Explore the vast range of titles available.

Emotion matching prediction between music and video segments is essential for intelligent mobile sensing systems, where multimodal affective cues collected from smart devices must be jointly analyzed for context-aware media understanding. However, traditional approaches relying on single-modality feature extraction struggle to capture complex cross-modal dependencies, resulting in a gap between low-level audiovisual signals and high-level affective semantics. To address these challenges, a dual-driven framework that integrates perceptual characteristics with objective feature representations is proposed for audiovisual affective matching prediction. The framework incorporates fine-grained affective states of audiovisual data to better characterize cross-modal correlations from an emotional distribution perspective. Moreover, a decoupled Deep Canonical Correlation Analysis approach is developed, incorporating discriminative sample-pairing criteria (matched/mismatched data discrimination) and separate modality-specific component extractors, which dynamically refine the feature projection space. To further enhance multimodal feature interaction, an Attention and Feature-Consistency-Loss-Based Multi-Task Learning Network is proposed. In addition, a feature-consistency loss function is introduced to impose joint constraints across dual semantic embeddings, ensuring both affective consistency and matching accuracy. Experiments on a self-collected benchmark dataset demonstrate that the proposed method achieves a mean absolute error of 0.109 in music–video matching score prediction, significantly outperforming existing approaches.

Using the multifractal detrended cross-correlation analysis (MF-DCCA) method and the Empirical Mode Decomposition (EMD)-MF-DCCA method, this study quantifies the dynamic interrelation between carbon emission allowance returns in the Chinese and EU markets. The cross-correlation statistics indicate a moderate acceptance of the cross-correlation between the two carbon markets. Applying the MF-DCCA and EMD-MF-DCCA methods to the two markets reveals that their cross-correlation exhibits a power-law nature. Moreover, the apparent persistence of the cross-correlation and notable Hurst index show that the cross-correlation between long-term trends of the returns of the Guangdong and EU carbon emission markets exhibits stronger fractality over the long term, whereas the cross-correlation between the short-term fluctuations of the Hubei and EU carbon emission markets demonstrates stronger fractality. Subsequent investigations show that both fat tails and long memory contribute to the various fractals of the cross-correlation between the returns of the Chinese and EU carbon emission markets, especially for the fractals between the Hubei and EU carbon emission markets. Ultimately, the sliding window analysis demonstrates that national policy, trading activity, and other factors can make the observed multiple fractals more sensitive. The aforementioned findings facilitate an understanding of the current state of the Chinese carbon emission market and inform strategies for its future development.

Objective: This article aims to analyze the connectivity of the fat ox and corn price indices in Bahia   Theoretical Framework: The State of Bahia is economically and internationally important in terms of agricultural production in Brazil. Therefore, studying economic indicators and understanding their dynamics is necessary in increasingly competitive markets.   Method: To this end, the daily price indices in the main regions of the State of Bahia will be investigated using network analysis weighted by the coefficient of , the formation of clusters, and degree distribution.   Results and Discussion: Strong connectivity in the fat ox networks was found for all time scales and for corn only for large scales. The results allowed the unification of the fat ox market to be identified and the trend for price indices to move. Meanwhile, the corn market only has these characteristics for large scales, allowing for better short-term business opportunities.   Research Implications: This research provides valuable information for developing public policies, local and international investors, researchers, and those interested in the subject. In addition, it can serve as a bridge to understanding the socioeconomic and environmental effects on the commodities market.

Cholangiocarcinoma (CCA) is a hepatic malignancy that has a rapidly increasing incidence. CCA is anatomically classified into intrahepatic (iCCA) and extrahepatic (eCCA), which is further divided into perihilar (pCCA) and distal (dCCA) subtypes, with higher incidence rates in Asia. Despite its rarity, CCA has a low 5-year survival rate and remains the leading cause of primary liver tumor-related death over the past 10–20 years. The systemic therapy section discusses gemcitabine-based regimens as primary treatments, along with oxaliplatin-based options. Second-line therapy is limited but may include short-term infusional fluorouracil (FU) plus leucovorin (LV) and oxaliplatin. The adjuvant therapy section discusses approaches to improve overall survival (OS) post-surgery. However, only a minority of CCA patients qualify for surgical resection. In comparison to adjuvant therapies, neoadjuvant therapy for unresectable cases shows promise. Gemcitabine and cisplatin indicate potential benefits for patients awaiting liver transplantation. The addition of immunotherapies to chemotherapy in combination is discussed. Nivolumab and innovative approaches like CAR-T cells, TRBAs, and oncolytic viruses are explored. We aim in this review to provide a comprehensive report on the systemic and locoregional therapies for CCA.

In recent years, the use of electroencephalography (EEG) has grown as a tool for diagnostic and brain function monitoring, being a simple and non-invasive method compared with other procedures like histological sampling. Typically, in order to extract functional brain responses from EEG signals, prolonged and repeated stimuli are needed because of the artifacts generated in recordings which adversely impact the stimulus-response analysis. To mitigate the artifact effect, correlation analysis (CA) methods are applied in the literature, where the predominant approaches focus on enhancing stimulus-response correlations through the use of linear analysis methods like canonical correlation analysis (CCA). This paper introduces a novel CA framework based on a neural network with a loss function specifically designed to maximize correlation between EEG and speech stimuli. Compared with other deep learning CA approaches (DCCAs) in the literature, this framework introduces a single multilayer perceptron (MLP) network instead of two networks for each stimulus. To validate the proposed approach, a comparison with linear CCA (LCCA) and DCCA was performed, using a dataset containing the EEG traces of subjects listening to speech stimuli. The experimental results show that the proposed method improves the overall Pearson correlation by 10.56% compared with the state-of-the-art DCCA method.

With the diversification and complexity of multimedia data on big data, it becomes increasingly important to realize accurate and effective mutual retrieval among cross-media knowledge service data. In this paper, we first improve the structure of cross-media knowledge deep relevance analysis and apply it to cross-media data to construct cross-media relevance learning evaluation metrics. Then deep learning is commonly used for training classification labels or mapping vectors to another vector space by supervision, and with the rapid growth of data size and hardware resources, the advantages of deep learning in handling large-scale complex data will become more and more obvious. According to the experimental scheme to extract the features of the original data of Wikipedia and NUS-WIDE and the comparative analysis of the results based on the CCA extension method, the performance of CMC-DCCA on the dataset is 0.319, 0.338, 0.363, and 0.372, respectively, and it outperforms the other four algorithms. This study constructs a correlation analysis model between different media data to mine the correlations between cross-media data, thus realizing cross-media knowledge discovery service research while spawning more intuitive and concrete multimedia information carriers so that users can obtain more comprehensive information.

The advancement in satellite image sensors has enabled the acquisition of high-resolution remote sensing (HRRS) images. However, interpreting these images accurately and obtaining the computational power needed to do so is challenging due to the complexity involved. This manuscript proposed a multi-stream convolutional neural network (CNN) fusion framework that involves multi-scale and multi-CNN integration for HRRS image recognition. The pre-trained CNNs were used to learn and extract semantic features from multi-scale HRRS images. Feature extraction using pre-trained CNNs is more efficient than training a CNN from scratch or fine-tuning a CNN. Discriminative canonical correlation analysis (DCCA) was used to fuse deep features extracted across CNNs and image scales. DCCA reduced the dimension of the features extracted from CNNs while providing a discriminative representation by maximizing the within-class correlation and minimizing the between-class correlation. The proposed model has been evaluated on NWPU-RESISC45 and UC Merced datasets. The accuracy associated with DCCA was 10% and 6% higher than discriminant correlation analysis (DCA) in the NWPU-RESISC45 and UC Merced datasets. The advantage of DCCA was better demonstrated in the NWPU-RESISC45 dataset due to the incorporation of richer within-class variability in this dataset. While both DCA and DCCA minimize between-class correlation, only DCCA maximizes the within-class correlation and, therefore, attains better accuracy. The proposed framework achieved higher accuracy than all state-of-the-art frameworks involving unsupervised learning and pre-trained CNNs and 2–3% higher than the majority of fine-tuned CNNs. The proposed framework offers computational time advantages, requiring only 13 s for training in NWPU-RESISC45, compared to a day for fine-tuning the existing CNNs. Thus, the proposed framework achieves a favourable balance between efficiency and accuracy in HRRS image recognition.

Drug–target interaction (DTIs) prediction plays a vital role in probing new targets for breast cancer research. Considering the multifaceted challenges associated with experimental methods identifying DTIs, the in silico prediction of such interactions merits exploration. In this study, we develop a feature-based method to infer unknown DTIs, called PsePDC-DTIs, which fuses information regarding protein sequences extracted by pseudo-position specific scoring matrix (PsePSSM), detrended cross-correlation analysis coefficient (DCCA coefficient), and an FP2 format molecular fingerprint descriptor of drug compounds. In addition, the synthetic minority oversampling technique (SMOTE) is employed for dealing with the imbalanced data after Lasso dimensionality reduction. Then, the processed feature vectors are put into a random forest classifier to perform DTIs predictions on four gold standard datasets, including nuclear receptors (NR), G-protein-coupled receptors (GPCR), ion channels (IC), and enzymes (E). Furthermore, we explore new targets for breast cancer treatment using its risk genes identified from large-scale genome-wide genetic studies using PsePDC-DTIs. Through five-fold cross-validation, the average values of accuracy in NR, GPCR, IC, and E datasets are 95.28%, 96.19%, 96.74%, and 98.22%, respectively. The PsePDC-DTIs model provides us with 10 potential DTIs for breast cancer treatment, among which erlotinib (DB00530) and FGFR2 (hsa2263), caffeine (DB00201) and KCNN4 (hsa3783), as well as afatinib (DB08916) and FGFR2 (hsa2263) are found with direct or inferred evidence. The PsePDC-DTIs model has achieved good prediction results, establishing the validity and superiority of the proposed method.

Distal cholangiocarcinoma (dCCA) arises from the distal bile duct and is anatomically embedded within the pancreatic head, adjacent to abundant autonomic nerve plexuses. This unique location renders dCCA particularly prone to perineural invasion (PNI), a pathological hallmark that contributes to its dismal prognosis. However, the spatial architecture and molecular drivers that orchestrate PNI remain poorly defined. Here, we applied Xenium subcellular resolution spatial transcriptomics platform to profile resected tumor tissues from dCCA patients stratified by PNI status pathologically. A spatially resolved atlas comprising a total of 20 cell types was generated, uncovering enrichment of Schwann cells, type 2 conventional dendritic cells (cDC2), M2-like macrophages, cancer associated fibroblasts (CAFs) and B/plasma cells in PNI-high tumors, along with depletion of exhausted CD8 + T cells. Heterogeneous malignant cells in PNI-high tumors demonstrated activation of extracellular matrix remodeling and axonogenesis pathways, in line with the initial pathological classification. Spatial mapping further revealed distinct PNI-associated niches, notably matrix-producing CAFs (mCAFs)-macrophage clusters exhibiting coordinated enrichment of inflammatory and fibrotic programs. We further identified the LAMB3-DAG1 axis as a potential mediator of dCCA cells-Schwann cell interaction, while the preferential proximity of arteries to Schwann cells suggested additional microenvironmental support for nerve invasion. Collectively, our study provides a comprehensive subcellular atlas of PNI in dCCA, uncovering coordinated epithelial, stromal, and immune remodeling that drives perineural invasion. The identified biomarkers not only hold promise for patient stratification but may also guide intraoperative navigation and surgical margin determination, offering new avenues for precision therapy.

In this paper, we introduce a method called multiscale multifractal detrended cross-correlation analysis (MM-DCCA) to describe the cross-correlation properties depend on the timescale in which the multifractality is computed. For traffic time series, we show that the fractal properties of cross-correlations have a relationship with the range of scale indicating the great necessity to study the cross-correlation properties between two time series at multiple scales. MM-DCCA gains a new insight into measuring different fractal properties of the cross-correlations between traffic series by sweeping all the range of scale, and it provides much richer information than multifractal detrended cross-correlation analysis (MF-DCCA). The Hurst surfaces present multifractal properties and strong long-range persistent cross-correlations between traffic series. By comparing Hurst surfaces before and after removing dominant periodicities, we find that periodicity is not the only reason which causes the crossover and dominates the cross-correlation. There are other interesting factors or underlying traffic mechanisms containing in the cross-correlation between traffic series. Moreover, the cross-correlation between the whole traffic series can be considered as a combination of both weekday and weekend parts. The results also suggest that the different periodic patterns hidden in the weekday and weekend patterns are the main distinction between them and play an important role in the Hurst surface of cross-correlation investigation.