Narrow Results

Deep learning (DL)-based natural language processing (NLP) technologies are applied and optimized for the analysis of emotional implications in text data. Understanding emotional intricacies in text has become necessary for a variety of industries, like social media analysis, customer feedback interpretation, and mental health monitoring, due to the exponential rise of digital material. The study emphasizes the difficulties associated with the automation of emotion analysis, including the diversities of the text formats and the NLP complexities. This research gathers that text-based emotion analysis data and pre-processing techniques are utilized for the analysis of emotion, including normalizing methods such as tokenization and lemmatization, along with the removal of stop words and punctuation. The model incorporates pre-trained embeddings and attention mechanisms to capture contextual information and emotional subtleties for any language. The study proposes the hybrid model using tuned crow search optimized dynamic graph neural network (TCSO-DGNN) on an automatically optimized process to enhance the accuracy of emotion detection. The proposed model results on emotion analysis data and superior results are obtained, indicating better performance in precision (94.72%), accuracy (96.43%), recall (94.36%), and F1-score (93.02%). The findings indicate that the inclusion of TCSO-DGNN significantly enhances the identification of complex emotional expressions concerning traditional NLP techniques. This study explores the potential of integrated DL approaches to refine the capabilities of emotion analysis tools, particularly in contexts requiring refined interpretations of human emotions expressed online. Additionally, it contributes to academic discourse in the field of emotion analysis while providing practical insights for developers of NLP applications across business and public service sectors.

In this paper, we focus on denoising seismic signals effectively to obtain high-quality data, which is crucially important for oil-gas reservoir prediction and seismic interpretation tasks. The rapid progress of deep learning has brought new development opportunities to seismic oil and gas exploration technologies. However, current deep learning-based seismic denoising models have limited learning ability due to insufficient extraction features in strong noise backgrounds. For this reason, we develop a new multi-scale generative adversarial networks (GAN) with transform prediction for seismic signal denoising. First, we develop a deep multi-scale diversion fusion (MSDF) network in a GAN generator considering the advantages of combining GAN and convolutional neural networks. MSDF network primarily contains several MSDF blocks that mine abundant long-short path features in multiple receptive fields to restore seismic signals in a rough to detailed manner. Additionally, current deep learning-based seismic denoising models only exploit features in the spatial domain, not considering high-frequency characteristics, which results in insufficient high-frequency details when reconstructing seismic data. So, we propose to predict high-frequency components during learning by employing the superior non-subsampled contourlet transform (NSCT), which further preserves the better global topological structure and local texture characteristics of MSDF in GAN generator than the spatial domain, promoting the discrimination ability in GAN discriminator. The qualitative and quantitative results on our constructed synthetic dataset and actual seismic data demonstrate that the proposed method surpasses other deep learning-based approaches in realizing higher signal-to-noise ratio, as well as mining more effective high-frequency signals.

Computed Tomography (CT) offers great visualization of the intricate internal body structures. To protect a patient from the potential radiation-related health risks, the acquisition of CT images should adhere to the “as low as reasonably allowed” (ALARA) standard. However, the acquired Low-dose CT (LDCT) images are inadvertently corrupted by artifacts and noise during the processes of acquisition, storage, and transmission, degrading the visual quality of the image and also causing the loss of image features and relevant information. Most recently, generative adversarial network (GAN) models based on deep learning (DL) have demonstrated ground-breaking performance to minimize image noise while maintaining high image quality. These models’ ability to adapt to uncertain noise distributions and representation-learning ability makes them highly desirable for the denoising of CT images. The state-of-the-art GANs used for LDCT image denoising have been comprehensively reviewed in this research paper. The aim of this paper is to highlight the potential of DL-based GAN for CT dose optimization and present future scope of research in the domain of LDCT image denoising.

A neuro-degenerative eye disease called Glaucoma is expanded due to the development of ocular pressure inside the retina. Worldwide, glaucoma is the second widest reason for increased vision loss. Moreover, glaucoma affects millions of people, and early intervention is crucial to prevent irreversible vision loss. The absence of an early diagnosis causes complete blindness to the individuals. Furthermore, screening entire populations for glaucoma is impractical due to its low prevalence in communities, the lack of cost-effectiveness, and various logistical challenges. Therefore, the above-mentioned issues are overcome by this investigation proposing an effectual detection scheme called Shepard Convolutional Kronecker Network (ShCKN) for glaucoma detection. Initially, the input glaucoma image undergoes preprocessing with a Gaussian filter to eliminate unwanted noise. The preprocessed image is then fed into the segmentation module, where optic disc segmentation is achieved using Bayesian fuzzy clustering (BFC), followed by blood vessel detection through a sparking process. Significant features, such as local gradient binary pattern (LGBP), convolutional neural network (CNN) features, and statistical features, are then extracted. Finally, glaucoma detection is conducted using the ShCKN, which integrates Shepard convolutional neural networks (ShCNN) and deep Kronecker network (DKN), with layer modifications based on fuzzy concepts. Furthermore, the proposed ShCKN model achieved the highest accuracy of 91.8%, specificity of 91.4%, sensitivity of 91.2%, and Matthews correlation coefficient (MCC) of 88.3%.

In medical science, appropriate monitoring of the function of the human body, physical representations, and accurate disease is a complex task. It helps in diagnosing and treating diseases by revealing internal structures hidden by the skin and bones. Image enhancement techniques play a crucial role in medical imaging, as they improve diagnostic accuracy and visualization. However, the conventional methods were limited by noise sensitivity, lack of adaptability, and computational complexity. Therefore, this paper proposes One-to-One Honey Optimization (OOHO) for medical image enhancement. At first, the input brain Magnetic Resonance Imaging (MRI) image is processed to find the histogram. After that, histogram clipping is performed based on the clipping threshold. Then, the exposure threshold is computed to create sub-histograms. Moreover, the Probability Density Function (PDF) is computed and updated. Likewise, the Cumulative Density Function (CDF) and the mapping function are computed in each sub-histogram. Therefore, the finest enhanced image is obtained by integrating all sub-images. The fitness value is computed utilizing the cost function. Furthermore, the optimal threshold is accomplished by utilizing OOHO, which is developed by the integration of a One-to-One-based Optimizer (OOBO) and Honey Badger optimization (HBO). The evaluation results show that the OOHO attained a Degree of Distortion (DD) of 0.270, a Pear Signal to Noise Ratio (PSNR) of 47.060 dB, a Mean Square Error (MSE) of 0.570, Structural Similarity Index Measure (SSIM) as 0.960, and Root Mean Square Error (RMSE) of 0.755. The high performance is noted by the devised model and the MSE of the OOBO, HBO, hybrid Simulated Annealing-Evaporation Rate-based Water Cycle (SA-ERWCA) algorithm, Particle Swarm Optimization combined with Histogram Equalization (PSO-HE), Genetic Algorithm-based Adaptive Histogram Equalization (GAAHE), High-Quality Guidance Network (HQG-Net), multi-scale attention generative adversarial network (MAGAN), and Gaussian Quantum-behaved Arithmetic Optimization Algorithm (GQAOA) methods are 0.657, 0.637, 0.637, 0.627, 0.670, 0.650, 0.650, and 0.640, respectively.

Epilepsy is one of the most common brain disorders worldwide. The most frequently used clinical tool to detect epileptic events and monitor epilepsy patients is the EEG recordings. There have been proposed many computer-aided diagnosis systems using EEG signals for the detection and prediction of seizures. In this study, a novel method based on Fourier-based Synchrosqueezing Transform (SST), which is a high-resolution time-frequency (TF) representation, and Convolutional Neural Network (CNN) is proposed to detect and predict seizure segments. SST is based on the reassignment of signal components in the TF plane which provides highly localized TF energy distributions. Epileptic seizures cause sudden energy discharges which are well represented in the TF plane by using the SST method. The proposed SST-based CNN method is evaluated using the IKCU dataset we collected, and the publicly available CHB-MIT dataset. Experimental results demonstrate that the proposed approach yields high average segment-based seizure detection precision and accuracy rates for both datasets (IKCU: 98.99% PRE and 99.06% ACC; CHB-MIT: 99.81% PRE and 99.63% ACC). Additionally, SST-based CNN approach provides significantly higher segment-based seizure prediction performance with 98.54% PRE and 97.92% ACC than similar approaches presented in the literature using the CHB-MIT dataset.

Due to stochastic wireless environment, the process of modulation classification has become a challenging task. Because of its powerful feature extraction ability and promising performance over the conventional schemes, deep learning (DL) models are employed to automatic modulation classification (AMC) problems. Most of the conventional models proposed are tested for the limited set of modulation schemes transmitted over additive white Gaussian noise (AWGN) channels without considering the effect of multipath fading and Doppler shift. The next-generation networks use adaptive and higher-order quadrature amplitude modulation (QAM) schemes for higher spectral efficiency. The classification accuracy of conventional DL-based AMC schemes drastically reduces, when different order QAM modulation schemes are accommodated. In this work, different scaling factors are selected for the generation of $M$-QAM frames. The combination of scaling factors, which maximize the classification accuracy is chosen. A convolutional neural network (CNN) with six stages is employed for AMC. The simulation results show that the classification accuracy of proposed scheme is higher than the conventional DL-based schemes under various signal-to-noise ratio (SNR) conditions. The proposed scheme shows at least 4% improvement in classification accuracy over the other DL-based schemes.

This paper relates to the field of Artificial Intelligence, specifically to image recognition, and provides an innovative method to take advantage of Blockchain Convolutional Neural Networks (BCNNs) in Emotion Recognitions (ERs) using audio–visual emotion patterns to determine a healthcare emergency to be attended. BCNN architectures were used to identify emergency patterns. The results obtained indicate that the proposed method is adequate for the classification and identification of audio–visual patterns using deep learning (DL) with Restricted Boltzmann Machines (RBMs). It is concluded that it is sufficient to consider the audio–visible key features obtained from the patient’s face and voice of the proposed model to recognize a healthcare emergency for immediate action. “Sense of urgency” and “with urgency but with self-control” are the emotion profiles considered for a healthcare emergency, and user personal emotion profiles are stored in the Blockchain ecosystem since they are deemed sensitive data.

This work was to explore the application of deep learning (DL) in identifying the neural mechanism of cardiovascular disease (CVD) and the role of physical exercise in the prevention and treatment of CVD. 200 cases of outpatient treatment in the hospital from January to December in 2021 were included as the research objects. 100 people with fitness exercise habits were sorted into the experiment group, and the other 100 cases without fitness exercise habit were in the control group. In addition, a DL-based CVD recognition model was constructed. The results showed that the detection effect of the back propagation (BP) algorithm under DL was better, with an average of over 99%. Heart rate variability (HRV) time domain analysis results showed that the Rrmaen, standard deviation of N-N interval (SDNN), and root mean square of the difference (RMSSD) of the experiment group were $846.3\pm 85.4$ms, $103.6\pm 22.5$ms, and $19.2\pm 4.8$ms, respectively. These were observably higher than those of the control group ($P<0.05$). In the HRV frequency domain analysis, the total frequency (TF) in the experiment group was $318.4\pm 128.6$MS², which was notably higher than that in the control group ($239.8\pm 102.5$MS², $P<0.05$). The scores of anxiety and depression in the experiment group before exercise intervention were $43.6\pm 3.8$ and $46.9\pm 4.2$, respectively, which were highly decreased after intervention ($P<0.05$). The CVD recognition model based on a DL algorithm could effectively identify CVD. Long-term regular exercise can effectively change the regulatory function of cardiovascular autonomic nerves and depression and anxiety states, which had popularization value.

In recent times, online users freely express their sentiments in different life aspects because of the huge increase in social networks. Sentiment Analysis (SA) is one of the main Natural Language Processing (NLP) fields thanks to its important role in identifying sentiment polarities and making decisions from the public’s opinions. The Arabic language is one of the most challenging languages for SA due to its various dialects, and morphological and syntactic complexities. Deep Learning (DL) models have shown significant capabilities, especially in SA. In particular, Long Short-Term Memory (LSTM) networks have proven perfect abilities to learn sequential data. This paper proposes a comparative study result of Word2Vec and FastText word embedding models that are used to create two Arabic SA (ASA) LSTM-based approaches. The experimental results confirm that the LSTM model with FastText can significantly ameliorate the Arabic classification accuracy.