Search Results (70)

This paper presents a novel deep learning-based automatic modulation recognition (AMR) model, designed to classify ten modulation types from complex I/Q signal data. The proposed architecture, named CV-CNN-TCN, integrates Complex-Valued Convolutional Neural Networks (CV-CNNs) with Temporal Convolutional Networks (TCNs) to jointly extract spatial and temporal features while preserving the inherent phase information of the signal. An enhanced variant, CV-CNN-TCN-DCC, incorporates dilated causal convolutions to further strengthen temporal representation. The models are trained and evaluated on the benchmark RadioML2016.10b dataset. At SNR = −10 dB, the CV-CNN-TCN achieves a classification accuracy of 37%, while the CV-CNN-TCN-DCC improves to 40%. In comparison, ResNet reaches 33%, and other models such as CLDNN (convolutional LSTM dense neural network) and SCRNN (Sequential Convolutional Recurrent Neural Network) remain below 30%. At 0 dB SNR, the CV-CNN-TCN-DCC achieves a Jaccard index of 0.58 and an MCC of 0.67, outperforming ResNet (0.55, 0.64) and CNN (0.53, 0.61). Furthermore, the CV-CNN-TCN-DCC achieves 75% accuracy at SNR = 10 dB and maintains over 90% classification accuracy for SNRs above 2 dB. These results demonstrate that the proposed architectures, particularly with dilated causal convolutional enhancements, significantly improve robustness and generalization under low-SNR conditions, outperforming state-of-the-art models in both accuracy and reliability. Full article

The Draw-A-Person Intellectual Ability test for children, adolescents, and adults is a widely used tool in psychology for assessing intellectual ability. This test relies on human drawings for initial raw scoring, with the subsequent conversion of data into IQ ranges through manual procedures. However, this manual scoring and IQ assessment process can be time-consuming, particularly for busy psychologists dealing with a high caseload of children and adolescents. Presently, DAP-IQ screening continues to be a manual endeavor conducted by psychologists. The primary objective of our research is to streamline the IQ screening process for psychologists by leveraging deep learning algorithms. In this study, we utilized the DAP-IQ manual to derive IQ measurements and categorized the entire dataset into seven distinct classes: Very Superior, Superior, High Average, Average, Below Average, Significantly Impaired, and Mildly Impaired. The dataset for IQ screening was sourced from primary to high school students aged from 8 to 17, comprising over 1100 sketches, which were subsequently manually classified under the DAP-IQ manual. Subsequently, the manual classified dataset was converted into digital images. To develop the artificial intelligence-based models, various deep learning algorithms were employed, including Convolutional Neural Network (CNN) and state-of-the-art CNN (Transfer Learning) models such as Mobile-Net, Xception, InceptionResNetV2, and InceptionV3. The Mobile-Net model demonstrated remarkable performance, achieving a classification accuracy of 98.68%, surpassing the capabilities of existing methodologies. This research represents a significant step towards expediting and enhancing the IQ screening for psychologists working with diverse age groups. Full article

Chronic pain is a prevalent public health concern in the United States, frequently managed with over-the-counter (OTC) painkillers without professional medical supervision. This study investigates household-level patterns of over-the-counter painkiller use utilizing a nationally representative dataset from NielsenIQ, focusing on how reported health conditions, whether self-identified or professionally diagnosed, affect purchasing behaviors. By linking consumer purchase data with self-reported ailment information, this study analyzed painkiller expenditures across different ailment types and demographic groups. Results show that over-the-counter painkiller purchases were highly symptom-driven, particularly for headache-related products, which were the most frequently purchased category across all household types. Nearly one-third of single-member households purchased over-the-counter painkillers for headaches, regardless of diagnosis type, indicating a strong role of perceived need in driving behavior. Females and older individuals more frequently reported ailments, with consistently higher proportions across both pain-related and other conditions. Nonetheless, a notable share of households reported over-the-counter painkiller use without any reported ailments. The findings suggest that diagnostic status plays a limited role in determining over-the-counter painkiller usage, emphasizing the need for improved public health messaging around safe self-medication. These insights can inform targeted education, labeling regulations, and policy interventions to support safer and more equitable pain management practices at the population level. Full article

This paper presents the key results of a large-scale empirical study on the adoption of Generative AI (GenAI) by the Joint Research Centre (JRC), the European Commission’s science-for-policy department. Since spring 2023, the JRC has developed and deployed GPT@JRC, a platform providing safe and compliant access to state-of-the-art Large Language Models for over 10,000 knowledge workers. While the literature highlighting the potential of GenAI to enhance productivity for knowledge-intensive tasks is abundant, there is a scarcity of empirical evidence on impactful use case types and success factors. This study addresses this gap and proposes the JRC GenAI Compass conceptual framework based on the lessons learnt from the JRC’s GenAI adoption journey. It includes the concept of AI-IQ, which reflects the complexity of a given GenAI system. This paper thus draws on a case study of enterprise-scale AI implementation in European public institutions to provide approaches to harness GenAI’s potential while mitigating the risks. Full article

Intrinsic connectivity of the fusiform face area (FFA) was assessed using resting-state functional magnetic resonance imaging (fMRI) to compare adults with autism spectrum disorder (ASD; n = 17) and age-, sex-, and IQ-matched typically developing controls (TD; n = 22). The FFA seed region was delineated in each participant using a functional localizer task. Whole brain analyses of FFA connectivity revealed increased connectivity between the right FFA and the vermis, sensorimotor cortex, and extended face-processing network in individuals with ASD compared to TD participants; the TD group did not demonstrate increased functional connectivity. No group differences were observed from the left FFA. The relationship between FFA connectivity and the ability to remember faces significantly differed between the groups. Better face memory performance was positively correlated with increased connectivity within general visual processing areas in the ASD participants; whereas for the TD group, better face memory performance was associated with increased connectivity with brain regions related to face encoding, recognition, and retrieval. FFA overconnectivity with face, emotion, and memory processing areas, along with atypical relationships between FFA–occipito-temporal connections and face memory performance highlights a possible mechanism underlying social dysfunction in individuals with ASD. Full article

The precise assessment of image quality that matches human perception is still a major challenge in the field of digital imaging. Digital images play a crucial role in many technological and media applications. The existing deep convolutional neural network (CNN)-based image quality assessment (IQA) methods have advanced considerably, but there remains a critical need to improve the performance of existing methods while maintaining explicit tolerance to visual texture resampling and texture similarity. This paper introduces DualNetIQ, a novel full-reference IQA method that leverages the strengths of deep learning architectures to exhibit robustness against resampling effects on visual textures. DualNetIQ includes two main stages: feature extraction from the reference and distorted images, and similarity measurement based on combining global texture and structure similarity metrics. In particular, DualNetIQ takes features from input images using a group of hybrid pre-trained multi-scale feature maps carefully chosen from VGG19 and SqueezeNet pre-trained CNN models to find differences in texture and structure between the reference image and the distorted image. The Grey Wolf Optimizer (GWO) calculates the weighted combination of global texture and structure similarity metrics to assess the similarity between reference and distorted images. The unique advantage of the proposed method is that it does not require training or fine-tuning the CNN deep learning model. Comprehensive experiments and comparisons on five databases, including various distortion types, demonstrate the superiority of the proposed method over state-of-the-art models, particularly in image quality prediction and texture similarity tasks. Full article

This paper aims to highlight non-invasive BCI-VR applied protocols as intervention paradigms on school-aged subjects with ASD. Computer-based interventions are considered appropriate for users with ASD as concentration on a screen reduces other stimuli from the environment that are likely to be distracting or disruptive. Since there are no social conditions for engagement in such processes and the responses of computing systems do not hold surprises for users, as the outputs are fully controlled, they are ideal for ASD subjects. Children and adolescents with ASD, when supported by BCI interventions through virtual reality applications, especially appear to show significant improvements in core symptoms, such as cognitive and social deficits, regardless of their age or IQ. We examined nine protocols applied from 2016 to 2023, focusing on the BCI paradigms, the procedure, and the outcomes. Our study is non-exhaustive but representative of the state of the art in the field. As concluded by the research, BCI-VR applied protocols have no side effects and are rather easy to handle and maintain, and despite the fact that there are research limitations, they hold promise as a tool for improving social and cognitive skills in school-aged individuals with ASD. Full article

This study investigates the effects of financial globalization, trade globalization, and information and communication technology on the ecological footprint in G9 industrial economies (China, the United States, Japan, Germany, India, South Korea, Italy, France, and the United Kingdom) from 2000Q1 to 2018Q4. A distinctive Method of Moments Quantile Regression (MMQR) model was employed to analyze these relationships, and the Bootstrap Quantile Regression (BSQR) model was used to validate the results. The findings reveal that financial globalization (FG), environmental tax (ETAX), and institutional quality (IQ) contribute to environmentally sustainable development by reducing the ecological footprint (ECOFP). In contrast, trade globalization, information and communication technology (ICT), and gross domestic product (GDP) have a significant positive impact on the ecological footprint, leading to increased environmental degradation. The BSQR results corroborate these findings, confirming the roles of financial globalization, institutional quality, environmental tax, trade globalization, information and communication technology, and gross domestic product in shaping the ecological footprint. Based on these results, policymakers in G9 industrial nations should promote financial globalization as a tool to reduce the ecological footprint by encouraging green financing and environmentally sustainable investments. For trade globalization, stricter environmental regulations and sustainable trade practices are essential to mitigate its adverse environmental effects. Also, efforts to minimize the ecological impact of information and communication technology should focus on integrating renewable energy into ICT infrastructure and advancing green technology innovations. Full article

This paper details the enhancement of the optoelectronic properties of Cu-(In, Ga)-Se₂ (CIGSe) solar cells through a two-segment process in the ultraviolet (UV)–visible spectral range. These include fine-tuning the DC sputtering power of the absorber layer (ranging from 20 to 40 W at segment I) and thoroughly checking the trace micro-chemistry composition of the absorber layer (CdS, ZnO/CdS, ZnMgO/CdS, and ZnMgO at segment II). After segment I of treatment, the optimal 30 W CIGSe absorber layer (i.e., with a 0.95 CGI ratio) can be obtained, it can be seen that the Cu-rich film exhibits the ability to significantly promote grain growth and can effectively reduce its trap state density. After the segment II process aimed at replacing toxic CdS, the optimal metal alloy (Zn_0.9Mg_0.1O) composition (buffer layer) achieved the highest conversion efficiency (η) of 8.70%, also emphasizing its role in environmental protection. Especially within the tunable bandgap range (2.48–3.62 eV), the developed overall internal and external quantum efficiency (IQE/EQE) is significantly improved by 13.15% at shorter wavelengths. A photovoltaic (PV) module designed with nine optimal CIGSe cells demonstrated commendable stability. Variation remained within ±5% throughout the 60-day experiment. The PV modules in this study represent a breakthrough benchmark toward a significant advance in the scientific understanding of renewable energy. Furthermore, this research clearly promotes the practical application of PV modules, harmonizes with sustainable goals, and actively contributes to the creation of eco-friendly communities. Full article

The measurement of human breathing is crucial for assessing the condition of the body. It opens up possibilities for various intelligent applications, like advanced medical monitoring and sleep analysis. Conventional approaches relying on wearable devices tend to be expensive and inconvenient for users. Recent research has shown that inexpensive WiFi devices commonly available in the market can be utilized effectively for non-contact breathing monitoring. WiFi-based breathing monitoring is highly sensitive to motion during the breathing process. This sensitivity arises because current methods primarily rely on extracting breathing signals from the amplitude and phase variations of WiFi Channel State Information (CSI) during breathing. However, these variations can be masked by body movements, leading to inaccurate counting of breathing cycles. To address this issue, we propose a method for extracting breathing signals based on the trajectories of two-chain CSI ratios on the I/Q plane. This method accurately monitors breathing by tracking and identifying the inflection points of the CSI ratio samples’ trajectories on the I/Q plane throughout the breathing cycle. We propose a dispersion model to label and filter out CSI ratio samples representing significant motion interference, thereby enhancing the robustness of the breathing monitoring system. Furthermore, to obtain accurate breathing waveforms, we propose a method for fitting the trajectory curve of the CSI ratio samples. Based on the fitted curve, a breathing segment extraction algorithm is introduced, enabling precise breathing monitoring. Our experimental results demonstrate that this approach achieves minimal error and significantly enhances the accuracy of WiFi-based breathing monitoring. Full article

Automatic modulation recognition (AMR) methods used in advanced wireless communications systems can identify unknown signals without requiring reference information. However, the acceptance of these methods depends on the accuracy, number of parameters, and computational complexity. This study proposes a hybrid convolutional transformer classifier (HCTC) for the classification of unknown signals. The proposed method utilizes a three-stage framework to extract features from in-phase/quadrature (I/Q) signals. In the first stage, spatial features are extracted using a convolutional layer. In the second stage, temporal features are extracted using a transformer encoder. In the final stage, the features are mapped using a deep-learning network. The proposed HCTC method is investigated using the benchmark RadioML database and compared with state-of-the-art methods. The experimental results demonstrate that the proposed method achieves a better performance in modulation signal classification. Additionally, the performance of the proposed method is evaluated when applied to different batch sizes and model configurations. Finally, open issues in modulation recognition research are addressed, and future research perspectives are discussed. Full article

Background/Objectives: The implementation of information quality value stream maps (IQ-VSMs) in food animal production systems can increase our understanding of the opportunities and challenges when using laboratory testing for antimicrobial resistance (AMR) to support antimicrobial stewardship (AMS). Our objectives were to (1) explore the implementation of information quality value stream mapping as a continuous improvement tool to inform decisions for bovine respiratory disease (BRD) management and AMS and (2) apply the information quality dimensions to identified Kaizen opportunities for the integration of laboratory data into BRD management systems to assess the appropriateness of BRD treatment plans in western Canadian feedlot production. Methods: A ‘Current State’ IQ-VSM outlined the processes, available information, information processing steps, and control decisions contributing to BRD management and treatment in commercial western Canadian feedlots, recognizing that laboratory BRD pathogens and AMR data are typically not part of BRD management. Results: The ‘Future State’ IQ-VSM incorporated Kaizen opportunities for improvement, including (i) the strategic collection of respiratory samples from representative samples of calves for laboratory analysis, regardless of clinical BRD status, (ii) compilation of laboratory data at the pen and feedlot levels, and (iii) analysis of pen- and feedlot-level laboratory data to inform the veterinarian’s assessment of the appropriateness of current BRD treatment plans. Conclusions: The IQ-VSMs provided a valuable framework to visualize the integration of BRD pathogen and AMR laboratory data to support AMS and address any potential future testing requirements. Full article

Multimodal automatic modulation recognition (MAMR) has emerged as a prominent research area. The effective fusion of features from different modalities is crucial for MAMR tasks. An effective multimodal fusion mechanism should maximize the extraction and integration of complementary information. Recently, fusion methods based on cross-modal attention have shown high performance. However, they overlook the differences in information intensity between different modalities, suffering from quadratic complexity. To this end, we propose an efficient Alignment before Fusion Transformer Network (AbFTNet) based on an in-phase quadrature (I/Q) and Fractional Fourier Transform (FRFT). Specifically, we first align and correlate the feature representations of different single modalities to achieve mutual information maximization. The single modality feature representations are obtained using the self-attention mechanism of the Transformer. Then, we design an efficient cross-modal aggregation promoting (CAP) module. By designing the aggregation center, we integrate two modalities to achieve the adaptive complementary learning of modal features. This operation bridges the gap in information intensity between different modalities, enabling fair interaction. To verify the effectiveness of the proposed methods, we conduct experiments on the RML2016.10a dataset. The experimental results show that multimodal fusion features significantly outperform single-modal features in classification accuracy across different signal-to-noise ratios (SNRs). Compared to other methods, AbFTNet achieves an average accuracy of 64.59%, with a 1.36% improvement over the TLDNN method, reaching the state of the art. Full article

This paper presents the state-of-the-art techniques employed in aerothermal modeling to respond to the current observatory design challenges, particularly those of the next generation of extremely large telescopes (ELTs), such as the European ELT, the Thirty Meter Telescope International Observatory (TIO), and the Giant Magellan Telescope (GMT). It reviews the various aerothermal simulation techniques, the synergy between modeling outputs and observatory integrating modeling, and recent applications. The suite of aerothermal modeling presented includes thermal network models, Computational Fluid Dynamics (CFD) models, solid thermal and deformation models, and conjugate heat transfer models (concurrent fluid/solid simulations). The aerothermal suite is part of the overall observatory integrated modeling (IM) framework, which also includes optics, dynamics, and controls. The outputs of the IM framework, nominally image quality (IQ) metrics for a specific telescope state, are fed into a stochastic framework in the form of a multidimensional array that covers the range of influencing operational parameters, thus providing a statistical representation of observatory performance. The applications of the framework range from site selection, ground layer characterization, and site development to observatory performance current best estimate and optimization, active thermal control design, structural analysis, and an assortment of cost–performance trade studies. Finally, this paper addresses planned improvements, the development of new ideas, attacking new challenges, and how it all ties to the “Computational Fluid Dynamics Vision 2030” initiative. Full article

Background: The origin of pre-existing cognitive impairment in stroke patients remains controversial, with a vascular or a degenerative hypothesis. Objective: To determine whether endothelial dysfunction is associated with pre-existing cognitive problems, lesion load and biological anomalies in stroke patients. Methods: Patients originated from the prospective STROKDEM study. The baseline cognitive state, assessed using the IQ-CODE, and risk factors for stroke were recorded at inclusion. Patients with an IQ-CODE score >64 were excluded. Endothelial function was determined 72 h after stroke symptom onset by non-invasive digital measurement of endothelium-dependent flow-mediated dilation and calculation of the reactive hyperemia index (RHI). RHI ≤ 1.67 indicated endothelial dysfunction. Different biomarkers of endothelial dysfunction were analysed in blood or plasma. All patients underwent MRI 72 h after stroke symptom onset. Results: A total of 86 patients were included (52 males; mean age 63.5 ± 11.5 years). Patients with abnormal RHI have hypertension or antihypertensive treatment more often. The baseline IQ-CODE was abnormal in 33 (38.4%) patients, indicating a pre-existing cognitive problem. Baseline IQ-CODE > 48 was observed in 15 patients (28.3%) with normal RHI and in 18 patients (54.6%) with abnormal RHI (p = 0.016). The RHI median was significantly lower in patients with abnormal IQ-CODE. Abnormal RHI was associated with a significantly higher median FAZEKAS score (2.5 vs. 2; p = 0.008), a significantly higher frequency of periventricular lesions (p = 0.015), more white matter lesions (p = 0.007) and a significantly higher cerebral atrophy score (p < 0.001) on MRI. Vascular biomarkers significantly associated with abnormal RHI were MCP-1 (p = 0.009), MIP_1a (p = 0.042), and homocysteinemia (p < 0.05). Conclusions: A vascular mechanism may be responsible for cognitive problems pre-existing stroke. The measurement of endothelial dysfunction after stroke could become an important element of follow-up, providing an indication of the functional and cognitive prognosis of stroke patients. Full article