating tube (EHT) heating temperature and interval with respect to objective functions of temperature uniformity and highest temperature. Through this automatic combined optimized scheme, the optimized heating temperature and EHT interval have been achieved to ensure maintaining a temperature difference of entire GS within 3 °C. All simulations are validated by experiments with an error below 10 %, especially below 1 % under multi EHTs. A heating scheme of 13 EHTs with a radius 6 mm, a distance 5 mm, a temperature 390 °C and an uniform interval 38.5 mm is suggested after final optimization. It is hoped that this study can advance heating technology of OLED GS to achieve the desired practical level.

Automated physics-based modeling of construction equipment through data fusion

Physics-based simulations are essential for designing autonomous construction equipment, but preparing models is time-consuming, requiring the integration of mechanical and geometric data. Current automatic modeling methods for modular robots are inadequate for construction equipment. This paper explores automating the modeling process by integrating mechanical data into 3D computer-aided design (CAD) models. A template library is developed with hierarchy and joint templates specific for equipment. During model generation, appropriate templates are selected based on the equipment type. Unspecified joint template data is extracted from technical specifications using a large language model (LLM). The 3D CAD model is then converted into a Universal Scene Description (USD) model. Users can adjust the part names and hierarchy within the USD model to align with the hierarchy template, and joint data is automatically integrated, resulting in a simulation-ready model. This method reduces modeling time by over 87 % compared to manual methods, while maintaining accuracy.

Automatic patient centering in computed tomography: a systematic review and meta-analysis.

To comprehensively examine the influence of auto-patient centering technologies on positioning accuracy, radiation dose, image quality, and time efficiency of computed tomography (CT) scans. A systematic search of peer-reviewed English publications was performed between January 2000 and November 2023 in PubMed, Embase, CINAHL, Scopus, and Web of Science. Two postgraduate students and an academic lecturer independently reviewed the articles to verify adherence to the inclusion criteria. The QUADAS-2 tool was employed to evaluate study quality. We derived summary estimates on positioning accuracy, radiation dose reduction, image quality, and time efficiency using proportion and meta-analysis methodologies. Nine studies were identified comparing automatic and manual CT positioning. Automatic positioning improved accuracy by reducing vertical offsets to 7 mm and 4 mm for thorax and abdominal CTs, compared to 19 mm and 18 mm with manual methods. Most studies showed significant reductions in radiation dose, ranging from 5.71 to 31%. Image quality results were mixed, automatic methods generally produced images with less noise, but differences were minimal. Time efficiency was better, with automatic positioning reducing preparation time from 0.48 min versus 0.67 min for manual positioning. This review confirms that automatic patient-centering technologies enhance positioning accuracy and decrease preparation times in CT scans. While reductions in radiation doses and some improvements in image quality were observed, the evidence remains mixed. Findings support integrating these technologies into clinical practice to optimize patient care. Question Does automatic patient centering in CT enhance positioning accuracy, reduce radiation exposure, and improve image quality? Findings Findings indicate that automatic centering can optimize image quality, reduce examination times and contribute to overall improvements in imaging efficiency. Clinical relevance Automatic patient centering in CT improves positioning accuracy, minimizes radiation exposure, enhances image quality, and accelerates imaging workflows, contributing to safer, more efficient imaging procedures that benefit patient care.

Automatic Detection and Tracking of Objects of Interest in Video Data with Global Motion

Introduction. At present, automatic capture and tracking of moving objects in video data obtained by a video camera mounted on a mobile carrier represents a relevant research task. Its successful solution is challenged by such factors, as a non-uniform background, object overlapping between one another and the background, significant and rapid changes in the size of the object of interest, abrupt changes in the movement trajectory of the mobile carrier.Aim. To develop an automatic method for detecting moving objects followed by their tracking in video data obtained under difficult observation conditions. An additional requirement imposed on the tracking stage consists in the restriction of computing resources.Materials and methods. The method is based on a convolutional neural network with a YOLO architecture. Due to the restriction of computing resources, object tracking is implemented without neural network solutions. In order to ensure stable tracking, several detectors are used simultaneously with the subsequent analysis of the data obtained. The tracking stage involves a detector based on histograms of oriented gradients (HOG), supplemented by a detector based on correlation filtering and motion trajectory prediction based on the Kalman filter.Results. At the automatic detection stage, the TPR, averaged over all video files participating in the experiments, was equal to 0.81, with the FPR corresponding to 0.10. At the tracking stage, the failure rate (tracking failures) was 6·10 –5.Conclusion. The proposed method can be successfully used to detect and track objects at a distance of 1500 m with an object projection size on the frame of 5 × 5 pixels under the conditions of global motion, a non-uniform background, and significant changes in the properties of the object of interest.

Radiomic features based automatic classification of CT lung findings for COVID-19 patients

Introduction. The lung CT images of COVID-19 patients can be typically characterized by three different findings- Ground Glass Opacity (GGO), consolidation and pleural effusion. GGOs have been shown to precede consolidations and has different heterogeneous appearance. Conventional severity scoring only uses total area of lung involvement ignoring appearance of the effected regions. This study proposes a baseline to select heterogeneity/radiomic features that can distinguish these three pathological lung findings.Methods. Four approaches were implemented to select features from a pool of 44 features. First one is a manual feature selection method. The rest are automatic feature selection methods based on Genetic Algorithm (GA) coupled with (1) K-Nearest-Neighbor (GA-KNN), (2) binary-decision-tree (GA-BDT) and (3) Artificial-Neural-Network (GA-ANN). For the purpose of validation, an ANN was trained using the selected features and tested on a completely independent data set.Results. Manual selection of nine radiomic features was found to provide the most accurate results with the highest sensitivity, specificity and accuracy (85.7% overall accuracy and 0.90 area under receiver operating characteristic curve) followed by GA-BDT, GA-KNN and GA-ANN (accuracy 78%, 77.5% and 76.8%).Conclusion. Manually selected nine radiomic features can be used in accurate severity scoring allowing the clinician to plan for more effective personalized treatment. They can also be useful for monitoring the progression of COVID-19 and response to therapy for clinical trials.

AECR: Automatic attack technique intelligence extraction based on fine-tuned large language model

Cyber Threat Intelligence (CTI) reports contain resourceful intelligence on cyber-attack campaigns, which provides great help for security analysts to infer attack trends and enhance their defenses. However, due to the diversity of report content and writing styles, current intelligence extraction is mostly based on time-consuming manual efforts. Moreover, existing automatic methods generally neglect the importance of background knowledge and produce inexact extraction results. These problems prevent the effective utilization and sharing of intelligence from CTI reports. In this paper, we primarily focus on the automatic extraction of attack technique (AT) intelligence, which reveals patterns of attack behaviors and hardly changes over time. We propose a novel automatic AT extraction pipeline for CTI reports (AECR). AECR explores the feasibility of extracting AT intelligence based on a fined-tuned large language model (LLM). Particularly, we endow the selected LLM with enhanced domain-specific knowledge to improve its comprehension of AT-relevant content and alleviate the hallucination problem. Experimental results demonstrate that AECR outperforms state-of-the-art methods by a wide margin with a reasonable time cost. Specifically, we improve the accuracy, precision, recall, and F1-score by 108%, 37.2%, 22.4%, and 67.5% respectively. To the best of our knowledge, AECR is the first to perform AT extraction based on fine-tuned LLM.

Minimally interactive segmentation of soft-tissue tumors on CT and MRI using deep learning.

Segmentations are crucial in medical imaging for morphological, volumetric, and radiomics biomarkers. Manual segmentation is accurate but not feasible in clinical workflow, while automatic segmentation generally performs sub-par. To develop a minimally interactive deep learning-based segmentation method for soft-tissue tumors (STTs) on CT and MRI. The interactive method requires the user to click six points near the tumor's extreme boundaries in the image. These six points are transformed into a distance map and serve, with the image, as input for a convolutional neural network. A multi-center public dataset with 514 patients and nine STT phenotypes in seven anatomical locations, with CT or T1-weighted MRI, was used for training and internal validation. For external validation, another public dataset was employed, which included five unseen STT phenotypes in extremities on CT, T1-weighted MRI, and T2-weighted fat-saturated (FS) MRI. Internal validation resulted in a dice similarity coefficient (DSC) of 0.85 ± 0.11 (mean ± standard deviation) for CT and 0.84 ± 0.12 for T1-weighted MRI. External validation resulted in DSCs of 0.81 ± 0.08 for CT, 0.84 ± 0.09 for T1-weighted MRI, and 0.88 ± 0.08 for T2-weighted FS MRI. Volumetric measurements showed consistent replication with low error internally (volume: 1 ± 28 mm3, r = 0.99; diameter: - 6 ± 14 mm, r = 0.90) and externally (volume: - 7 ± 23 mm3, r = 0.96; diameter: - 3 ± 6 mm, r = 0.99). Interactive segmentation time was considerably shorter (CT: 364 s, T1-weighted MRI: 258s) than manual segmentation (CT: 1639s, T1-weighted MRI: 1895s). The minimally interactive segmentation method effectively segments STT phenotypes on CT and MRI, with robust generalization to unseen phenotypes and imaging modalities. Question Can this deep learning-based method segment soft-tissue tumors faster than can be done manually and more accurately than other automatic methods? Findings The minimally interactive segmentation method achieved accurate segmentation results in internal and external validation, and generalized well across soft-tissue tumor phenotypes and imaging modalities. Clinical relevance This minimally interactive deep learning-based segmentation method could reduce the burden of manual segmentation, facilitate the integration of imaging-based biomarkers (e.g., radiomics) into clinical practice, and provide a fast, semi-automatic solution for volume and diameter measurements (e.g., RECIST).

Precision and Efficiency in Dam Crack Inspection: A Lightweight Object Detection Method Based on Joint Distillation for Unmanned Aerial Vehicles (UAVs)

Dams in their natural environment will gradually develop cracks and other forms of damage. If not detected and repaired in time, the structural strength of the dam may be reduced, and it may even collapse. Repairing cracks and defects in dams is very important to ensure their normal operation. Traditional detection methods rely on manual inspection, which consumes a lot of time and labor, while deep learning methods can greatly alleviate this problem. However, previous studies have often focused on how to better detect crack defects, with the corresponding image resolution not being particularly high. In this study, targeting the scenario of real-time detection by drones, we propose an automatic detection method for dam crack targets directly on high-resolution remote sensing images. First, for high-resolution remote sensing images, we designed a sliding window processing method and proposed corresponding methods to eliminate redundant detection frames. Then, we introduced a Gaussian distribution in the loss function to calculate the similarity of predicted frames and incorporated a self-attention mechanism in the spatial pooling module to further enhance the detection performance of crack targets at various scales. Finally, we proposed a pruning-after-distillation scheme, using the compressed model as the student and the pre-compression model as the teacher and proposed a joint distillation method that allows more efficient distillation under this compression relationship between teacher and student models. Ultimately, a high-performance target detection model can be deployed in a more lightweight form for field operations such as UAV patrols. Experimental results show that our method achieves an mAP of 80.4%, with a parameter count of only 0.725 M, providing strong support for future tasks such as UAV field inspections.

Метод автоматичного виділення прямокутної зони об'єктів зображення

T The automatic selection method for rectangle bounding of image objrcts exposed on the film is described, to optimize the image's size and increase the decoding efficiency. The image processing algorithm has been developed to improve the efficiency of grouping individual elements into a single structure. The simulation was carried out to prove the effectiveness of the proposed method. The main tool that has been used for developing the application is the open-source library of functions and algorithms of computer vision, image processing, and numerical algorithms of general-purpose OpenCV.

Research on Automatic Alignment for Corn Harvesting Based on Euclidean Clustering and K-Means Clustering

Aiming to meet the growing need for automated harvesting, an automatic alignment method based on Euclidean clustering and K-means clustering is proposed to address issues of driver fatigue and inaccurate driving in manually operated corn harvesters. Initially, the corn field environment is scanned using LiDAR to obtain point cloud data, which are then subjected to pass-through filtering and statistical filtering to remove noise and non-corn contour points. Subsequently, Euclidean clustering and K-means clustering methods are applied to the filtered point cloud data. To validate the impact of Euclidean clustering on subsequent clustering, two separate treatments of the obtained point cloud data were conducted during experimental validation: the first used the K-means clustering algorithm directly, while the second involved performing Euclidean clustering followed by K-means clustering. The results demonstrate that the combined method of Euclidean clustering and K-means clustering achieved a success rate of 81.5%, representing a 26.5% improvement over traditional K-means clustering. Additionally, the Rand index increased by 0.575, while accuracy improved by 57% and recall increased by 61%.

UniProt: the Universal Protein Knowledgebase in 2025

Abstract The aim of the UniProt Knowledgebase (UniProtKB; https://www.uniprot.org/) is to provide users with a comprehensive, high-quality and freely accessible set of protein sequences annotated with functional information. In this publication, we describe ongoing changes to our production pipeline to limit the sequences available in UniProtKB to high-quality, non-redundant reference proteomes. We continue to manually curate the scientific literature to add the latest functional data and use machine learning techniques. We also encourage community curation to ensure key publications are not missed. We provide an update on the automatic annotation methods used by UniProtKB to predict information for unreviewed entries describing unstudied proteins. Finally, updates to the UniProt website are described, including a new tab linking protein to genomic information. In recognition of its value to the scientific community, the UniProt database has been awarded Global Core Biodata Resource status.

Automated Gas Influx Handling Model and Mechanisms During Deep High-Temperature and High-Pressure Well Drilling

The exploration and development of oil and gas resources in deep formations is a key strategic priority for national energy production. However, manual methods for handling gas kicks suffer from low operating accuracy and inefficiency during high-temperature and high-pressure deep well drilling. To address the need for real-time bottomhole pressure prediction and control, an efficient gas–liquid–solid computing model was developed based on the gas slip model and cuttings settling velocity model. By integrating this model with an automatic choke adjustment system, an automatic gas kick attenuation model for deep well drilling was established. Results show that, compared to the driller’s and wait-and-weight methods, the automatic gas kick attenuation method significantly reduces peak choke pressure due to its larger frictional pressure drop and higher cuttings hydrostatic pressure. The automatic attenuation method not only leads to an average reduction of 28.42% in maximum choke/casing pressure but also accelerates gas removal, achieving gas kick attenuation ten times faster than the driller’s method and seven times faster than the wait-and-weight method. The study also investigates the influence of gas solubility, well depth, gas influx volume, formation permeability, and drilling fluid volumetric flow rate on gas kick attenuation characteristics. The findings provide a solid foundation for improving the efficiency of gas kick management in deep well drilling operations.