Search

In brain–computer interface motor imagery (BCI-MI) systems, convolutional neural networks (CNNs) have traditionally dominated as the deep learning method of choice, demonstrating significant advancements in state-of-the-art studies. Recently, Transformer models with attention mechanisms have emerged as a sophisticated technique, enhancing the capture of long-term dependencies and intricate feature relationships in BCI-MI. This research investigates the performance of EEG-TCNet and EEG-Conformer models, which are trained and validated using various hyperparameters and bandpass filters during preprocessing to assess improvements in model accuracy. Additionally, this study introduces EEG-TCNTransformer, a novel model that integrates the convolutional architecture of EEG-TCNet with a series of self-attention blocks employing a multi-head structure. EEG-TCNTransformer achieves an accuracy of 83.41% without the application of bandpass filtering.

Face recognition (FR) uses a passive approach to person authentication that avoids face-to-face contact. Among different FR techniques, most FR approaches place little emphasis on reducing powerful cryptography and instead concentrate on increasing recognition rates. In this paper, we have proposed the Hidden Markov Model (HMM) and convolutional Neural Network (CNN) models for FR by using ORL and Yale datasets. Facial images from the given data sets are divided into 3 portions, 4 portions, 5 portions, and 6 portions corresponding to their respective HMM hidden states being used in the HMM model. Quantized levels of eigenvalues and eigenvector coefficients of overlapping blocks of facial images define the observation states of the HMM model. For image selection and rejection, a threshold is calculated using singular value decomposition (SVD). After training HMM on 3 states HMM, 4 states HMM, 5 states HMM, and 6 states HMM, the recognition accuracies are 96.5%, 98.5%, 98.5%, and 99.5%, respectively, on the ORL database and 90.6667%, 94.6667%, 94.6667%, and 94.6667% on the Yale database. The CNN model uses convolutional layers, a max-pooling layer, a flattening layer, a dense layer, and a dropout layer. Relu is used as the activation function in all layers except in the last layer, where softmax is used as the activation function. Cross entropy is used as a loss function, and we have used the Adam optimizer in our proposed algorithm. The proposed CNN model has given 100% training and testing accuracy on the ORL data set. While using the Yale data set, the CNN model has a training accuracy of 100% and a testing accuracy of 85.71%. In this paper, our proposed model showed that the HMM model is cost-effective with lesser accuracy, while the CNN model is more accurate as compared to HMM but has a higher computational cost.

Background: Evaluating chest X-rays is a complex and high-demand task due to the intrinsic challenges associated with diagnosing a wide range of pulmonary conditions. Therefore, advanced methodologies are required to categorize multiple conditions from chest X-ray images accurately. Methods: This study introduces an optimized deep learning approach designed for the multi-label categorization of chest X-ray images, covering a broad spectrum of conditions, including lung opacity, normative pulmonary states, COVID-19, bacterial pneumonia, viral pneumonia, and tuberculosis. An optimized deep learning model based on the modified VGG16 architecture with SE blocks was developed and applied to a large dataset of chest X-ray images. The model was evaluated against state-of-the-art techniques using metrics such as accuracy, F1-score, precision, recall, and area under the curve (AUC). Results: The modified VGG16-SE model demonstrated superior performance across all evaluated metrics. The model achieved an accuracy of 98.49%, an F1-score of 98.23%, a precision of 98.41%, a recall of 98.07% and an AUC of 98.86%. Conclusion: This study provides an effective deep learning approach for categorizing chest X-rays. The model’s high performance across various lung conditions suggests its potential for integration into clinical workflows, enhancing the accuracy and speed of pulmonary disease diagnosis.

Inundation maps that show water depths that occur in the event of a flood are essential for protection. Especially information on timings is crucial. Creating a dynamic inundation map with depth data in temporal resolution is a major challenge and is not possible with physical models, as these are too slow for real-time predictions. To provide a dynamic inundation map in real-time, we developed a data-driven multi-step inundation forecast system for fluvial flood events. The forecast system is based on a convolutional neural network (CNN), feature-informed dense layers, and a recursive connection from the predicted inundation at timestep t as a new input for timestep t + 1. The forecast system takes a hydrograph as input, cuts it at desired timesteps (t), and outputs the respective inundation for each timestep, concluding in a dynamic inundation map with a temporal resolution (t). The prediction shows a Critical Success Index (CSI) of over 90%, an average Root Mean Square Error (RMSE) of 0.07, 0.12, and 0.15 for the next 6 h, 12 h, and 24 h, respectively, and an individual RMSE value below 0.3 m, for all test datasets when compared with the results from a physically based model.

Background: Machine learning models can provide quick and reliable assessments in place of medical practitioners. With over 50 million adults in the United States suffering from osteoarthritis, there is a need for models capable of interpreting musculoskeletal ultrasound images. However, machine learning requires lots of data, which poses significant challenges in medical imaging. Therefore, we explore two strategies for enriching a musculoskeletal ultrasound dataset independent of these limitations: traditional augmentation and diffusion-based image synthesis. Methods: First, we generate augmented and synthetic images to enrich our dataset. Then, we compare the images qualitatively and quantitatively, and evaluate their effectiveness in training a deep learning model for detecting thickened synovium and knee joint recess distension. Results: Our results suggest that synthetic images exhibit some anatomical fidelity, diversity, and help a model learn representations consistent with human opinion. In contrast, augmented images may impede model generalizability. Finally, a model trained on synthetically enriched data outperforms models trained on un-enriched and augmented datasets. Conclusions: We demonstrate that diffusion-based image synthesis is preferable to traditional augmentation. Our study underscores the importance of leveraging dataset enrichment strategies to address data scarcity in medical imaging and paves the way for the development of more advanced diagnostic tools.

Accurate aboveground vegetation biomass forecasting is essential for livestock management, climate impact assessments, and ecosystem health. While artificial intelligence (AI) techniques have advanced time series forecasting, a research gap in predicting aboveground biomass time series beyond single values persists. This study introduces RECMO and DirRecMO, two multi-output methods for forecasting aboveground vegetation biomass. Using convolutional neural networks, their efficacy is evaluated across short-, medium-, and long-term horizons on six Kenyan grassland biomass datasets, and compared with that of existing single-output methods (Recursive, Direct, and DirRec) and multi-output methods (MIMO and DIRMO). The results indicate that single-output methods are superior for short-term predictions, while both single-output and multi-output methods exhibit a comparable effectiveness in long-term forecasts. RECMO and DirRecMO outperform established multi-output methods, demonstrating a promising potential for biomass forecasting. This study underscores the significant impact of multi-output size on forecast accuracy, highlighting the need for optimal size adjustments and showcasing the proposed methods’ flexibility in long-term forecasts. Short-term predictions show less significant differences among methods, complicating the identification of the best performer. However, clear distinctions emerge in medium- and long-term forecasts, underscoring the greater importance of method choice for long-term predictions. Moreover, as the forecast horizon extends, errors escalate across all methods, reflecting the challenges of predicting distant future periods. This study suggests advancing hybrid models (e.g., RECMO and DirRecMO) to improve extended horizon forecasting. Future research should enhance adaptability, investigate multi-output impacts, and conduct comparative studies across diverse domains, datasets, and AI algorithms for robust insights.

The development of robust circuit structures remains a pivotal milestone in electronic device research. This article proposes an integrated hardware–software system designed for the acquisition, processing, and analysis of surface electromyographic (sEMG) signals. The system analyzes sEMG signals to understand muscle function and neuromuscular control, employing convolutional neural networks (CNNs) for pattern recognition. The electrical signals analyzed on healthy and unhealthy subjects are acquired using a meticulously developed integrated circuit system featuring biopotential acquisition electrodes. The signals captured in the database are extracted, classified, and interpreted by the application of CNNs with the aim of identifying patterns indicative of neuromuscular problems. By leveraging advanced learning techniques, the proposed method addresses the non-stationary nature of sEMG recordings and mitigates cross-talk effects commonly observed in electrical interference patterns captured by surface sensors. The integration of an AI algorithm with the signal acquisition device enhances the qualitative outcomes by eliminating redundant information. CNNs reveals their effectiveness in accurately deciphering complex data patterns from sEMG signals, identifying subjects with neuromuscular problems with high precision. This paper contributes to the landscape of biomedical research, advocating for the integration of advanced computational techniques to unravel complex physiological phenomena and enhance the utility of sEMG signal analysis.

The rapid evolution of digital technology and the increasing integration of artificial intelligence in agriculture have paved the way for groundbreaking solutions in plant identification. This research pioneers the development and training of a deep learning model to identify three aromatic plants—rosemary, mint, and bay leaf—using advanced computer-aided detection within the You Only Look Once (YOLO) framework. Employing the Cross Industry Standard Process for Data Mining (CRISP-DM) methodology, the study meticulously covers data understanding, preparation, modeling, evaluation, and deployment phases. The dataset, consisting of images from diverse devices and annotated with bounding boxes, was instrumental in the training process. The model’s performance was evaluated using the mean average precision at a 50% intersection over union (mAP50), a metric that combines precision and recall. The results demonstrated that the model achieved a precision of 0.7 or higher for each herb, though recall values indicated potential over-detection, suggesting the need for database expansion and methodological enhancements. This research underscores the innovative potential of deep learning in aromatic plant identification and addresses both the challenges and advantages of this technique. The findings significantly advance the integration of artificial intelligence in agriculture, promoting greater efficiency and accuracy in plant identification.

The safety and robustness of convolutional neural networks (CNNs) have raised increasing concerns, especially in safety-critical areas, such as medical applications. Although CNNs are efficient in image classification, their predictions are often sensitive to minor, for human observers, invisible modifications of the image. Thus, a modified, corrupted image can be visually equal to the legitimate image for humans but fool the CNN and make a wrong prediction. Such modified images are called adversarial images throughout this paper. A popular method to generate adversarial images is backpropagating the loss gradient to modify the input image. Usually, only the direction of the gradient and a given step size were used to determine the perturbations (FGSM, fast gradient sign method), or the FGSM is applied multiple times to craft stronger perturbations that change the model classification (i-FGSM). On the contrary, if the step size is too large, the minimum perturbation of the image may be missed during the gradient search. To seek exact and minimal input images for a classification change, in this paper, we suggest starting the FGSM with a small step size and adapting the step size with iterations. A few decay algorithms were taken from the literature for comparison with a novel approach based on an index tracking the loss status. In total, three tracking functions were applied for comparison. The experiments show our loss adaptive decay algorithms could find adversaries with more than a 90% success rate while generating fewer perturbations to fool the CNNs.

Motivated by the growing threat of distributed denial-of-service (DDoS) attacks and the emergence of quantum computing, this study introduces a novel “quanvolutional autoencoder” architecture for learning representations. The architecture leverages the computational advantages of quantum mechanics to improve upon traditional machine learning techniques. Specifically, the quanvolutional autoencoder employs randomized quantum circuits to analyze time-series data from DDoS attacks, offering a robust alternative to classical convolutional neural networks. Experimental results suggest that the quanvolutional autoencoder performs similarly to classical models in visualizing and learning from DDoS hive plots and leads to faster convergence and learning stability. These findings suggest that quantum machine learning holds significant promise for advancing data analysis and visualization in cybersecurity. The study highlights the need for further research in this fast-growing field, particularly for unsupervised anomaly detection.

Background: Deep learning models have shown potential in improving cancer diagnosis and treatment. This study aimed to develop a convolutional neural network (CNN) model to predict Invasive Ductal Carcinoma (IDC), a common type of breast cancer. Additionally, a user-friendly interface was designed to facilitate the use of the model by healthcare professionals. Methods: The CNN model was trained and tested using a dataset of high-resolution microscopic images derived from 162 whole-mount slide images of breast cancer specimens. These images were meticulously scanned at 40× magnification using a state-of-the-art digital slide scanner to capture detailed information. Each image was then divided into 277,524 patches of 50 × 50 pixels, resulting in a diverse dataset containing 198,738 IDC-negative and 78,786 IDC-positive patches. Results: The model achieved an accuracy of 98.24% in distinguishing between benign and malignant cases, demonstrating its effectiveness in cancer detection. Conclusions: This study suggests that the developed CNN model has promising potential for clinical applications in breast cancer diagnosis and personalized treatment strategies. Our study further emphasizes the importance of accurate and reliable cancer detection methods for timely diagnosis and treatment. This study establishes a foundation for utilizing deep learning models in future cancer treatment research by demonstrating their effectiveness in analyzing large and complex datasets. This approach opens exciting avenues for further research and potentially improves our understanding of cancer and its treatment.

Brain tumors necessitate swift detection and classification for optimal patient outcomes. Deep learning has been extensively utilized to recognize complex tumor patterns in magnetic resonance imaging (MRI) images, aiding in tumor diagnosis, treatment, and prognostication. However, model complexity and limited generalizability with unfamiliar data hinder appropriate clinical integration. The objective of this study is to develop a clean-energy cloud-based deep learning platform to classify brain tumors. Three datasets of a total of 2611 axial MRI images were used to train our multi-layer convolutional neural network (CNN). Our platform automatically optimized every transfer learning and data augmentation feature combination to provide the highest predictive accuracy for our classification task. Our proposed system identified and classified brain tumors successfully and efficiently with an overall precision value of 96.8% [95% CI; 93.8–97.6]. Using clean energy supercomputing resources and cloud platforms cut our workflow to 103 min, $0 in total cost, and a negligible carbon footprint (0.0014 kg eq CO2). By leveraging automated optimized learning, we developed a cost-effective deep learning (DL) platform that accurately classified brain tumors from axial MRI images of different levels. Although studies have identified machine learning tools to overcome these obstacles, only some are cost-effective, generalizable, and usable regardless of experience.

Background: Accurate detection of axillary lymph node (ALN) metastases in breast cancer is crucial for clinical staging and treatment planning. This study aims to develop a deep learning model using clinical implication-applied preprocessed computed tomography (CT) images to enhance the prediction of ALN metastasis in breast cancer patients. Methods: A total of 1128 axial CT images of ALN (538 malignant and 590 benign lymph nodes) were collected from 523 breast cancer patients who underwent preoperative CT scans between January 2012 and July 2022 at Hallym University Medical Center. To develop an optimal deep learning model for distinguishing metastatic ALN from benign ALN, a CT image preprocessing protocol with clinical implications and two different cropping methods (fixed size crop [FSC] method and adjustable square crop [ASC] method) were employed. The images were analyzed using three different convolutional neural network (CNN) architectures (ResNet, DenseNet, and EfficientNet). Ensemble methods involving and combining the selection of the two best-performing CNN architectures from each cropping method were applied to generate the final result. Results: For the two different cropping methods, DenseNet consistently outperformed ResNet and EfficientNet. The area under the receiver operating characteristic curve (AUROC) for DenseNet, using the FSC and ASC methods, was 0.934 and 0.939, respectively. The ensemble model, which combines the performance of the DenseNet121 architecture for both cropping methods, delivered outstanding results with an AUROC of 0.968, an accuracy of 0.938, a sensitivity of 0.980, and a specificity of 0.903. Furthermore, distinct trends observed in gradient-weighted class activation mapping images with the two cropping methods suggest that our deep learning model not only evaluates the lymph node itself, but also distinguishes subtler changes in lymph node margin and adjacent soft tissue, which often elude human interpretation. Conclusions: This research demonstrates the promising performance of a deep learning model in accurately detecting malignant ALNs in breast cancer patients using CT images. The integration of clinical considerations into image processing and the utilization of ensemble methods further improved diagnostic precision.

The early and precise identification of the different phenological stages of the bean (Phaseolus vulgaris L.) allows for the determination of critical and timely moments for the implementation of certain agricultural activities that contribute in a significant manner to the output and quality of the harvest, as well as the necessary actions to prevent and control possible damage caused by plagues and diseases. Overall, the standard procedure for phenological identification is conducted by the farmer. This can lead to the possibility of overlooking important findings during the phenological development of the plant, which could result in the appearance of plagues and diseases. In recent years, deep learning (DL) methods have been used to analyze crop behavior and minimize risk in agricultural decision making. One of the most used DL methods in image processing is the convolutional neural network (CNN) due to its high capacity for learning relevant features and recognizing objects in images. In this article, a transfer learning approach and a data augmentation method were applied. A station equipped with RGB cameras was used to gather data from images during the complete phenological cycle of the bean. The information gathered was used to create a set of data to evaluate the performance of each of the four proposed network models: AlexNet, VGG19, SqueezeNet, and GoogleNet. The metrics used were accuracy, precision, sensitivity, specificity, and F1-Score. The results of the best architecture obtained in the validation were those of GoogleNet, which obtained 96.71% accuracy, 96.81% precision, 95.77% sensitivity, 98.73% specificity, and 96.25% F1-Score.

Deep neural networks have demonstrated outstanding performances in agriculture production. Agriculture production is one of the most important sectors because it has a direct impact on the economy and social life of any society. Plant disease identification is a big challenge for agriculture production, for which we need a fast and accurate technique to identify plant disease. With the recent advancement in deep learning, we can develop a robust and accurate system. This research investigated the use of deep learning for accurate and fast tomato plant disease identification. In this research, we have used individual and merged datasets of tomato plants with 10 diseases (including healthy plants). The main aim of this work is to check the accuracy of the existing convolutional neural network models such as Visual Geometry Group, Residual Net, and DenseNet on tomato plant disease detection and then design a custom deep neural network model to give the best accuracy in case of the tomato plant. We have trained and tested our models with datasets containing over 18,000 and 25,000 images with 10 classes. We achieved over 99% accuracy with our custom model. This high accuracy was achieved with less training time and lower computational cost compared to other CNNs. This research demonstrates the potential of deep learning for efficient and accurate tomato plant disease detection, which can benefit farmers and contribute to improved agricultural production. The custom model’s efficient performance makes it promising for practical implementation in real-world agricultural settings.

Alzheimer’s disease (AD) is a pressing global issue, demanding effective diagnostic approaches. This systematic review surveys the recent literature (2018 onwards) to illuminate the current landscape of AD detection via deep learning. Focusing on neuroimaging, this study explores single- and multi-modality investigations, delving into biomarkers, features, and preprocessing techniques. Various deep models, including convolutional neural networks (CNNs), recurrent neural networks (RNNs), and generative models, are evaluated for their AD detection performance. Challenges such as limited datasets and training procedures persist. Emphasis is placed on the need to differentiate AD from similar brain patterns, necessitating discriminative feature representations. This review highlights deep learning’s potential and limitations in AD detection, underscoring dataset importance. Future directions involve benchmark platform development for streamlined comparisons. In conclusion, while deep learning holds promise for accurate AD detection, refining models and methods is crucial to tackle challenges and enhance diagnostic precision.

Human behavior is regarded as one of the most complex notions present nowadays, due to the large magnitude of possibilities. These behaviors and actions can be distinguished as normal and abnormal. However, abnormal behavior is a vast spectrum, so in this work, abnormal behavior is regarded as human aggression or in another context when car accidents occur on the road. As this behavior can negatively affect the surrounding traffic participants, such as vehicles and other pedestrians, it is crucial to monitor such behavior. Given the current prevalent spread of cameras everywhere with different types, they can be used to classify and monitor such behavior. Accordingly, this work proposes a new optimized model based on a novel integrated wavelet-based channel augmentation unit for classifying human behavior in various scenes, having a total number of trainable parameters of 5.3 m with an average inference time of 0.09 s. The model has been trained and evaluated on four public datasets: Real Live Violence Situations (RLVS), Highway Incident Detection (HWID), Movie Fights, and Hockey Fights. The proposed technique achieved accuracies in the range of 92% to 99.5% across the used benchmark datasets. Comprehensive analysis and comparisons between different versions of the model and the state-of-the-art have been performed to confirm the model’s performance in terms of accuracy and efficiency. The proposed model has higher accuracy with an average of 4.97%, and higher efficiency by reducing the number of parameters by around 139.1 m compared to other models trained and tested on the same benchmark datasets.

Chest X-ray imaging plays a vital and indispensable role in the diagnosis of lungs, enabling healthcare professionals to swiftly and accurately identify lung abnormalities. Deep learning (DL) approaches have attained popularity in recent years and have shown promising results in automated medical image analysis, particularly in the field of chest radiology. This paper presents a novel DL framework specifically designed for the multi-class diagnosis of lung diseases, including fibrosis, opacity, tuberculosis, normal, viral pneumonia, and COVID-19 pneumonia, using chest X-ray images, aiming to address the need for efficient and accessible diagnostic tools. The framework employs a convolutional neural network (CNN) architecture with custom blocks to enhance the feature maps designed to learn discriminative features from chest X-ray images. The proposed DL framework is evaluated on a large-scale dataset, demonstrating superior performance in the multi-class diagnosis of the lung. In order to evaluate the effectiveness of the presented approach, thorough experiments are conducted against pre-existing state-of-the-art methods, revealing significant accuracy, sensitivity, and specificity improvements. The findings of the study showcased remarkable accuracy, achieving 98.88%. The performance metrics for precision, recall, F1-score, and Area Under the Curve (AUC) averaged 0.9870, 0.9904, 0.9887, and 0.9939 across the six-class categorization system. This research contributes to the field of medical imaging and provides a foundation for future advancements in DL-based diagnostic systems for lung diseases.

The primary strategy for mitigating lost productivity entails promptly, accurately, and efficiently detecting plant pests. Although detection by humans can be useful in detecting certain pests, it is often slower compared to automated methods, such as machine learning. Hence, this study employs a Convolutional Neural Network (CNN) model, specifically GoogleNet, to detect pests within mobile applications. The technique of detection involves the input of images depicting plant pests, which are subsequently subjected to further processing. This study employed many experimental methods to determine the most effective model. The model exhibiting a 93.78% accuracy stands out as the most superior model within the scope of this investigation. The aforementioned model has been included in a smartphone application with the purpose of facilitating Indonesian farmers in the identification of pests affecting their crops. The implementation of an Indonesian language application is a contribution to this research. Using this local language makes it easier for Indonesian farmers to use it. The potential impact of this application on Indonesian farmers is anticipated to be significant. By enhancing pest identification capabilities, farmers may employ more suitable pest management strategies, leading to improved crop yields in the long run.

During the 1950s, the Gros Michel species of bananas were nearly wiped out by the incurable Fusarium Wilt, also known as Panama Disease. Originating in Southeast Asia, Fusarium Wilt is a banana pandemic that has been threatening the multi-billion-dollar banana industry worldwide. The disease is caused by a fungus that spreads rapidly throughout the soil and into the roots of banana plants. Currently, the only way to stop the spread of this disease is for farmers to manually inspect and remove infected plants as quickly as possible, which is a time-consuming process. The main purpose of this study is to build a deep Convolutional Neural Network (CNN) using a transfer learning approach to rapidly identify Fusarium wilt infections on banana crop leaves. We chose to use the ResNet50 architecture as the base CNN model for our transfer learning approach owing to its remarkable performance in image classification, which was demonstrated through its victory in the ImageNet competition. After its initial training and fine-tuning on a data set consisting of 600 healthy and diseased images, the CNN model achieved near-perfect accuracy of 0.99 along with a loss of 0.46 and was fine-tuned to adapt the ResNet base model. ResNet50’s distinctive residual block structure could be the reason behind these results. To evaluate this CNN model, 500 test images, consisting of 250 diseased and healthy banana leaf images, were classified by the model. The deep CNN model was able to achieve an accuracy of 0.98 and an F-1 score of 0.98 by correctly identifying the class of 492 of the 500 images. These results show that this DCNN model outperforms existing models such as Sangeetha et al., 2023’s deep CNN model by at least 0.07 in accuracy and is a viable option for identifying Fusarium Wilt in banana crops.

Early rice disease detection is vital in preventing damage to agricultural product output and quantity in the agricultural field. Manual observations of rice diseases are tedious, costly, and time-consuming, especially when classifying disease patterns and color while dealing with non-native diseases. Hence, image processing and Machine Learning (ML) techniques are used to detect rice disease early and within a relatively brief time period. This article aims to demonstrate the performance of different ML algorithms in rice disease detection through image processing. We critically examined different techniques, and the outcomes of previous research have been reviewed to compare the performance of rice disease classifications. Performance has been evaluated based on the criteria of feature extraction, clustering, segmentation, noise reduction, and level of accuracy of disease detection techniques. This paper also showcases various algorithms for different datasets in terms of training methods, image preprocessing with clustering and filtering criteria, and testing with reliable outcomes. Through this review, we provide valuable insights into the current state of ML-based approaches for the early detection of rice diseases, and assist future research and improvement. In addition, we discuss several challenges that must be overcome in order to achieve effective identification of rice diseases.

The proliferation of Android-based devices has brought about an unprecedented surge in mobile application usage, making the Android ecosystem a prime target for cybercriminals. In this paper, a new method for Android malware classification is proposed. The method implements a convolutional neural network for malware classification using images. The research presents a novel approach to transforming the Android Application Package (APK) into a grayscale image. The image creation utilizes natural language processing techniques for text cleaning, extraction, and fuzzy hashing to represent the decompiled code from the APK in a set of hashes after preprocessing, where the image is composed of n fuzzy hashes that represent an APK. The method was tested on an Android malware dataset with 15,493 samples of five malware types. The proposed method showed an increase in accuracy compared to others in the literature, achieving up to 98.24% in the classification task.

The widespread use of electric vehicles necessitates meticulous planning for the placement of charging stations (CSs) in already crowded cities so that they can efficiently meet the charging demand while adhering to various real-world constraints such as the total budget, queuing time, electrical regulations, etc. Many classical and metaheuristic-based approaches provide good solutions, but they are not intuitive, and they do not scale well for large cities and complex constraints. Many classical solution techniques often require prohibitive amounts of memory and their solutions are not easily explainable. We analyzed the layouts of the 50 most populous cities of the world and observed that any city can be represented as a composition of five basic primitive shapes (stretched to different extents). Based on this insight, we use results from classical topology to design a new charging station placement algorithm. The first step is a topological clustering algorithm to partition a large city into small clusters and then use precomputed solutions for each basic shape to arrive at a solution for each cluster. These cluster-level solutions are very intuitive and explainable. Then, the next step is to combine the small solutions to arrive at a full solution to the problem. Here, we use a surrogate function and repair-based technique to fix any resultant constraint violations (after all the solutions are combined). The third step is optional, where we show that the second step can be extended to incorporate complex constraints and secondary objective functions. Along with creating a full software suite, we perform an extensive evaluation of the top 50 cities and demonstrate that our method is not only 30 times faster but its solution quality is also 36.62% better than the gold standard in this area—an integer linear programming (ILP) approach with a practical timeout limit.

Underwater acoustic target recognition (UATR) is crucial for defense and ocean environment monitoring. Although traditional methods and deep learning approaches based on time–frequency domain features have achieved high recognition rates in certain tasks, they rely on manually designed feature extraction processes, leading to information loss and limited adaptability to environmental changes. To overcome these limitations, we proposed a novel end-to-end underwater acoustic target recognition model, 1DCTN. This model directly used raw time-domain signals as input, leveraging one-dimensional convolutional neural networks (1D CNNs) to extract local features and combining them with Transformers to capture global dependencies. Our model simplified the recognition process by eliminating the need for complex feature engineering and effectively addressed the limitations of LSTM in handling long-term dependencies. Experimental results on the publicly available ShipsEar dataset demonstrated that 1DCTN achieves a remarkable accuracy of 96.84%, setting a new benchmark for end-to-end models on this dataset. Additionally, 1DCTN stood out among lightweight models, achieving the highest recognition rate, making it a promising direction for future research in underwater acoustic recognition.

Rare-earth magnesium alloys exhibit higher comprehensive mechanical properties compared to other series of magnesium alloys, effectively expanding their applications in aerospace, weapons, and other fields. In this work, the tensile strength, yield strength, and elongation of a Mg-Gd-Y-Zn-Zr rare-earth magnesium alloy under different process conditions were determined, and a large number of microstructure observations and analyses were carried out for the tensile specimens; a prediction model of the corresponding mechanical properties was established by using a convolutional neural network (CNN), in which the metallographic diagram of the rare-earth magnesium alloy was taken as the input, and the corresponding tensile strength, yield strength, elongation, and three mechanical properties were taken as the output. The stochastic gradient descent (SGD) algorithm was used for parameter optimization and experimental validation, and the results showed that the average relative errors of the tensile strength and yield strength prediction results were 1.90% and 3.14%, respectively, which were smaller than the expected error of 5%.

The morphology of particles formed in different environments contains critical information. Thus, the rapid and effective reconstruction of their three-dimensional (3D) morphology is crucial. This study reconstructs the 3D morphology from two-dimensional (2D) images of particles using artificial intelligence (AI). More than 100,000 particles were sampled from three sources: naturally formed particles (desert sand), manufactured particles (lunar soil simulant), and numerically generated digital particles. A deep learning approach based on a voxel representation of the morphology and multi-dimensional convolutional neural networks was proposed to rapidly upscale and reconstruct particle morphology. The trained model was tested using the three particle types and evaluated using different multi-scale morphological descriptors. The results demonstrated that the statistical properties of the morphological descriptors were consistent for the real 3D particles and those derived from the 2D images and the model. This finding confirms the model’s validity and generalizability in upscaling and reconstructing diverse particle samples. This study provides a method for generating 3D numerical representations of geological particles, facilitating in-depth analysis of properties, such as mechanical behavior and transport characteristics, from 2D images.

Based on the perspective of identifying the water army, this paper uses the methods of machine learning and data visualization to analyze the helpfulness of online produce reviews, portray product portraits, and provide real and helpful product reviews. In order to identify and eliminate the water army, the Term Frequency-Inverse Document Frequency Model (TF-IDF) and Latent Semantic Index Model (LSI) are used. After eliminating the water army, three classification methods were selected to perform sentimental analysis, including logistics, SnowNLP, and Convolutional Neural Network for text(TextCNN). The TextCNN has the highest F1 score among the three classification methods. At the same time, the Latent Dirichlet Allocation mode (LDA) is used to extract the topics of various reviews. Finally, targeted countermeasures are proposed to manufacturers, consumers, and regulators.

A robust biometric system is essential to mitigate various security threats. Electroencephalography (EEG) brain signals present a promising alternative to other biometric traits due to their sensitivity, non-duplicability, resistance to theft, and individual-specific dynamics. However, existing EEG-based biometric systems employ deep neural networks, such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs), which face challenges such as high parameter complexity, limiting their practical application. Additionally, their ability to generalize across a large number of subjects remains unclear. Moreover, they have been validated on datasets collected in controlled environments, which do not accurately reflect real-world scenarios involving diverse brain conditions. To overcome these challenges, we propose a lightweight neural network model, GCT–EEGNet, which is based on the design ideas of a CNN model and incorporates an attention mechanism to pay attention to the appropriate frequency bands for extracting discriminative features relevant to the identity of a subject despite diverse brain conditions. First, a raw EEG signal is decomposed into frequency bands and then passed to GCT–EEGNet for feature extraction, which utilizes a gated channel transformation (GCT) layer to selectively emphasize informative features from the relevant frequency bands. The extracted features were used for subject recognition through a cosine similarity metric that measured the similarity between feature vectors of different EEG trials to identify individuals. The proposed method was evaluated on a large dataset comprising 263 subjects. The experimental results demonstrated that the method achieved a correct recognition rate (CRR) of 99.23% and an equal error rate (EER) of 0.0014, corroborating its robustness against different brain conditions. The proposed model maintains low parameter complexity while keeping the expressiveness of representations, even with unseen subjects.

Semantic segmentation forms the foundation for understanding very high resolution (VHR) remote sensing images, with extensive demand and practical application value. The convolutional neural networks (CNNs), known for their prowess in hierarchical feature representation, have dominated the field of semantic image segmentation. Recently, hierarchical vision transformers such as Swin have also shown excellent performance for semantic segmentation tasks. However, the hierarchical structure enlarges the receptive field to accumulate features and inevitably leads to the blurring of object boundaries. We introduce a novel object-aware network, Embedding deep SuperPixel, for VHR image semantic segmentation called ESPNet, which integrates advanced ConvNeXt and the learnable superpixel algorithm. Specifically, the developed task-oriented superpixel generation module can refine the results of the semantic segmentation branch by preserving object boundaries. This study reveals the capability of utilizing deep convolutional neural networks to accomplish both superpixel generation and semantic segmentation of VHR images within an integrated end-to-end framework. The proposed method achieved mIoU scores of 84.32, 90.13, and 55.73 on the Vaihingen, Potsdam, and LoveDA datasets, respectively. These results indicate that our model surpasses the current advanced methods, thus demonstrating the effectiveness of the proposed scheme.

The present work proposes a new model based on a convolutional neural network (CNN) to retrieve solar shortwave (SW) irradiance via the estimation of the cloud modification factor (CMF) from daytime sky images captured by all-sky cameras; this model is named CNN-CMF. To this end, a total of 237,669 sky images paired with SW irradiance measurements obtained by using pyranometers were selected at the following three sites: Valladolid and Izaña, Spain, and Lindenberg, Germany. This dataset was randomly split into training and testing sets, with the latter excluded from the training model in order to validate it using the same locations. Subsequently, the test dataset was compared with the corresponding SW irradiance measurements obtained by the pyranometers in scatter density plots. The linear fit shows a high determination coefficient (R2) of 0.99. Statistical analyses based on the mean bias error (MBE) values and the standard deviation (SD) of the SW irradiance differences yield results close to −2% and 9%, respectively. The MBE indicates a slight underestimation of the CNN-CMF model compared to the measurement values. After its validation, model performance was evaluated at the Antarctic station of Marambio (Argentina), a location not used in the training process. A similar comparison between the model-predicted SW irradiance and pyranometer measurements yielded R2=0.95, with an MBE of around 2% and an SD of approximately 26%. Although the precision provided by the SD at the Marambio station is lower, the MBE shows that the model’s accuracy is similar to previous results but with a slight overestimation of the SW irradiance. Finally, the determination coefficient improved to 0.99, and the MBE and SD are about 3% and 11%, respectively, when the CNN-CMF model is used to estimate daily SW irradiation values.

Millimeter-wave radars are widely used in various environments due to their excellent detection capabilities. However, the detection performance in severe weather environments is still an important research challenge. In this paper, the propagation characteristics of millimeter-wave radar in a rainfall environment are thoroughly investigated, and the modeling of the millimeter-wave radar echo signal in a rainfall environment is completed. The effect of rainfall on radar detection performance is verified through experiments, and an anti-rain clutter interference method based on a convolutional neural network is proposed. The method combines image recognition and classification techniques to effectively distinguish target signals from rain clutter in radar echo signals based on feature differences. In addition, this paper compares the recognition results of the proposed method with VGGnet and Resnet. The experimental results show that the proposed convolutional neural network method significantly improves the target detection capability of the radar system in a rainfall environment, verifying the method’s effectiveness and accuracy. This study provides a new solution for the application of millimeter-wave radar in severe weather conditions.

To address the challenges of sparse point clouds in current MIMO millimeter-wave radar environmental mapping, this paper proposes a dense 3D millimeter-wave radar point cloud environmental mapping algorithm. In the preprocessing phase, a radar SLAM-based approach is introduced to construct local submaps, which replaces the direct use of radar point cloud frames. This not only reduces data dimensionality but also enables the proposed method to handle scenarios involving vehicle motion with varying speeds. Building on this, a 3D-RadarHR cross-modal learning network is proposed, which uses LiDAR as the target output to train the radar submaps, thereby generating a dense millimeter-wave radar point cloud map. Experimental results across multiple scenarios, including outdoor environments and underground tunnels, demonstrate that the proposed method can increase the point cloud density of millimeter-wave radar environmental maps by over 50 times, with a point cloud accuracy better than 0.1 m. Compared to existing algorithms, the proposed method achieves superior environmental map reconstruction performance while maintaining a real-time processing rate of 15 Hz.

In clinical settings, computed tomography (CT), magnetic resonance imaging (MRI), or positron emission tomography (PET) are commonly employed in brain imaging to assist clinicians in determining the type of stroke in patients. However, these modalities are associated with potential hazards or limitations. In contrast, microwave imaging emerges as a promising technique, offering advantages such as non-ionizing radiation, low cost, lightweight, and portability. The primary challenges faced by microwave tomography include the severe ill-posedness of the electromagnetic inverse scattering problem and the time-consuming nature and unsatisfactory resolution of iterative quantitative algorithms. This paper proposes a learning electric field enhancement imaging method (LEFEIM) to achieve quantitative brain imaging based on a microwave tomography system. LEFEIM comprises two cascaded networks. The first, based on a convolutional neural network, utilizes the electric field from the receiving antenna to predict the electric field distribution within the imaging domain. The second network employs the electric field distribution as input to learn the dielectric constant distribution, thereby realizing quantitative brain imaging. Compared to the Born Iterative Method (BIM), LEFEIM significantly improves imaging time, while enhancing imaging quality and goodness-of-fit to a certain extent. Simultaneously, LEFEIM exhibits anti-noise capabilities.

Cloud computing has recently gained widespread attention owing to its use in applications involving the Internet of Things (IoT). However, the transmission of massive volumes of data to a cloud server often results in overhead. Fog computing has emerged as a viable solution to address this issue. This study implements an Artificial Intelligence of Things (AIoT) system based on fog computing on a smart farm. Three experiments are conducted to evaluate the performance of the AIoT system. First, network traffic volumes between systems employing and not employing fog computing are compared. Second, the performance of the communication protocols—hypertext transport protocol (HTTP), message queuing telemetry transport protocol (MQTT), and constrained application protocol (CoAP)—commonly used in IoT applications is assessed. Finally, a convolutional neural network-based algorithm is introduced to determine the maturity level of coffee tree images. Experimental data are collected over ten days from a coffee tree farm in the Republic of Korea. Notably, the fog computing system demonstrates a 26% reduction in the cumulative data volume compared with a non-fog system. MQTT exhibits stable results in terms of the data volume and loss rate. Additionally, the maturity level determination algorithm performed on coffee fruits provides reliable results.

In the field of gesture recognition technology, accurately detecting human gestures is crucial. In this research, ultrasonic transducers were utilized for gesture recognition. Due to the wide beamwidth of ultrasonic transducers, it is difficult to effectively distinguish between multiple objects within a single beam. However, they are effective at accurately identifying individual objects. To leverage this characteristic of the ultrasonic transducer as an advantage, this research involved constructing an ultrasonic array. This array was created by arranging eight transmitting transducers in a circular formation and placing a single receiving transducer at the center. Through this, a wide beam area was formed extensively, enabling the measurement of unrestricted movement of a single hand in the X, Y, and Z axes. Hand gesture data were collected at distances of 10 cm, 30 cm, 50 cm, 70 cm, and 90 cm from the array. The collected data were trained and tested using a customized Convolutional Neural Network (CNN) model, demonstrating high accuracy on raw data, which is most suitable for immediate interaction with computers. The proposed system achieved over 98% accuracy.

In order to solve the issue of low efficiency in retired battery clustering, a method for quickly obtaining a charging curve and Incremental Capacity (IC) curve based on Convolutional Neural Networks (CNN) is proposed. By training a CNN model, the method enables accurate prediction of complete IC curves and V-Q curves from local charging curves starting at any beginning. The prediction accuracy was validated using the Oxford battery degradation dataset, and transfer learning was conducted by fine-tuning the model trained on LCO batteries for use with LFP batteries, which reduced the RMSE of the estimation and validated the generalizability of the model. Peak parameters were extracted from both the original and predicted IC curves for clustering, and the t-test was applied to eliminate outliers, which significantly reduced the time required to obtain clustering features and improved clustering efficiency.

Mapping and monitoring crops are the most complex and difficult tasks for experts processing and analyzing remote sensing (RS) images. Classifying crops using RS images is the most expensive task, and it requires intensive labor, especially in the sample collection phase. Fieldwork requires periodic visits to collect data about the crop’s physiochemical characteristics and separating them using the known conventional machine learning algorithms and remote sensing images. As the problem becomes more complex because of the diversity of crop types and the increase in area size, sample collection becomes more complex and unreliable. To avoid these problems, a new segmentation model was created that does not require sample collection or high-resolution images and can successfully distinguish wheat from other crops. Moreover, UNet is a well-known Convolutional Neural Network (CNN), and the semantic method was adjusted to become more powerful, faster, and use fewer resources. The new model was named Fast-UNet and was used to improve the segmentation of wheat crops. Fast-UNet was compared to UNet and Google’s newly developed semantic segmentation model, DeepLabV3+. The new model was faster than the compared models, and it had the highest average accuracy compared to UNet and DeepLabV3+, with values of 93.45, 93.05, and 92.56 respectively. Finally, new datasets of time series NDVI images and ground truth data were created. These datasets, and the newly developed model, were made available publicly on the Web.

Rice spikelet size, spikelet length and spikelet width, are very important traits directly related to a rice crop’s yield. The accurate measurement of these parameters is quite significant in research such as breeding, yield evaluation and variety improvement for rice crops. Traditional measurement methods still mainly rely on manual labor, which is time-consuming, labor-intensive and error-prone. In this study, a novel method, dubbed the “SSM-Method”, based on convolutional neural network and traditional image processing technology has been developed for the efficient and precise measurement of rice spikelet size parameters on rice panicle structures. Firstly, primary branch images of rice panicles were collected at the same height to build an image database. The spikelet detection model using convolutional neural network was then established for spikelet recognition and localization. Subsequently, the calibration value was obtained through traditional image processing technology. Finally, the “SSM-Method” integrated with a spikelet detection model and calibration value was developed for the automatic measurement of spikelet sizes. The performance of the developed SSM-Method was evaluated through testing 60 primary branch images. The test results showed that the root mean square error (RMSE) of spikelet length for two rice varieties (Huahang15 and Qingyang) were 0.26 mm and 0.30 mm, respectively, while the corresponding RMSE of spikelet width was 0.27 mm and 0.31 mm, respectively. The proposed algorithm can provide an effective, convenient and low-cost tool for yield evaluation and breeding research.

Gender recognition is an important part of the duck industry. Currently, the gender identification of ducks mainly relies on manual labor, which is highly labor-intensive. This study aims to propose a novel method for distinguishing between males and females based on the characteristic sound parameters for day-old ducks. The effective data from the sounds of day-old ducks were recorded and extracted using the endpoint detection method. The 12-dimensional Mel-frequency cepstral coefficients (MFCCs) with first-order and second-order difference coefficients in the effective sound signals of the ducks were calculated, and a total of 36-dimensional feature vectors were obtained. These data were used as input information to train three classification models, include a backpropagation neural network (BPNN), a deep neural network (DNN), and a convolutional neural network (CNN). The training results show that the accuracies of the BPNN, DNN, and CNN were 83.87%, 83.94%, and 84.15%, respectively, and that the three classification models could identify the sounds of male and female ducks. The prediction results showed that the prediction accuracies of the BPNN, DNN, and CNN were 93.33%, 91.67%, and 95.0%, respectively, which shows that the scheme for distinguishing between male and female ducks via sound had high accuracy. Moreover, the CNN demonstrated the best recognition effect. The method proposed in this study can provide some support for developing an efficient technique for gender identification in duck production.

In the field of hyperspectral image classification, deep learning technology, especially convolutional neural networks, has achieved remarkable progress. However, convolutional neural network models encounter challenges in hyperspectral image classification due to limitations in their receptive fields. Conversely, the global modeling capability of Transformers has garnered attention in hyperspectral image classification. Nevertheless, the high computational cost and inadequate local feature extraction hinder its widespread application. In this study, we propose a novel fusion model of convolutional neural networks and Transformers to enhance performance in hyperspectral image classification, namely the dual-branch multi-granularity convolutional cross-substitution Transformer (DMCCT). The proposed model adopts a dual-branch structure to separately extract spatial and spectral features, thereby mitigating mutual interference and information loss between spectral and spatial data during feature extraction. Moreover, a multi-granularity embedding module is introduced to facilitate multi-scale and multi-level local feature extraction for spatial and spectral information. In particular, the improved convolutional cross-substitution Transformer module effectively integrates convolution and Transformer, reducing the complexity of attention operations and enhancing the accuracy of hyperspectral image classification tasks. Subsequently, the proposed method is evaluated against existing approaches using three classical datasets, namely Pavia University, Kennedy Space Center, and Indian Pines. Experimental results demonstrate the efficacy of the proposed method, achieving significant classification results on these datasets with overall classification accuracies of 98.57%, 97.96%, and 96.59%, respectively. These results establish the superiority of the proposed method in the context of hyperspectral image classification under similar experimental conditions.

This paper presents a novel multi-initialization model for recognizing road surface damage, e.g. potholes and cracks, on video using convolutional neural networks (CNNs) in real-time for fast damage recognition. The model is trained by the latest Road Damage Detection dataset, which includes four types of road damage. In addition, the CNN model is updated using pseudo-labeled images from semi-learned methods to improve the performance of the pavement damage detection technique. This study describes the use of the YOLO architecture and optimizes it according to the selected parameters, demonstrating high efficiency and accuracy. The results obtained can enhance the safety and efficiency of road pavement and, hence, its traffic quality and contribute to decision-making for the maintenance and restoration of road infrastructure.

Computer tomography (CT) scans’ capabilities in detecting lesions have been increasing remarkably in the past decades. In this paper, we propose a multi-organ lesion detection (MOLD) approach to better address real-life chest-related clinical needs. MOLD is a challenging task, especially within a large, high resolution image volume, due to various types of background information interference and large differences in lesion sizes. Furthermore, the appearance similarity between lesions and other normal tissues demands more discriminative features. In order to overcome these challenges, we introduce depth-aware (DA) and skipped-layer hierarchical training (SHT) mechanisms with the novel Dense 3D context enhanced (Dense 3DCE) lesion detection model. The novel Dense 3DCE framework considers the shallow, medium, and deep-level features together comprehensively. In addition, equipped with our SHT scheme, the backpropagation process can now be supervised under precise control, while the DA scheme can effectively incorporate depth domain knowledge into the scheme. Extensive experiments have been carried out on a publicly available, widely used DeepLesion dataset, and the results prove the effectiveness of our DA-SHT Dense 3DCE network in the MOLD task.

Diagnosing atypical pigmented facial lesions (aPFLs) is a challenging topic for dermatologists. Accurate diagnosis of these lesions is crucial for effective patient management, especially in dermatology, where visual assessment plays a central role. Incorrect diagnoses can result in mismanagement, delays in appropriate interventions, and potential harm. AI, however, holds the potential to enhance diagnostic accuracy and provide reliable support to clinicians. This work aimed to evaluate and compare the effectiveness of machine learning (logistic regression of lesion features and patient metadata) and deep learning (CNN analysis of images) models in dermoscopy diagnosis and the management of aPFLs. This study involved the analysis of 1197 dermoscopic images of facial lesions excised due to suspicious and histologically confirmed malignancy, classified into seven classes (lentigo maligna—LM; lentigo maligna melanoma—LMM; atypical nevi—AN; pigmented actinic keratosis—PAK; solar lentigo—SL; seborrheic keratosis—SK; and seborrheic lichenoid keratosis—SLK). Image samples were collected through the Integrated Dermoscopy Score (iDScore) project. The statistical analysis of the dataset shows that the patients mean age was 65.5 ± 14.2, and the gender was equally distributed (580 males—48.5%; 617 females—51.5%). A total of 41.7% of the sample constituted malignant lesions (LM and LMM). Meanwhile, the benign lesions were mainly PAK (19.3%), followed by SL (22.2%), AN (10.4%), SK (4.0%), and SLK (2.3%). The lesions were mainly localised in the cheek and nose areas. A stratified analysis of the assessment provided by the enrolled dermatologists was also performed, resulting in 2445 evaluations of the 1197 images (2.1 evaluations per image on average). The physicians demonstrated higher accuracy in differentiating between malignant and benign lesions (71.2%) than in distinguishing between the seven specific diagnoses across all the images (42.9%). The logistic regression model obtained a precision of 39.1%, a sensitivity of 100%, a specificity of 33.9%, and an accuracy of 53.6% on the test set, while the CNN model showed lower sensitivity (58.2%) and higher precision (47.0%), specificity (90.8%), and accuracy (59.5%) for melanoma diagnosis. This research demonstrates how AI can enhance the diagnostic accuracy in complex dermatological cases like aPFLs by integrating AI models with clinical data and evaluating different diagnostic approaches, paving the way for more precise and scalable AI applications in dermatology, showing their critical role in improving patient management and the outcomes in dermatology.

Objectives: We aimed to study classical, publicly available convolutional neural networks (3D-CNNs) using a combination of several cine-MR orientation planes for the estimation of left ventricular ejection fraction (LVEF) without contour tracing. Methods: Cine-MR examinations carried out on 1082 patients from our institution were analysed by comparing the LVEF provided by the CVI42 software (V5.9.3) with the estimation resulting from different 3D-CNN models and various combinations of long- and short-axis orientation planes. Results: The 3D-Resnet18 architecture appeared to be the most favourable, and the results gradually and significantly improved as several long-axis and short-axis planes were combined. Simply pasting multiple orientation views into composite frames increased performance. Optimal results were obtained by pasting two long-axis views and six short-axis views. The best configuration provided an R2 = 0.83, a mean absolute error (MAE) = 4.97, and a root mean square error (RMSE) = 6.29; the area under the ROC curve (AUC) for the classification of LVEF < 40% was 0.99, and for the classification of LVEF > 60%, the AUC was 0.97. Internal validation performed on 149 additional patients after model training provided very similar results (MAE 4.98). External validation carried out on 62 patients from another institution showed an MAE of 6.59. Our results in this area are among the most promising obtained to date using CNNs with cardiac magnetic resonance. Conclusion: (1) The use of traditional 3D-CNNs and a combination of multiple orientation planes is capable of estimating LVEF from cine-MRI data without segmenting ventricular contours, with a reliability similar to that of traditional methods. (2) Performance significantly improves as the number of orientation planes increases, providing a more complete view of the left ventricle.

Background/Objectives: Head and neck cancer (HNC), predominantly squamous cell carcinoma (SCC), presents a significant global health burden. Conventional diagnostic approaches often face challenges in terms of achieving early detection and accurate diagnosis. This review examines recent advancements in hyperspectral imaging (HSI), integrated with computer-aided diagnostic (CAD) techniques, to enhance HNC detection and diagnosis. Methods: A systematic review of seven rigorously selected studies was performed. We focused on CAD algorithms, such as convolutional neural networks (CNNs), support vector machines (SVMs), and linear discriminant analysis (LDA). These are applicable to the hyperspectral imaging of HNC tissues. Results: The meta-analysis findings indicate that LDA surpasses other algorithms, achieving an accuracy of 92%, sensitivity of 91%, and specificity of 93%. CNNs exhibit moderate performance, with an accuracy of 82%, sensitivity of 77%, and specificity of 86%. SVMs demonstrate the lowest performance, with an accuracy of 76% and sensitivity of 48%, but maintain a high specificity level at 89%. Additionally, in vivo studies demonstrate superior performance when compared to ex vivo studies, reporting higher accuracy (81%), sensitivity (83%), and specificity (79%). Conclusion: Despite these promising findings, challenges persist, such as HSI’s sensitivity to external conditions, the need for high-resolution and high-speed imaging, and the lack of comprehensive spectral databases. Future research should emphasize dimensionality reduction techniques, the integration of multiple machine learning models, and the development of extensive spectral libraries to enhance HSI’s clinical utility in HNC diagnostics. This review underscores the transformative potential of HSI and CAD techniques in revolutionizing HNC diagnostics, facilitating more accurate and earlier detection, and improving patient outcomes.

The precision of robotic manipulators in the industrial or medical field is very important, especially when it comes to repetitive or exhaustive tasks. Geometric deformations are the most common in this field. For this reason, new robotic vision techniques have been proposed, including 3D methods that made it possible to determine the geometric distances between the parts of a robotic manipulator. The aim of this work is to measure the angular position of a robotic arm with six degrees of freedom. For this purpose, a stereo camera and a convolutional neural network algorithm are used to reduce the degradation of precision caused by geometric errors. This method is not intended to replace encoders, but to enhance accuracy by compensating for degradation through an intelligent visual measurement system. The camera is tested and the accuracy is about one millimeter. The implementation of this method leads to better results than traditional and simple neural network methods.

Glaucoma represents a significant global contributor to blindness. Accurately segmenting the optic disc (OD) and optic cup (OC) to obtain precise CDR is essential for effective screening. However, existing convolutional neural network (CNN)-based segmentation techniques are often limited by high computational demands and long inference times. This paper proposes an efficient end-to-end method for OD and OC segmentation, utilizing the lightweight MobileNetv3 network as the core feature-extraction module. Our approach combines boundary branches with adversarial learning, to achieve multi-label segmentation of the OD and OC. We validated our proposed approach across three public available datasets: Drishti-GS, RIM-ONE-r3, and REFUGE. The outcomes reveal that the Dice coefficients for the segmentation of OD and OC within these datasets are 0.974/0.900, 0.966/0.875, and 0.962/0.880, respectively. Additionally, our method substantially lowers computational complexity and inference time, thereby enabling efficient and precise segmentation of the optic disc and optic cup.

Real-world datasets often follow a long-tailed distribution, where a few majority (head) classes contain a large number of samples, while many minority (tail) classes contain significantly fewer samples. This imbalance creates an information disparity between head and tail classes, which can negatively impact the performance of deep networks. Some transfer knowledge techniques attempt to mitigate this gap by generating additional minority samples, either through oversampling the tail classes or transferring knowledge from the head classes to the tail classes. However, these methods often restrict the diversity of the generated minority samples, primarily focusing on transferring information only to the tail classes. This paper introduces a simple yet effective method for long-tailed classification, called mixed mutual transfer (MMT), which facilitates the mutual transfer of knowledge between head and tail classes by blending samples. The core idea of our method is to create new samples by blending head and tail samples. Head samples are selected using a uniform sampler that retains the long-tailed distribution, while tail samples are selected using a differential sampler that reverses the long-tailed distribution to alleviate imbalance. Our approach aims to diversify both tail and head classes. During the training phase, we utilize the generated samples to update the original dataset for training deep networks. Mixed mutual transfer simultaneously enhances the performance of both head and tail classes. Experimental results on various class-imbalanced datasets show that the proposed method significantly outperforms existing methods, demonstrating its effectiveness in improving the performance of long-tailed deep networks.

In extremely dark conditions, low-light imaging may offer spectators a rich visual experience, which is important for both military and civic applications. However, the images taken in ultra-micro light environments usually have inherent defects such as extremely low brightness and contrast, a high noise level, and serious loss of scene details and colors, which leads to great challenges in the research of low-light image and object detection and classification. The low-light night vision image used as the study object in this work has an excessively dim overall picture and very little information about the screen’s features. Three algorithms, HE, AHE, and CLAHE, were used to enhance and highlight the image. The effectiveness of these image enhancement methods is evaluated using metrics such as the peak signal-to-noise ratio and mean square error, and CLAHE was selected after comparison. The target image includes vehicles, people, license plates, and objects. The gray-level co-occurrence matrix (GLCM) was used to extract the texture features of the enhanced images, and the extracted image texture features were used as input to construct a backpropagation (BP) neural network classification model. Then, low-light image classification models were developed based on VGG16 and ResNet50 convolutional neural networks combined with low-light image enhancement algorithms. The experimental results show that the overall classification accuracy of the VGG16 convolutional neural network model is 92.1%. Compared with the BP and ResNet50 neural network models, the classification accuracy was increased by 4.5% and 2.3%, respectively, demonstrating its effectiveness in classifying low-light night vision targets.

Great advances in stitching high-quality retinal images have been made in recent years. On the other hand, very few studies have been carried out on low-resolution retinal imaging. This work investigates the challenges of low-resolution retinal images obtained by the D-EYE smartphone-based fundus camera. The proposed method uses homography estimation to register and stitch low-quality retinal images into a cohesive mosaic. First, a Siamese neural network extracts features from a pair of images, after which the correlation of their feature maps is computed. This correlation map is fed through four independent CNNs to estimate the homography parameters, each specializing in different corner coordinates. Our model was trained on a synthetic dataset generated from the Microsoft Common Objects in Context (MSCOCO) dataset; this work added an important data augmentation phase to improve the quality of the model. Then, the same is evaluated on the FIRE retina and D-EYE datasets for performance measurement using the Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM). The obtained results are promising: the average PSNR was 26.14 dB, with an SSIM of 0.96 on the D-EYE dataset. Compared to the method that uses a single neural network for homography calculations, our approach improves the PSNR by 7.96 dB and achieves a 7.86% higher SSIM score.