Search

In view of the inevitable error problems caused by peak extraction, calibration curve fitting and other essential operations in the traditional data processing of chromatic confocal microscopy (CCM), a regression model based on convolutional neural network (CNN) is proposed so that the above necessary operations are no longer required. This CNN-based regression model draws on the core concepts of the AlexNet model and has been moderately customized and optimized to make it more suitable for CCM application scenarios. The proposed method has been v alidated using a completely homemade CCM apparatus The experimental results showed that the CNN-based method is feasible for the CCM measurement and exhibits better stability and higher axial resolution than traditional methods, indicating that deep learning has good application value in the data processing of CCM. • A new chromatic confocal sensing method based on artificial intelligence is proposed. A model based on the deep convolutional neural network (CNN) has been established. • With the new method, traditional data processing methods, i.e., the peak extraction algorithm and the calibration procedure, required in chromatic confocal microscopy are no longer needed. • An experimental prototype has been built, and detailed experimental tests for resolution, accuracy, and 3D topography were carried out. • The measurement results proved that the CNN-based model can improve the axial resolution and measurement precision. [ABSTRACT FROM AUTHOR]

Recent technological developments have enabled computers to identify and categorize facial expressions to determine a person's emotional state in an image or a video. This process, called "Facial Expression Recognition (FER)", has become one of the most popular research areas in computer vision. In recent times, deep FER systems have primarily concentrated on addressing two significant challenges: the problem of overfitting due to limited training data availability, and the presence of expression-unrelated variations, including illumination, head pose, image resolution, and identity bias. In this paper, a comprehensive survey is provided on deep FER, encompassing algorithms and datasets that offer insights into these intrinsic problems. Initially, this paper presents a detailed timeline showcasing the evolution of methods and datasets in deep facial expression recognition (FER). This timeline illustrates the progression and development of the techniques and data resources used in FER. Then, a comprehensive review of FER methods is introduced, including the basic principles of FER (components such as preprocessing, feature extraction and classification, and methods, etc.) from the pro-deep learning era (traditional methods using handcrafted features, i.e., SVM and HOG, etc.) to the deep learning era. Moreover, a brief introduction is provided related to the benchmark datasets (there are two categories: controlled environments (lab) and uncontrolled environments (in the wild)) used to evaluate different FER methods and a comparison of different FER models. Existing deep neural networks and related training strategies designed for FER, based on static images and dynamic image sequences, are discussed. The remaining challenges and corresponding opportunities in FER and the future directions for designing robust deep FER systems are also pinpointed. [ABSTRACT FROM AUTHOR]

Identifying quartz, feldspar and cuttings in sandstone is of great significance for judging the dispositive environment, but traditional manual identification methods have problems of strong subjectivity and high dependence on experience. An image recognition method of sandstone micro-components based on Faster R-CNN targeted detection algorithm was constructed by using deep learning, convolutional neural network and other technologies. Intelligent recognition of three components, the average recognition accuracy of the three components is 89. 28% . In order to verify the reliability of the model, the experiments compared different algorithms and feature extraction networks. The results show that the recognition effect of the Faster R-CNN targets detection algorithm is better than that of YOLOv3, YOLOv4, and YOLOv5s, and the performance of the ResNet50 features extraction network is better than VGG16. The Faster R-CNN targets detection model using the ResNet50 features extraction network has significant advantages, it can better meet the identification requirements of rock slices, and provide intelligent technical solutions to traditional manual methods. [ABSTRACT FROM AUTHOR]

In the evolving landscape of Internet of Things (IoT) and Industrial IoT (IIoT) security, novel and efficient intrusion detection systems (IDSs) are paramount. In this article, we present a groundbreaking approach to intrusion detection for IoT-based electric vehicle charging stations (EVCS), integrating the robust capabilities of convolutional neural network (CNN), long short-term memory (LSTM), and gated recurrent unit (GRU) models. The proposed framework leverages a comprehensive real-world cybersecurity dataset, specifically tailored for IoT and IIoT applications, to address the intricate challenges faced by IoT-based EVCS. We conducted extensive testing in both binary and multiclass scenarios. The results are remarkable, demonstrating a perfect 100% accuracy in binary classification, an impressive 97.44% accuracy in six-class classification, and 96.90% accuracy in fifteen-class classification, setting new benchmarks in the field. These achievements underscore the efficacy of the CNN-LSTM-GRU ensemble architecture in creating a resilient and adaptive IDS for IoT infrastructures. The ensemble algorithm, accessible via GitHub, represents a significant stride in fortifying IoT-based EVCS against a diverse array of cybersecurity threats. [ABSTRACT FROM AUTHOR]

In this work, an exhaustive analysis of the partial discharges that originate in the bubbles present in dielectric mineral oils is carried out. To achieve this, a low-cost, high-resolution CMOS image sensor is used. Partial discharge measurements using that image sensor are validated by a standard electrical detection system that uses a discharge capacitor. In order to accurately identify the images corresponding to partial discharges, a convolutional neural network is trained using a large set of images captured by the image sensor. An image classification model is also developed using deep learning with a convolutional network based on a TensorFlow and Keras model. The classification results of the experiments show that the accuracy achieved by our model is around 95% on the validation set and 82% on the test set. As a result of this work, a non-destructive diagnosis method has been developed that is based on the use of an image sensor and the design of a convolutional neural network. This approach allows us to obtain information about the state of mineral oils before breakdown occurs, providing a valuable tool for the evaluation and maintenance of these dielectric oils. [ABSTRACT FROM AUTHOR]

Obstructive sleep apnea (OSA), a prevalent sleep disorder, is intimately associated with various other diseases, particularly cardiovascular conditions. The conventional diagnostic method, nocturnal polysomnography (PSG), despite its widespread use, faces challenges due to its high cost and prolonged duration. Recent developments in electrocardiogram-based diagnostic techniques have opened new avenues for addressing these challenges, although they often require a deep understanding of feature engineering. In this study, we introduce an innovative method for OSA classification that combines a composite deep convolutional neural network model with a multimodal strategy for automatic feature extraction. This approach involves transforming the original dataset into scalogram images that reflect heart rate variability attributes and Gramian angular field matrix images that reveal temporal characteristics, aiming to enhance the diversity and richness of data features. The model comprises automatic feature extraction and feature enhancement components and has been trained and validated on the PhysioNet Apnea-ECG database. The experimental results demonstrate the model's exceptional performance in diagnosing OSA, achieving an accuracy of 96.37%, a sensitivity of 94.67%, a specificity of 97.44%, and an AUC of 0.96. These outcomes underscore the potential of our proposed model as an efficient, accurate, and convenient tool for OSA diagnosis. [ABSTRACT FROM AUTHOR]

In Internet of Things-based smart grids, smart meters record and report a massive number of power consumption data at certain intervals to the data center of the utility for load monitoring and energy management. Energy theft is a big problem for smart meters and causes non-technical losses. Energy theft attacks can be launched by malicious consumers by compromising the smart meters to report manipulated consumption data for less billing. It is a global issue causing technical and financial damage to governments and operators. Deep learning-based techniques can effectively identify consumers involved in energy theft through power consumption data. In this study, a hybrid convolutional neural network (CNN)-based energy-theft-detection system is proposed to detect data-tampering cyber-attack vectors. CNN is a commonly employed method that automates the extraction of features and the classification process. We employed CNN for feature extraction and traditional machine learning algorithms for classification. In this work, honest data were obtained from a real dataset. Six attack vectors causing data tampering were utilized. Tampered data were synthetically generated through these attack vectors. Six separate datasets were created for each attack vector to design a specialized detector tailored for that specific attack. Additionally, a dataset containing all attack vectors was also generated for the purpose of designing a general detector. Furthermore, the imbalanced dataset problem was addressed through the application of the generative adversarial network (GAN) method. GAN was chosen due to its ability to generate new data closely resembling real data, and its application in this field has not been extensively explored. The data generated with GAN ensured better training for the hybrid CNN-based detector on honest and malicious consumption patterns. Finally, the results indicate that the proposed general detector could classify both honest and malicious users with satisfactory accuracy. [ABSTRACT FROM AUTHOR]

Real-time and high-precision land cover classification is the foundation for efficient and quantitative research on grassland degradation using remote sensing techniques. In view of the shortcomings of manual surveying and satellite remote sensing, this study focuses on the identification and classification of grass species indicating grassland degradation. We constructed a UAV-based hyperspectral remote sensing system and collected field data in grassland areas. By applying artificial intelligence technology, we developed a 3D_RNet-O model based on convolutional neural networks, effectively addressing technical challenges in hyperspectral remote sensing identification and classification of grassland degradation indicators, such as low reflectance of vegetation, flat spectral curves, and sparse distribution. The results showed that the model achieved a classification accuracy of 99.05% by optimizing hyperparameter combinations based on improving residual block structures. The establishment of the UAV-based hyperspectral remote sensing system and the proposed 3D_RNet-O classification model provide possibilities for further research on low-altitude hyperspectral remote sensing in grassland ecology. [ABSTRACT FROM AUTHOR]

Background: Lysine crotonylation (Kcr) is a crucial protein post-translational modification found in histone and non-histone proteins. It plays a pivotal role in regulating diverse biological processes in both animals and plants, including gene transcription and replication, cell metabolism and differentiation, as well as photosynthesis. Despite the significance of Kcr, detection of Kcr sites through biological experiments is often time-consuming, expensive, and only a fraction of crotonylated peptides can be identified. This reality highlights the need for efficient and rapid prediction of Kcr sites through computational methods. Currently, several machine learning models exist for predicting Kcr sites in humans, yet models tailored for plants are rare. Furthermore, no downloadable Kcr site predictors or datasets have been developed specifically for plants. To address this gap, it is imperative to integrate existing Kcr sites detected in plant experiments and establish a dedicated computational model for plants. Results: Most plant Kcr sites are located on non-histones. In this study, we collected non-histone Kcr sites from five plants, including wheat, tabacum, rice, peanut, and papaya. We then conducted a comprehensive analysis of the amino acid distribution surrounding these sites. To develop a predictive model for plant non-histone Kcr sites, we combined a convolutional neural network (CNN), a bidirectional long short-term memory network (BiLSTM), and attention mechanism to build a deep learning model called PlantNh-Kcr. On both five-fold cross-validation and independent tests, PlantNh-Kcr outperformed multiple conventional machine learning models and other deep learning models. Furthermore, we conducted an analysis of species-specific effect on the PlantNh-Kcr model and found that a general model trained using data from multiple species outperforms species-specific models. Conclusion: PlantNh-Kcr represents a valuable tool for predicting plant non-histone Kcr sites. We expect that this model will aid in addressing key challenges and tasks in the study of plant crotonylation sites. [ABSTRACT FROM AUTHOR]

Brain tumors can have fatal consequences, affecting many body functions. For this reason, it is essential to detect brain tumor types accurately and at an early stage to start the appropriate treatment process. Although convolutional neural networks (CNNs) are widely used in disease detection from medical images, they face the problem of overfitting in the training phase on limited labeled and insufficiently diverse datasets. The existing studies use transfer learning and ensemble models to overcome these problems. When the existing studies are examined, it is evident that there is a lack of models and weight ratios that will be used with the ensemble technique. With the framework proposed in this study, several CNN models with different architectures are trained with transfer learning and fine-tuning on three brain tumor datasets. A particle swarm optimization-based algorithm determined the optimum weights for combining the five most successful CNN models with the ensemble technique. The results across three datasets are as follows: Dataset 1, 99.35% accuracy and 99.20 F1-score; Dataset 2, 98.77% accuracy and 98.92 F1-score; and Dataset 3, 99.92% accuracy and 99.92 F1-score. We achieved successful performances on three brain tumor datasets, showing that the proposed framework is reliable in classification. As a result, the proposed framework outperforms existing studies, offering clinicians enhanced decision-making support through its high-accuracy classification performance. [ABSTRACT FROM AUTHOR]

Alzheimer's disease (AD) is a debilitating neurodegenerative disorder that requires accurate diagnosis for effective management and treatment. In this article, we propose an architecture for a convolutional neural network (CNN) that utilizes magnetic resonance imaging (MRI) data from the Alzheimer's disease Neuroimaging Initiative (ADNI) dataset to categorize AD. The network employs two separate CNN models, each with distinct filter sizes and pooling layers, which are concatenated in a classification layer. The multi-class problem is addressed across three, four, and five categories. The proposed CNN architecture achieves exceptional accuracies of 99.43%, 99.57%, and 99.13%, respectively. These high accuracies demonstrate the efficacy of the network in capturing and discerning relevant features from MRI images, enabling precise classification of AD subtypes and stages. The network architecture leverages the hierarchical nature of convolutional layers, pooling layers, and fully connected layers to extract both local and global patterns from the data, facilitating accurate discrimination between different AD categories. Accurate classification of AD carries significant clinical implications, including early detection, personalized treatment planning, disease monitoring, and prognostic assessment. The reported accuracy underscores the potential of the proposed CNN architecture to assist medical professionals and researchers in making precise and informed judgments regarding AD patients. [ABSTRACT FROM AUTHOR]

AbstractObstructive sleep apnea (OSA) is associated with various health complications, and snoring is a prominent characteristic of this disorder. Therefore, the exploration of a concise and effective method for detecting snoring has consistently been a crucial aspect of sleep medicine. As the easily accessible data, the identification of snoring through sound analysis offers a more convenient and straightforward method. The objective of this study was to develop a convolutional neural network (CNN) for classifying snoring and non-snoring events based on audio. This study utilized Mel-frequency cepstral coefficients (MFCCs) as a method for extracting features during the preprocessing of raw data. In order to extract multi-scale features from the frequency domain of sound sources, this study proposes the utilization of a multi-branch convolutional neural network (MBCNN) for the purpose of classification. The network utilized asymmetric convolutional kernels to acquire additional information, while the adoption of one-hot encoding labels aimed to mitigate the impact of labels. The experiment tested the network’s performance by utilizing a publicly available dataset consisting of 1,000 sound samples. The test results indicate that the MBCNN achieved a snoring detection accuracy of 99.5%. The integration of multi-scale features and the implementation of MBCNN, based on audio data, have demonstrated a substantial improvement in the performance of snoring classification. [ABSTRACT FROM AUTHOR]

BACKGROUND: In the contemporary food industry, accurate and rapid differentiation of oolong tea varieties holds paramount importance for traceability and quality control. However, achieving this remains a formidable challenge. This study addresses this lacuna by employing machine learning algorithms – namely support vector machines (SVMs) and convolutional neural networks (CNNs) – alongside computer vision techniques for the automated classification of oolong tea leaves based on visual attributes. RESULTS: An array of 13 distinct characteristics, encompassing color and texture, were identified from five unique oolong tea varieties. To fortify the robustness of the predictive models, data augmentation and image cropping methods were employed. A comparative analysis of SVM‐ and CNN‐based models revealed that the ResNet50 model achieved a high Top‐1 accuracy rate exceeding 93%. This robust performance substantiates the efficacy of the implemented methodology for rapid and precise oolong tea classification. CONCLUSION: The study elucidates that the integration of computer vision with machine learning algorithms constitutes a promising, non‐invasive approach for the quick and accurate categorization of oolong tea varieties. The findings have significant ramifications for process monitoring, quality assurance, authenticity validation and adulteration detection within the tea industry. © 2023 Society of Chemical Industry. [ABSTRACT FROM AUTHOR]

Purpose: MRI-based radiomics models can predict genetic markers in pediatric low-grade glioma (pLGG). These models usually require tumour segmentation, which is tedious and time consuming if done manually. We propose a deep learning (DL) model to automate tumour segmentation and build an end-to-end radiomics-based pipeline for pLGG classification. Methods: The proposed architecture is a 2-step U-Net based DL network. The first U-Net is trained on downsampled images to locate the tumour. The second U-Net is trained using image patches centred around the located tumour to produce more refined segmentations. The segmented tumour is then fed into a radiomics-based model to predict the genetic marker of the tumour. Results: Our segmentation model achieved a correlation value of over 80% for all volume-related radiomic features and an average Dice score of.795 in test cases. Feeding the auto-segmentation results into a radiomics model resulted in a mean area under the ROC curve (AUC) of.843, with 95% confidence interval (CI) [.78-.906] and.730, with 95% CI [.671-.789] on the test set for 2-class (BRAF V600E mutation BRAF fusion) and 3-class (BRAF V600E mutation BRAF fusion and Other) classification, respectively. This result was comparable to the AUC of.874, 95% CI [.829-.919] and.758, 95% CI [.724-.792] for the radiomics model trained and tested on the manual segmentations in 2-class and 3-class classification scenarios, respectively. Conclusion: The proposed end-to-end pipeline for pLGG segmentation and classification produced results comparable to manual segmentation when it was used for a radiomics-based genetic marker prediction model. [ABSTRACT FROM AUTHOR]

Lung cancer is a deadly type of malignancy that poses a significant threat to human health. Accurately identifying the subtypes of lung cancer is critical for effective treatment. However, conventional methods for determining subtypes, such as histological examination, are invasive and time-consuming. In order to overcome this problem, a non-invasive approach for predicting lung cancer subtypes using multi-modality images with a hybrid model is proposed in this study. The model combines attention-based Convolutional Neural Networks (CNNs) and machine learning classifiers to achieve this objective. The model employs a soft-attention mechanism to focus on the pathological areas and extract both global and local features from the images. The stack-based ensemble classifier employs logistic regression as a meta-learner and four machine learning classifiers, including Support Vector Machine (SVM), Naive Bayes, Random Forest, and J48. The classifier categorizes lung cancer into Adenocarcinoma (ADC), Squamous Cell Carcinoma (SQC), and Small Cell Carcinoma (SCC). The suggested model validated using publically accessible datasets (Lung-PET-CT-DX, Lung1, and Lung3) achieved superior performance, with a validation accuracy of 98.8%, an F1-score of 0.986, and a Matthews Correlation Coefficient (MCC) of 0.988. [ABSTRACT FROM AUTHOR]

Steganography refers to hiding a secret message from various sources, such as images, videos, audio and so on. The advantage of steganography is to avoid data hacking in transmission medium during the transmission of information sources. Video steganography is superior to image steganography since the videos can hide a substantial quantity of secret messages more than the image. Hence, this research introduced the video stereography technique, Arnold Transform with SqueezeNet-based Pelican Whale Optimization Algorithm (AT + SqueezeNet_PWOA), for concealing the secret image on the video. To hide the secret image on the video, the proposed method follows three steps: key frame and feature extraction, pixel prediction and embedding. The extraction of the key frame process is carried out by the Structural Similarity Index Measure (SSIM), and then the neighborhood features and convolutional neural network (CNN) features are extracted from the frame to improve the robustness of the embedding process. Moreover, the pixel prediction is completed by the SqueezeNet model, wherein the learning factors are tuned by the PWOA. In addition, the embedding process is completed by applying the Arnold transform on the predicted pixel, and the transformed regions are combined with the secret image using the embedding function. Likewise, the extraction process extracts the secret image from the embedded video by substituting the predicted pixel and Arnold transform on the embedded video. The proposed method is used to hide chunks of secret data in the form of video sequences and it improves the performance. The Arnold transform used in this work provides security by encrypting the data. The use of SqueezeNet makes the proposed model a simple design and this reduces the computational time. Thus, the AT + SqeezeNet_PWOA attained better correlation coefficient (CC), peak signal-to-noise ratio (PSNR) and mean square error (MSE) of 0.908, 48.66 and 0.001 dB with the Gaussian noise. [ABSTRACT FROM AUTHOR]

This paper proposes an efficient method for rectifying distorted document images via deep learning, ultimately improving the legibility of graphics and text in documents. The framework comprises two interconnected UNets, working in tandem to predict a 3D coordinate map and a forward map for the input distorted document image, respectively. At the beginning of the process, a page mask is predicted and used as input to both U-Nets to help improve the performance of their tasks. In the last step, the predicted forward map is transformed into a corresponding backward map, which is utilized to rectify the distorted image. The experimental results not only reveal that the predicted page masks and 3D coordinate maps significantly enhance the accuracy of predicting forward maps for subsequent rectification but also demonstrate satisfactory results both globally and locally. [ABSTRACT FROM AUTHOR]

Since the objects in remote sensing images have the characteristics of arbitrary direction, dense distribution and large scale differences, object detection in remote sensing images has become a challenging problem. Aiming at this problem, this paper systematically reviewed the work related to rotating object detection in deep learning remote sensing images in recent three years. Firstly, this paper introduced the representation method and characteristics of rotated bounding box. Then, it analyzed the existing methods of rotating object detection in remote sensing images from four aspects: the feature extraction network, the generation of rotating anchor and candidate bounding box, the label allocation and sampling strategy, and the loss function. Next, it introduced the commonly used remote sensing image datasets for rotated object detection, and compared the performance of different algorithms. Finally, this paper prospected rotating object detection in remote sensing images. [ABSTRACT FROM AUTHOR]

Recent advances in high-throughput electron microscopy (EM) have revolutionized the examination of microstructures by enabling fast EM image generation. However, accurately segmenting EM images remains challenging due to inherent characteristics, including low contrast and subtle grayscale variations. Moreover, as manually annotated EM images are limited, it is usually impractical to utilize deep learning techniques for EM image segmentation. To address these challenges, the pyramid multiscale channel attention network (PmcaNet) is specifically designed. PmcaNet employs a convolutional neural network-based architecture and a multiscale feature pyramid to effectively capture global context information, enhancing its ability to comprehend the intricate structures within EM images. To enable the rapid extraction of channel-wise dependencies, a novel attention module is introduced to enhance the representation of intricate nonlinear features within the images. The performance of PmcaNet is evaluated on two general EM image segmentation datasets as well as a homemade dataset of superalloy materials, regarding pixel-wise accuracy and mean intersection over union (mIoU) as evaluation metrics. Extensive experiments demonstrate that PmcaNet outperforms other models on the ISBI 2012 dataset, achieving 87.85% pixel-wise accuracy and 73.11% mean intersection over union (mIoU), while also advancing results on the Kathuri and SEM-material datasets. [ABSTRACT FROM AUTHOR]

During robot operation, robot components like links and joints may experience collisions or excess loads that can lead to structural damages or cracks. A crack in a structural component can degrade the overall performance of the structure. This study examines the influence of cracks on the vibration characteristics of a baseline robot link. The approach uses the finite element method to simulate the dynamics of planar robot link models with and without artificial cracks with different sizes, locations, and orientations in the ABAQUS software. The robot link models include one intact model and five defective models with cracks. A rectangular crack with a fixed length of 1 mm and a varying width from 0.001 to 0.1 mm is applied to a specific location along the robot link. Finite element analysis and machine learning are used to simulate and characterize the vibration of each robot link with one fixed end and one free end. The vibration responses are measured at the free end. The measured vibration data are then transformed into two-dimensional (2D) image data using the Gramian Angular Summation Field method. A convolutional neural network is then trained with the image data for crack detection and analysis. The results indicate that the proposed method demonstrates 98.25% accuracy on the data generated by the simulation experiments. [ABSTRACT FROM AUTHOR]

The integration of artificial intelligence (AI) with Digital Twins (DTs) has emerged as a promising approach to revolutionize healthcare, particularly in terms of diagnosis and management of thoracic disorders. This study proposes a comprehensive framework, named Lung-DT, which leverages IoT sensors and AI algorithms to establish the digital representation of a patient's respiratory health. Using the YOLOv8 neural network, the Lung-DT system accurately classifies chest X-rays into five distinct categories of lung diseases, including "normal", "covid", "lung_opacity", "pneumonia", and "tuberculosis". The performance of the system was evaluated employing a chest X-ray dataset available in the literature, demonstrating average accuracy of 96.8%, precision of 92%, recall of 97%, and F1-score of 94%. The proposed Lung-DT framework offers several advantages over conventional diagnostic methods. Firstly, it enables real-time monitoring of lung health through continuous data acquisition from IoT sensors, facilitating early diagnosis and intervention. Secondly, the AI-powered classification module provides automated and objective assessments of chest X-rays, reducing dependence on subjective human interpretation. Thirdly, the twin digital representation of the patient's respiratory health allows for comprehensive analysis and correlation of multiple data streams, providing valuable insights as to personalized treatment plans. The integration of IoT sensors, AI algorithms, and DT technology within the Lung-DT system demonstrates a significant step towards improving thoracic healthcare. By enabling continuous monitoring, automated diagnosis, and comprehensive data analysis, the Lung-DT framework has enormous potential to enhance patient outcomes, reduce healthcare costs, and optimize resource allocation. [ABSTRACT FROM AUTHOR]

Mobile fitness applications provide the opportunity to show users real-time feedback on their current fitness activity. For such applications, it is essential to accurately track the user's current fitness activity using available mobile sensors, such as inertial measurement units (IMUs). Convolutional neural networks (CNNs) have been shown to produce strong results in different time series classification tasks, including the recognition of daily living activities. However, fitness activities can present unique challenges to the human activity recognition task (HAR), including greater similarity between individual activities and fewer available data for model training. In this paper, we evaluate the applicability of CNNs to the fitness activity recognition task (FAR) using IMU data and determine the impact of input data size and sensor count on performance. For this purpose, we adapted three existing CNN architectures to the FAR task and designed a fourth CNN variant, which we call the scaling fully convolutional network (Scaling-FCN). We designed a preprocessing pipeline and recorded a running exercise data set with 20 participants, in which we evaluated the respective recognition performances of the four networks, comparing them with three traditional machine learning (ML) methods commonly used in HAR. Although CNN architectures achieve at least 94% test accuracy in all scenarios, two traditional ML architectures surpass them in the default scenario, with support vector machines (SVMs) achieving 99.00 ± 0.34% test accuracy. The removal of all sensors except one foot sensor reduced the performance of traditional ML architectures but improved the performance of CNN architectures on our data set, with our Scaling-FCN reaching the highest accuracy of 99.86 ± 0.11% on the test set. Our results suggest that CNNs are generally well suited for fitness activity recognition, and noticeable performance improvements can be achieved if sensors are dropped selectively, although traditional ML architectures can still compete with or even surpass CNNs when favorable input data are utilized. [ABSTRACT FROM AUTHOR]

The present study presents a novel approach for identifying epileptogenic tubers in patients with tuberous sclerosis complex (TSC) and automating tuber segmentation using a three-dimensional convolutional neural network (3D CNN). The study retrospectively included 31 TSC patients whose lesions were manually annotated from multiparametric neuroimaging data. Epileptogenic tubers were determined via presurgical evaluation and stereoelectroencephalography recording. Neuroimaging metrics were extracted and compared between epileptogenic and non-epileptogenic tubers. Additionally, five datasets with different preprocessing strategies were used to construct and train 3D CNNs for automated tuber segmentation. The normalized positron emission tomography (PET) metabolic value was significantly lower in epileptogenic tubers defined via presurgical evaluation (p = 0.001). The CNNs showed high performance for localizing tubers, with an accuracy between 0.992 and 0.994 across the five datasets. The automated segmentations were highly correlated with clinician-based features. The neuroimaging characteristics for epileptogenic tubers were demonstrated, increasing surgical confidence in clinical practice. The validated deep learning detection algorithm yielded a high performance in determining tubers with an excellent agreement with reference clinician-based segmentation. Collectively, when coupled with our investigation of minimal input requirements, the approach outlined in this study represents a clinically invaluable tool for the management of TSC. [ABSTRACT FROM AUTHOR]

Human activity recognition (HAR) by deep learning is a challenging and interesting topic. Although there are robust models, there is also a bunch of parameters and variables, which affect the performance such as the number of layers, pooling type. This study presents a new deep learning architecture that is obtained by fine-tuning of the conventional CNN-LSTM model, namely, CNN (+3)-LSTM. Three changes are made to the conventional model to increase the accuracy. Firstly, kernel size is set to 1×1 to extract more information. Secondly, three convolutional layers are added to the model. Lastly, average pooling is used instead of max-pooling. Performance analysis of the proposed model is conducted on the KTH dataset and implemented on Keras. In addition to the overall accuracy of the proposed model, the contribution of each change is observed individually. Results show that adding layers made the highest contribution followed by kernel size and pooling, respectively. The proposed model is compared with state-of-art and outperformed some of the recent studies with a 94.1% recognition rate. [ABSTRACT FROM AUTHOR]

As a new generation of high-performance batteries, lithium-ion batteries have found extensive applications in electric cars, as well as energy storage systems and various other industries. State of charge (SOC) estimation is one of the most important indicators. SOC estimation model of lithium-ion battery based on deep learning neural networks employs diverse external measurement parameters and internal battery parameters as input information, and adopts feed-forward neural network (FNN), convolutional neural network (CNN) and long short-term memory network (LSTM) as predictors to realize the accurate SOC estimation. The model based on deep learning neural networks takes into account the influence of various input parameters and can understand the state of the battery more comprehensively. By using FNN, CNN and LSTM networks, the influence of noise and instability of battery data on SOC estimation can be effectively avoided. After many times of training and verification, the high accuracy and stability of the model can meet the need of SOC estimation for lithium-ion batteries. [ABSTRACT FROM AUTHOR]

Recently, deep learning methods have been widely applied to single image super-resolution reconstruction tasks and achieved great improvement in both quantitative and qualitative evaluations. However, most of the existing convolutional neural network-based methods commonly reduce the number of layers or channels to obtain lightweight model. These strategies may reduce the representation ability of informative features and degrade the network performance. To address this issue, we propose a sparse lightweight attention network (SLAN), a novel SISR algorithm to keep informative features between layers. Specially, a sparse attention feature fusion module with lightweight attention and sparse extracting modules is developed to expand feature extracting receptive field and enhance the informative feature extracting ability. To take advantage of the multi-level features and keep fewer multi-adds, cross fusion is used and demonstrated usefully. Extensive experimental results on public datasets demonstrate the superior performance of our proposed SLAN. The average PSNR(dB)/SSIM values of SLAN are about 0.04/0.0004, 0.06/0.0009, and 0.09/0.0018 larger than the competitors under scaling factors of × 2, × 3 and × 4, respectively. Our SLAN benefits from its model size and low computation cost and may be deployed on mobile platforms. [ABSTRACT FROM AUTHOR]

• We develop a program capable of multi-threading and automatically running REGEN 3.3 software to generate 181,440 regenerator performance data. • We use a simple and novel one-dimensional convolutional neural network model to predict regenerator performance. • This model exhibits excellent performance and high reliability in both fitting and generalization. • This model computes diverse ruleless regenerator parameters, reducing compute time from 2500 min for 1000 data with REGEN 3.3 software to 130 ms. The regenerator is the core component of the regenerative cryogenic refrigerator, for its structure sizes, operating parameters and phase characteristics at the cold and hot ends co-determine the power and efficiency of the refrigerator, and the design parameters of other coupled components. Efficiently predicting the regenerator performance can reduce the design period of cryogenic refrigerators. Addressing the long computational time constraints in the traditional numerical simulation methods, a novel approach based on a one-dimensional convolutional neural network (1D-CNN) was proposed. Initially, a program capable of multi-threading and automatically running the specialized regenerator calculation software REGEN 3.3 was developed. The performance of the regenerators with various parameter combinations at the cold end temperature of 60–120 K were calculated and 181,440 pieces of data were obtained. Subsequently, the architecture and hyperparameters of the model were determined. The trained model exhibits an average relative error of 3.83% for predicting regenerator power, 0.13% for predicting pressure ratio at the hot end, and 1.55% for predicting the coefficient of performance (COP). The model's generalization ability was confirmed by generating data points beyond the original dataset. Additionally, the model allows for the simultaneous calculation of multiple sets of irregular regenerator parameters, and reduces the calculation time from 2500 min for 1000 pieces using REGEN 3.3 software to just 130 ms, representing a decrease by nearly six orders of magnitude. This approach effectively resolves the long computation time associated with traditional numerical simulation methods, and will present a new solution for the rapid and precise design of regenerators. [ABSTRACT FROM AUTHOR]

Introduction: We developed a convolutional neural network (CNN) model to predict treatment outcomes of transforaminal epidural steroid injection (TFESI) for controlling cervical radicular pain due to cervical foraminal stenosis. Methods: We retrospectively recruited 293 patients with cervical TFESI due to cervical radicular pain caused by cervical foraminal stenosis. We obtained a single oblique cervical radiograph from each patient. We cut each oblique cervical radiograph image into a square shape, including the foramen that was targeted for TFESI, the intervertebral disc, the facet joint of the corresponding level with the targeted foramen, and the pedicles of the vertebral bodies just above and below the targeted foramen. Therefore, images including the targeted foramen and structures around the targeted foramen were used as input data. A favorable outcome was defined as a ≥ 50% reduction in the numeric rating scale (NRS) score at 2 months post TFESI compared to the pretreatment NRS score. A poor outcome was defined as a < 50% reduction in the NRS score at 2 months post TFESI vs. the pretreatment score. Results: The area under the curve of our developed model for predicting the treatment outcome of cervical TFESI in patients with cervical foraminal stenosis was 0.823. Conclusion: A CNN model trained using oblique cervical radiographs can be helpful in predicting treatment outcomes after cervical TFESI in patients with cervical foraminal stenosis. If the predictive accuracy is increased, we believe that the deep learning model using cervical radiographs as input data can be easily and widely used in clinics or hospitals. [ABSTRACT FROM AUTHOR]

Flash event generates enormous traffic and the cloud service providers use sustaining techniques like scaling and content delivery network to up their services. One of the main bottlenecks that the cloud service providers still find difficult to tackle is flash attacks. Illegitimate users send craftily designed packets to land up inside the server for wreaking havoc. As deep learning autoencoder has the potential to detect malicious traffic it has been used in this research study to develop an ensemble. Convolutional neural network is efficacious in overcoming the issue of overfitting; deep autoencoder is proficient in extracting features through dimensionality reduction. In order to obtain both these advantages it was decided to develop an ensemble keeping denoising autoencoder as the core element. The process of addressing a flash attack requires first detecting the presence of bot in malicious traffic, second studying its nature by observing its behavioral manifestations. Detection of botnet was achieved by three ensembles, namely, DAE_CNN, DAE_MLP, and DAE_XGB. But capturing its external manifested behavior is challenging, because the bot signatures are always in a state of flux. The simulated empirical study yielded an appreciable outcome. Its accuracy rate was 99.9% for all the three models and the false positive rates were 0, 0.006, and 0.001, respectively. [ABSTRACT FROM AUTHOR]

Several deep learning and transformer models have been recommended in previous research to deal with the classification of hyperspectral images (HSIs). Among them, one of the most innovative is the bidirectional encoder representation from transformers (BERT), which applies a distance-independent approach to capture the global dependency among all pixels in a selected region. However, this model does not consider the local spatial-spectral and spectral sequential relations. In this paper, a dual-dimensional (i.e., spatial and spectral) BERT (the so-called D2BERT) is proposed, which improves the existing BERT model by capturing more global and local dependencies between sequential spectral bands regardless of distance. In the proposed model, two BERT branches work in parallel to investigate relations among pixels and spectral bands, respectively. In addition, the layer intermediate information is used for supervision during the training phase to enhance the performance. We used two widely employed datasets for our experimental analysis. The proposed D2BERT shows superior classification accuracy and computational efficiency with respect to some state-of-the-art neural networks and the previously developed BERT model for this task. [ABSTRACT FROM AUTHOR]

Objectives: The quality of network structure is the key to obtain excellent classification performance. Most of existing deep models are designed by the trial-and-error method. When the same model processes different hyperspectral images (HSIs), different network hyperparameters need to be set artificially to adapt to the data characteristics of different images, greatly increasing the difficulty of network design. Aiming at the problem of deep learning model design in hyperspectral image classification, this paper proposes an automatic network structure search method for HSI classification. Methods: In the proposed method, the differentiable structure search technology is first used to search the network structure on the source HSI data sets, then the deep network model is constructed by stacking cells, and finally the classification performance is evaluated on the target HSIs. It should be pointed out that, the proposed method only conducts one search on the source HSI data sets, and the obtained deep model can be applied to other target HSI classification tasks, which can effectively improve the utilization rate of the model. To improve the generalization ability and classification accuracy of the model, multi-source and multi-resolution HSIs are employed to construct the precollected data set, and the partial channel connection operation is introduced to further improve the search efficiency. Results: Three HSIs inculding Houston, Chikusei and Pavia Center are selected as source data. Four HSIs including University of Pavia, Indian Pines, Salinas and Houston 2018 are selected as the target HSIs. The source data is used to determine the network model and the target HSIs are used to evaluate the classification performance. Experimental results show that the proposed method can automatically search out a universal deep network model suitable for HSI classification task, and this model can achieve better classification performance than the conventional deep learning models. The overall classification accuracies of University of Pavia, Indian Pines, Salinas and Houston 2018 are 98.15%, 98.74%, 97.30% and 74.47%, respectively. Conclusions: Compared with other deep models, the model in this paper is automatically determined by differentiable architecture search algorithm according to the data characteristics of HSIs, effectively improving the classification accuracy of target HSIs while avoiding complex network design. Extensive experiments verify the applicability and effectiveness of the proposed method in HSI classification. [ABSTRACT FROM AUTHOR]

Exposing a healthy wound bed for skin grafting is an important step during burn surgery to ensure graft take and maintain good functional outcomes. Currently, the removal of non-viable tissue in the burn wound bed during excision is determined by expert clinician judgment. Using a porcine model of tangential burn excision, we investigated the effectiveness of an intraoperative multispectral imaging device combined with artificial intelligence to aid clinician judgment for the excision of non-viable tissue. Multispectral imaging data was obtained from serial tangential excisions of thermal burn injuries and used to train a deep learning algorithm to identify the presence and location of non-viable tissue in the wound bed. Following algorithm development, we studied the ability of two surgeons to estimate wound bed viability, both unaided and aided by the imaging device. The deep learning algorithm was 87% accurate in identifying the viability of a burn wound bed. When paired with the surgeons, this device significantly improved their abilities to determine the viability of the wound bed by 25% (p = 0.03). Each time a surgeon changed their decision after seeing the AI model output, it was always a change from an incorrect decision to excise more tissue to a correct decision to stop excision. This study provides insight into the feasibility of image-guided burn excision, its effect on surgeon decision making, and suggests further investigation of a real-time imaging system for burn surgery could reduce over-excision of burn wounds. • Study demonstrates usefulness of an imaging device in providing surgical guidance to aid clinicians in burn excision. • Multispectral imaging and artificial intelligence correctly identified the viable and non-viable wound bed. • Multispectral imaging and artificial intelligence helped surgeons stop at the correct depth during burn excision surgery. • Multispectral imaging and artificial intelligence could eventually provide reliable guidance during burn surgery. [ABSTRACT FROM AUTHOR]

Timely forecasting of aboveground vegetation biomass is crucial for effective management and ensuring food security. However, research on predicting aboveground biomass remains scarce. Artificial intelligence (AI) methods could bridge this research gap and provide early warning to planners and stakeholders. This study evaluates the effectiveness of deep learning (DL) algorithms in predicting aboveground vegetation biomass with limited-size data. It employs an iterative forecasting procedure for four target horizons, comparing the performance of DL models—multi-layer perceptron (MLP), long short-term memory (LSTM), gated recurrent unit (GRU), convolutional neural network (CNN), and CNN-LSTM—against the traditional seasonal autoregressive integrated moving average (SARIMA) model, serving as a benchmark. Five limited-size vegetation biomass time series from Kenyan grasslands with values at 15-day intervals over a 20-year period were chosen for this purpose. Comparing the outcomes of these models revealed significant differences (p < 0.05); however, none of the models proved superior among the five time series and the four horizons evaluated. The SARIMA, CNN, and CNN-LSTM models performed best, with the statistical model slightly outperforming the other two. Additionally, the accuracy of all five models varied significantly according to the prediction horizon (p < 0.05). As expected, the accuracy of the models decreased as the prediction horizon increased, although this relationship was not strictly monotonic. Finally, this study indicated that, in limited-size aboveground vegetation biomass time series, there is no guarantee that deep learning methods will outperform traditional statistical methods. [ABSTRACT FROM AUTHOR]

Invasive plant species pose significant biodiversity and ecosystem threats. Real-time identification of invasive plants is a crucial prerequisite for early and timely prevention. While deep learning has shown promising results in plant recognition, the use of deep learning models often involve a large number of parameters and high data requirements for training. Unfortunately, the available data for various invasive plant species are often limited. To address this challenge, this study proposes a lightweight deep learning model called IPMCNet for the identification of multiple invasive plant species. IPMCNet attains high recognition accuracy even with limited data and exhibits strong generalizability. Simultaneously, by employing depth-wise separable convolutional kernels, splitting channels, and eliminating fully connected layer, the model's parameter count is lower than that of some existing lightweight models. Additionally, the study explores the impact of different loss functions, and the insertion of various attention modules on the model's accuracy. The experimental results reveal that, compared with eight other existing neural network models, IPMCNet achieves the highest classification accuracy of 94.52%. Furthermore, the findings suggest that focal loss is the most effective loss function. The performance of the six attention modules is suboptimal, and their insertion leads to a decrease in model accuracy. [ABSTRACT FROM AUTHOR]

Peanut leaf spot is a worldwide disease whose prevalence poses a major threat to peanut yield and quality, and accurate prediction models are urgently needed for timely disease management. In this study, we proposed a novel peanut leaf spot prediction method based on an improved long short-term memory (LSTM) model and multi-year meteorological data combined with disease survey records. Our method employed a combination of convolutional neural networks (CNNs) and LSTMs to capture spatial–temporal patterns from the data and improve the model's ability to recognize dynamic features of the disease. In addition, we introduced a Squeeze-and-Excitation (SE) Network attention mechanism module to enhance model performance by focusing on key features. Through several hyper-parameter optimization adjustments, we identified a peanut leaf spot disease condition index prediction model with a learning rate of 0.001, a number of cycles (Epoch) of 800, and an optimizer of Adma. The results showed that the integrated model demonstrated excellent prediction ability, obtaining an RMSE of 0.063 and an R 2 of 0.951, which reduced the RMSE by 0.253 and 0.204, and raised the R 2 by 0.155 and 0.122, respectively, compared to the single CNN and LSTM. Predicting the occurrence and severity of peanut leaf spot disease based on the meteorological conditions and neural networks is feasible and valuable to help growers make accurate management decisions and reduce disease impacts through optimal fungicide application timing. [ABSTRACT FROM AUTHOR]

Automated building extraction is imperative for several geospatial applications such as monitoring disaster-affected buildings and urban planning. Existing deep learning (DL)-based building extraction methods fail to capture high-level semantic features due to the complex nature and diverse appearance of visually similar structures. To address this issue, in this letter, we propose an enhanced multi-scale attentive feature fusion network (EMAFF-Net) for building extraction from remote sensing (RS) images. EMAFF-Net is an end-to-end DL architecture based on U-Net that includes: i) an encoder; ii) an enhanced multi-scale feature fusion (EMFF) module; iii) a refined multi-scale convolutional block attention (RM-CBAM) module and iv) a decoder with refinement layers. To extract multi-scale contextual information, we incorporate an RM-CBAM module into the lateral connections of encoder-decoder layers of EMAFF-Net. Further, a novel EMFF module is integrated to obtain fine-grained features from the lowest encoder layer with minimal trainable parameters required. We evaluate the performance of the proposed network on two benchmark datasets: Massachusetts (MAS) and WHU building datasets. The experimental results show that the proposed approach outperforms the existing reference methods showcasing its potential in practical applications. [ABSTRACT FROM AUTHOR]

Accurate and reliable estimation of the pelvic tilt is one of the essential pre-planning factors for total hip arthroplasty to prevent common post-operative complications such as implant impingement and dislocation. Inspired by the latest advances in deep learning-based systems, our focus in this paper has been to present an innovative and accurate method for estimating the functional pelvic tilt (PT) from a standing anterior–posterior (AP) radiography image. We introduce an encoder–decoder-style network based on a concurrent learning approach called VGG-UNET (VGG embedded in U-NET), where a deep fully convolutional network known as VGG is embedded at the encoder part of an image segmentation network, i.e., U-NET. In the bottleneck of the VGG-UNET, in addition to the decoder path, we use another path utilizing light-weight convolutional and fully connected layers to combine all extracted feature maps from the final convolution layer of VGG and thus regress PT. In the test phase, we exclude the decoder path and consider only a single target task i.e., PT estimation. The absolute errors obtained using VGG-UNET, VGG, and Mask R-CNN are 3.04 ± 2.49, 3.92 ± 2.92, and 4.97 ± 3.87, respectively. It is observed that the VGG-UNET leads to a more accurate prediction with a lower standard deviation (STD). Our experimental results demonstrate that the proposed multi-task network leads to a significantly improved performance compared to the best-reported results based on cascaded networks. [ABSTRACT FROM AUTHOR]

While the term task load (TL) refers to external task demands, the amount of work, or the number of tasks to be performed, mental workload (MWL) refers to the individual's effort, mental capacity, or cognitive resources utilized while performing a task. MWL in multitasking scenarios is often closely linked with the quantity of tasks a person is handling within a given timeframe. In this study, we challenge this hypothesis from the perspective of electroencephalography (EEG) using a deep learning approach. We conducted an EEG experiment with 50 participants performing NASA Multi-Attribute Task Battery II (MATB-II) under 4 different task load levels. We designed a convolutional neural network (CNN) to help with two distinct classification tasks. In one setting, the CNN was used to classify EEG segments based on their task load level. In another setting, the same CNN architecture was trained again to detect the presence of individual MATB-II subtasks. Results show that, while the model successfully learns to detect whether a particular subtask is active in a given segment (i.e., to differentiate between different subtasks-related EEG patterns), it struggles to differentiate between the two highest levels of task load (i.e., to distinguish MWL-related EEG patterns). We speculate that the challenge comes from two factors: first, the experiment was designed in a way that these two highest levels differed only in the quantity of work within a given timeframe; and second, the participants' effective adaptation to increased task demands, as evidenced by low error rates. Consequently, this indicates that under such conditions in multitasking, EEG may not reflect distinct enough patterns to differentiate higher levels of task load. [ABSTRACT FROM AUTHOR]

Simple Summary: Diagnosing glioblastoma multiforme (GBM), solitary brain metastases (SBM), and primary central nervous system lymphoma (PCNSL) in malignant tumors of the central nervous system using multi-modal magnetic resonance imaging (MRI) is significantly important in helping physicians develop treatment plans and enhance patient prognosis. In this paper, MFFC-Net is developed and validated using deep learning methods to predict these three tumor categories from multi-modal MRI without the manual region of interest (ROI). MFFC-Net first uses a multi-encoder with DenseBlocks to extract deep features from multi-modal MRI. Then, the feature fusion layer fuses the deep information between different modalities and tissues. Finally, the spatial-channel attention module suppresses redundant new information and activates tumor classification-related features. Compared with radiomics models, MFFC-Net demonstrated higher accuracy. In addition, the results in the different sequences provide important references for future clinical work on MRI image acquisition. We believe that MFFC-Net has the potential to assist in the diagnosis and treatment of brain tumors in the future. (1) Background: Diagnosis of glioblastoma (GBM), solitary brain metastases (SBM), and primary central nervous system lymphoma (PCNSL) plays a decisive role in the development of personalized treatment plans. Constructing a deep learning classification network to diagnose GBM, SBM, and PCNSL with multi-modal MRI is important and necessary. (2) Subjects: GBM, SBM, and PCNSL were confirmed by histopathology with the multi-modal MRI examination (study from 1225 subjects, average age 53 years, 671 males), 3.0 T T2 fluid-attenuated inversion recovery (T2-Flair), and Contrast-enhanced T1-weighted imaging (CE-T1WI). (3) Methods: This paper introduces MFFC-Net, a classification model based on the fusion of multi-modal MRIs, for the classification of GBM, SBM, and PCNSL. The network architecture consists of parallel encoders using DenseBlocks to extract features from different modalities of MRI images. Subsequently, an L 1 − n o r m feature fusion module is applied to enhance the interrelationships among tumor tissues. Then, a spatial-channel self-attention weighting operation is performed after the feature fusion. Finally, the classification results are obtained using the full convolutional layer (FC) and Soft-max. (4) Results: The ACC of MFFC-Net based on feature fusion was 0.920, better than the radiomics model (ACC of 0.829). There was no significant difference in the ACC compared to the expert radiologist (0.920 vs. 0.924, p = 0.774). (5) Conclusions: Our MFFC-Net model could distinguish GBM, SBM, and PCNSL preoperatively based on multi-modal MRI, with a higher performance than the radiomics model and was comparable to radiologists. [ABSTRACT FROM AUTHOR]

The role of capsule endoscopy and enteroscopy in managing various small-bowel pathologies is well-established. However, their broader application has been hampered mainly by their lengthy reading times. As a result, there is a growing interest in employing artificial intelligence (AI) in these diagnostic and therapeutic procedures, driven by the prospect of overcoming some major limitations and enhancing healthcare efficiency, while maintaining high accuracy levels. In the past two decades, the applicability of AI to gastroenterology has been increasing, mainly because of the strong imaging component. Nowadays, there are a multitude of studies using AI, specifically using convolutional neural networks, that prove the potential applications of AI to these endoscopic techniques, achieving remarkable results. These findings suggest that there is ample opportunity for AI to expand its presence in the management of gastroenterology diseases and, in the future, catalyze a game-changing transformation in clinical activities. This review provides an overview of the current state-of-the-art of AI in the scope of small-bowel study, with a particular focus on capsule endoscopy and enteroscopy. [ABSTRACT FROM AUTHOR]

Simple Summary: This systematic review demonstrated that artificial intelligence (AI) can help detect metastatic prostate cancer with or without lymph node involvement on prostate-specific membrane antigen (PSMA) PET scans with high accuracy. Additional benefits of AI include the ability to estimate the volume of metastatic cancer, prognosticate, and differentiate bony metastasis from post-radiotherapy bone changes. AI can also improve workflow by helping to standardize reporting and automate time-consuming tasks. However, given the variable sensitivity and positive predictive value of AI, it is recommended that an experienced nuclear medicine physician proofread the final report. Larger studies producing more consistent results are needed before AI can be fully integrated into PSMA reporting. Early detection of metastatic prostate cancer (mPCa) is crucial. Whilst the prostate-specific membrane antigen (PSMA) PET scan has high diagnostic accuracy, it suffers from inter-reader variability, and the time-consuming reporting process. This systematic review was registered on PROSPERO (ID CRD42023456044) and aims to evaluate AI's ability to enhance reporting, diagnostics, and predictive capabilities for mPCa on PSMA PET scans. Inclusion criteria covered studies using AI to evaluate mPCa on PSMA PET, excluding non-PSMA tracers. A search was conducted on Medline, Embase, and Scopus from inception to July 2023. After screening 249 studies, 11 remained eligible for inclusion. Due to the heterogeneity of studies, meta-analysis was precluded. The prediction model risk of bias assessment tool (PROBAST) indicated a low overall risk of bias in ten studies, though only one incorporated clinical parameters (such as age, and Gleason score). AI demonstrated a high accuracy (98%) in identifying lymph node involvement and metastatic disease, albeit with sensitivity variation (62–97%). Advantages included distinguishing bone lesions, estimating tumour burden, predicting treatment response, and automating tasks accurately. In conclusion, AI showcases promising capabilities in enhancing the diagnostic potential of PSMA PET scans for mPCa, addressing current limitations in efficiency and variability. [ABSTRACT FROM AUTHOR]

The travel time of ambient noise cross-correlation is widely used in geophysics, but traditional methods for picking the travel time of correlation are either difficult to be applied to data with low signal-to-noise ratio (SNR), or make some assumptions which fail to be achieved in many realistic situations, or require a lot of complex calculations. Here, we present a neural network based on convolutional neural networks (CNN) and Transformer for the travel time picking of ambient noise cross-correlation. CNNs expand the dimension of the vector of each time step for the input of Transformer. Transformer focuses the model's attention on the key parts of the sequence. Model derives the travel time according to the attention. 102,000 cross-correlations are used to train the network. Compared with traditional methods, the approach is easy to use and has a better performance, especially for the low SNR data. Then, we test our model on another ambient noise cross-correlation dataset, which contains cross-correlations from different regions and at different scales. The model has good performance on the test dataset. It can be seen from the experiment that the travel time of the cross-correlation function of ambient noise with an average SNR as low as 9.3 can be picked. 97.2% of the picked travel times are accurate, and the positive and negative travel time of most cross-correlations are identical (90.2%). Our method can be applied to seismic instrument performance verification, seismic velocity imaging, source location and other applications for its good ability to pick travel time accurately. [ABSTRACT FROM AUTHOR]

Objectives: With smartphones and wearable devices becoming ubiquitous, they offer an opportunity for large-scale voice sampling. This systematic review explores the application of deep learning models for the automated analysis of voice samples to detect vocal cord pathologies. Methods: We conducted a systematic literature review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting guidelines. We searched MEDLINE and Embase databases for original publications on deep learning applications for diagnosing vocal cord pathologies between 2002 and 2022. Risk of bias was assessed using Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2). Results: Out of the 14 studies that met the inclusion criteria, data from a total of 3037 patients were analyzed. All studies were retrospective. Deep learning applications targeted Reinke's edema, nodules, polyps, cysts, unilateral cord paralysis, and vocal fold cancer detection. Most pathologies had detection accuracy above 90%. Thirteen studies (93%) exhibited a high risk of bias and concerns about applicability. Conclusions: Technology holds promise for enhancing the screening and diagnosis of vocal cord pathologies. While current research is limited, the presented studies offer proof of concept for developing larger-scale solutions. [ABSTRACT FROM AUTHOR]

Gas tungsten arc welding (GTAW) is a popular technology for joining metallic parts with high stability and quality. However, many GTAW-related processes are conducted manually, which is time-consuming, and the weld quality strongly depends on the skill of the welder. Although various automatic GTAW methods and systems have been developed, controlling the weld quality remains a challenge because of the difficulties involved in predicting the weld quality, such as the penetration depth and backside bead geometry, during welding. Hence, this paper proposes an accurate and effective method for estimating the penetration depth through weld pool monitoring using a convolutional neural network (CNN) trained on weld pool images. The weld pool images contained several objects, each influencing the prediction accuracy. The CNN architecture and the structure of the fully connected layers (FCLs) also affected the prediction accuracy. To optimize the performance of the estimation model, the effects of each object in the weld pool image and the structure of the CNN architecture were analyzed and evaluated. The structure of the FCLs that outputted a quantitative penetration depth was optimized and evaluated through hyperparameter tuning. With the proposed method, the optimized model could quantitatively predict the penetration depth; the mean absolute error was 0.0516 mm, with an R2 value of 0.998. Accurately predicting the penetration depth can be employed in real-time weld quality control to ensure a sound weld back bead. [ABSTRACT FROM AUTHOR]

Cardiovascular diseases (CVD) represent a significant global health challenge, often remaining undetected until severe cardiac events, such as heart attacks or strokes, occur. In regions like Qatar, research focused on non-invasive CVD identification methods, such as retinal imaging and dual-energy X-ray absorptiometry (DXA), is limited. This study presents a groundbreaking system known as Multi-Modal Artificial Intelligence for Cardiovascular Disease (M2AI-CVD), designed to provide highly accurate predictions of CVD. The M2AI-CVD framework employs a four-fold methodology: First, it rigorously evaluates image quality and processes lower-quality images for further analysis. Subsequently, it uses the Entropy-based Fuzzy C Means (EnFCM) algorithm for precise image segmentation. The Multi-Modal Boltzmann Machine (MMBM) is then employed to extract relevant features from various data modalities, while the Genetic Algorithm (GA) selects the most informative features. Finally, a ZFNet Convolutional Neural Network (ZFNetCNN) classifies images, effectively distinguishing between CVD and Non-CVD cases. The research’s culmination, tested across five distinct datasets, yields outstanding results, with an accuracy of 95.89%, sensitivity of 96.89%, and specificity of 98.7%. This multi-modal AI approach offers a promising solution for the accurate and early detection of cardiovascular diseases, significantly improving the prospects of timely intervention and improved patient outcomes in the realm of cardiovascular health. [ABSTRACT FROM AUTHOR]

Convolutional neural networks (CNNs) are currently one of the most popular image classification technologies. Their excellent image classification ability enables the prediction of the ash content in clean coal using coal flotation froth images. Herein, a scheme is proposed to predict the ash content in coal flotation concentrate using coal flotation froth images and CNN architecture. According to the ash content of the concentrate, the coal flotation froth image data set is divided into seven interval categories, and data augmentation is used to expand the data set. Then, several recently trained CNNs (ResNet, EfficientNet, EfficientNetV2, ConvNeXt) are used to classify the coal flotation froth images with different concentrate ash content interval (integer ±1%). The classification performance of each network model, relationship between model performance and hyperparameters, and the abstract pixel features of the best model are visualized. The EfficientNetV2-L network achieves the highest classification accuracy (99%) after fine-tuning. The results indicate that the CNNs are effective in predicting the ash content of clean coal by classifying coal flotation froth images. The optimized CNN model is applied to froth images in the industry, resulting in high classification performance, which demonstrates that the CNN has high potential for industrial applications. [ABSTRACT FROM AUTHOR]

Convolutional Neural Networks (CNN) with different architectures have shown promising results in skin cancer diagnosis. However, CNN has a high computational cost, which makes the need for a light version of CNN a desirable step. This version can be used on small devices, such as mobile phones or tablets. A light version can be created using pruning techniques. In this study, iterative magnitude pruning (IMP) is utilized. This method depends on pruning the network iteratively. The IMP method is applied on AlexNet with transfer learning (TL) and data augmentation. The proposed IMP AlexNet with TL is applied on three different skin cancer datasets which are PAD-UFES-20, MED-NODE, and PH2 dataset. The datasets used are a combination of smartphone, dermoscopic, and non-dermoscopic images. Different CNN versions are applied on the same datasets for comparison with IMP AlexNet. The CNNs used are VGG-16, ShuffleNet, SqueezNet, DarkNet-19, DarkNet-53, and Inception-v3. The proposed IMP AlexNet achieved accuracies of 97.62%, 96.79%, and 96.75%, with accuracy losses of 1.53%, 2.3%, and 2.2%, respectively, compared to the original AlexNet. In addition, the proposed IMP AlexNet requires less running time and memory usage than the traditional AlexNet. The average running time for IMP AlexNet is 0.45 min, 0.28 min, and 0.3 min, for PAD-UFES-20, MED-NODE, and PH2 datasets, respectively. The average RAM usage with IMP AlexNet is 1.8 GB, 1.6 GB, and 1.7 GB, respectively. IMP AlexNet accelerates the average running time by approximately 15 times that of the traditional AlexNet and reduces the average RAM used by 40%. [ABSTRACT FROM AUTHOR]

Wireless physical layer authentication has emerged as a promising approach to wireless security. The topic of wireless node classification and recognition has experienced significant advancements due to the rapid development of deep learning techniques. The potential of using deep learning to address wireless security issues should not be overlooked due to its considerable capabilities. Nevertheless, the utilization of this approach in the classification of wireless nodes is impeded by the lack of available datasets. In this study, we provide two models based on a data-driven approach. First, we used generative adversarial networks to design an automated model for data augmentation. Second, we applied a convolutional neural network to classify wireless nodes for a wireless physical layer authentication model. To verify the effectiveness of the proposed model, we assessed our results using an original dataset as a baseline and a generated synthetic dataset. The findings indicate an improvement of approximately 19% in classification accuracy rate. [ABSTRACT FROM AUTHOR]

Automatic extraction of building contours from high-resolution images is of great significance in the fields of urban planning, demographics, and disaster assessment. Network models based on convolutional neural network (CNN) and transformer technology have been widely used for semantic segmentation of buildings from high resolution remote sensing images (HRSI). However, the fixed geometric structure and the local receptive field of the convolutional kernel are not good at global feature extraction, and the transformer technique with self-attention mechanism introduces computational redundancies and extracts local feature details poorly in the process of modeling the global contextual information. In this paper, a dual-branch fused reconstructive transformer network, DFRTNet, is proposed for efficient and accurate building extraction. In the encoder, the traditional transformer is reconfigured by designing the local and global feature extraction module (LGFE); the branch of global feature extraction (GFE) performs dynamic range attention (DRA) based on the idea of top-k attention for extracting global features; furthermore, the branch of local feature extraction (LFE) is used to obtain fine-grained features. The multilayer perceptron (MLP) is employed to efficiently fuse the local and global features. In the decoder, a simple channel attention module (CAM) is used in the up-sampling part to enhance channel dimension features. Our network achieved the best segmentation accuracy on both the WHU and Massachusetts building datasets when compared to other mainstream and state-of-the-art methods. [ABSTRACT FROM AUTHOR]

• The deep learning model can better extract the deeper features of the multi-point deformation sequence. • MVMD algorithm reduces the complexity of multi-point displacement sequence. • The model considers the spatiotemporal characteristics of dam deformation, the prediction accuracy is obviously improved. It is difficult for single measuring point model to synchronously characterize the spatial correlation of deformation sequence of concrete dam and the similarity of displacements of different measuring points. Considering the limitation of conventional model in dealing with spatial-temporal data structure, In this study, a deep learning technique for multi-point and multi-output deformation prediction of concrete dams is proposed, which is based on Multivariate Variational Mode Decomposition (MVMD) and three-dimensional convolutional neural network (3D-CNN) combined model. Firstly, the multi-point sequence is decomposed into several relatively stable sub-sequences with different frequency scales by MVMD, which reduces the complexity of the multi-point displacement sequence. Then, In order to enhance the model 's ability to fit the time and spatial features of the sequence, the 3D-CNN model is used to predict the decomposed components and reconstruct the displacement data of the measuring points. This study focuses on the principle, process and calculation steps of the proposed model, and verifies the proposed model through the deformation data of three typical measuring points. The test results show that the prediction model proposed in this paper is an effective method for dam deformation prediction. [ABSTRACT FROM AUTHOR]