Multidimensional Data Research Articles

In the past decade, breast pathology in China has made significant progress in diagnostic standards, technological applications, scientific research, and discipline development. The histopathological diagnostic system has been continuously refined, with the implementation of relevant guidelines and expert consensus enhancing standardization and reproducibility of diagnostic results. Immunohistochemistry and molecular testing technologies have become increasingly sophisticated, with emerging biomarkers such as low HER2 expression and PIK3CA mutations gradually integrated into clinical decision-making, promoting the advancement of precision therapy. The application of digital pathology and image-assisted analysis has steadily expanded, providing new tools to improve diagnostic efficiency and consistency. The national breast pathology group has actively advanced the development of tiered diagnostic systems, workforce training, and public education, effectively strengthening diagnostic capabilities at the grassroots level. Looking ahead, the integration of multidimensional data, optimization of auxiliary diagnostic systems, and interdisciplinary collaboration are expected to drive the continued development of breast pathology in China.

Low-rank tensor recovery methods within the tensor singular value decomposition (t-SVD) framework have demonstrated considerable success by leveraging the inherent low-dimensional structures of multi-dimensional data. However, previous approaches in this framework often rely on linear transforms or, in some cases, nonlinear transforms constructed with fully connected networks (FCNs). These methods typically promote a global low-rank structure, which may not fully exploit the nature of multiple subspaces in real-world data. In this work, we propose a nonlinear transform to capture long-range dependencies and diverse patterns across multiple subspaces of the data within the t-SVD framework. This approach provides a richer and more nuanced representation compared to the localized processing typically seen in FCN-based transforms. In the transform domain, we construct a low-rank self-representation layer that fully exploits the multi-subspace structure inherent in tensor data. Instead of merely enforcing overall low-rankness, our method minimizes the nuclear norm of a self-representation tensor, allowing for a more precise and joint characterization of multiple subspaces. This results in a more accurate representation of the data's intrinsic low-dimensional structures, leading to superior recovery performance. This new framework, termed the DEep Low-rank Tensor representAtion (DELTA), is evaluated across several typical multi-dimensional data recovery applications, including tensor completion, robust tensor completion, and spectral snapshot imaging. Experiments on various real-world multi-dimensional data illustrate the superior performance of our DELTA.

Spatial metabolomics is a rapidly advancing field offering powerful insights into metabolic heterogeneity in biological tissues. However, its widespread adoption is hindered by fragmented tools and the lack of comprehensive, open-source GUI software covering the full analytical workflow (quality control, preprocessing, identification, pattern, and differential analysis). To address this, we developed SMAnalyst, an open-source, integrated web-based platform. SMAnalyst consolidates core functionalities, including multi-dimensional data quality assessment (background consistency, intensity, missing values), a comprehensive metabolite annotation scoring system (mass accuracy, isotopic similarity, adduct evidence), and dual-dimension spatial pattern discovery (metabolite co-expression and pixel clustering). It also offers flexible differential analysis (cluster- or user-defined regions). With its intuitive GUI and modular workflow, SMAnalyst significantly lowers the analysis barrier, by providing a unified solution that eliminates the need for tool switching and advanced computational skills. Tested with a mouse brain dataset, SMAnalyst efficiently handles large-scale data (e.g., >14,000 pixels, >3000 ion peaks), effectively filling a critical gap in integrated analytical solutions for spatial metabolomics.

Molecular glues (MGs) represent a unique class of small molecules that modulate protein-protein interactions by altering target protein surface properties, enabling targeted degradation, pathway modulation, or functional control of proteins, including traditionally undruggable targets. Currently, 17 MG-based drugs are approved globally with over 40 in clinical trials, underscoring their therapeutic potential. Despite these advances, the lack of a dedicated database integrating structural, pharmacological, and computational data for such compounds hinders rational drug design. To address this gap, we developed MGDB, a specialized open-access repository integrating rigorously curated multidimensional data for MGs. MGDB contains 7396 curated MGs being sourced from 162 peer-reviewed publications and 156 patents. It consolidates structural data, 9728 experimental bioactivity data points (covering degradation efficiency, binding affinity, cellular/animal activity) across 201 targets and 108 effectors, 115 296 computed physicochemical properties, and 270 785 ADMET profiles. The database supports text-based and chemical structure-based queries and interoperability with external resources (e.g. PubChem, ChEMBL, DrugBank, UniProt, and WIPO) via hyperlinks. By centralizing and standardizing specialized MG information, MGDB empowers researchers to rapidly explore MG research landscapes and provides high-quality datasets for artificial intelligence-driven rational therapeutic design. MGDB is freely available at http://mgdb.idruglab.cn/.

Plug-in Hybrid Electric Vehicles (PHEVs) are increasingly favored for their low emissions and freedom from range anxiety, combining electric efficiency with the extended range of a gasoline engine. While previous research on PHEV energy consumption has predominantly focused on powertrain design and energy management strategies, there is growing recognition of the critical role played by driver behavior in determining real-world energy efficiency. Based on multi-mode vehicle data collected from real-world driving scenarios, we propose a novel dual-layer LSTM-Transformer model, named DLLT, for real-time prediction of energy consumption and driving dynamics in multi-mode PHEVs. The first layer employs an LSTM network to perform mode clustering, while the second layer conducts energy consumption regression using a Transformer model with integrated mode information. This hierarchical architecture enables adaptation to diverse driving and braking modes, significantly enhancing the model’s ability to accurately identify vehicle operation modes and precisely predict energy consumption. To more accurately validate the effectiveness of DLLT in modeling eco-driving behavior for PHEVs, we collected a large amount of multidimensional time-series data from real-world PHEVs. Experimental results demonstrate that the model achieves a 93% accuracy rate in vehicle mode prediction. Under unseen driving conditions, it attains R2 values of 0.99 for fuel consumption, 0.86 for acceleration, and 0.81 for electric power, outperforming existing models across all evaluation metrics. With its high prediction accuracy and robust generalization capability, DLLT shows great potential for applications in PHEV eco-driving behavior analysis, intelligent energy management systems, and future autonomous driving control strategies.

Currently, thermal power units undertake the task of peak and frequency regulation, and their internal equipment is in a non-conventional environment, which could very easily fail and thus lead to unplanned shutdown of the unit. To realize the condition monitoring and early warning of the key equipment inside coal power units, this study proposes a deep learning-based equipment condition anomaly detection model, which combines the deep autoencoder (DAE), Transformer, and Gaussian mixture model (GMM) to establish an anomaly detection model. DAE and the Transformer encoder extract static and time-series features from multi-dimensional operation data, and GMM learns the feature distribution of normal data to realize anomaly detection. Based on the data verification of boiler superheater equipment and turbine bearings in real power plants, the model is more capable of detecting equipment anomalies in advance than the traditional method and is more stable with fewer false alarms. When applied to the superheater equipment, the proposed model triggered early warnings approximately 90 h in advance compared to the actual failure time, with a lower false negative rate, reducing the missed detection rate by 70% compared to the Transformer-GMM (TGMM) model, which verifies the validity of the model and its early warning capability.

With the rapid development of e-commerce and social media, consumer-generated visual and textual content has become a vital resource for understanding purchasing behavior. However, traditional unimodal sentiment analysis (text-only or image-only) struggles to fully capture emotional context, leading to prediction biases. Multimodal fusion technology integrates multidimensional data such as text and images to simulate human decision-making processes that synthesize multiple cues, providing a more holistic perspective for consumer sentiment analysis. This study proposes a multimodal analysis framework integrating BERT (text) and ResNet-50 (image), validated on datasets to predict consumer purchase tendencies. Experimental results demonstrate that the model achieves an accuracy of 85%a 12% improvement over single-modal baselines. Five-fold cross-validation tests on concatenated features and statistical significance tests (p<0.05) validate the model's effectiveness. The findings confirm the potential of multimodal fusion in optimizing consumer insight extraction and enhancing recommendation system performance, offering new approaches for intelligent services on e-commerce platforms.

The integration of high-penetration distributed renewable energy sources into new power systems introduces significant challenges, particularly frequent reverse power flows that threaten substation security. To address this issue, this paper proposes a novel safety assessment method based on a system dynamics (SD) framework. This approach uniquely emphasizes the critical roles of electrical interconnections among substation equipment and the fluctuations in distributed power output. The methodology involves analyzing operational characteristics to establish equipment correlations, developing a comprehensive fault probability function for each equipment by integrating multi-dimensional monitoring data and fault propagation factors, and constructing a system dynamics model using an adjacency matrix to represent operational relationships. The effectiveness of the proposed method is validated through a case study on a regional substation. Results demonstrate its capability to dynamically and accurately evaluate both equipment-level and system-wide safety status under reverse power flow conditions, providing a robust tool for enhancing the security and resilience of modern power systems.

ABSTRACT Accurate stock market prediction is crucial for investors to formulate correct investment strategies. However, the non‐linearity, high dimensionality, and volatility of financial data pose significant challenges to existing stock market prediction models. To effectively address the complex datasets faced by stock market prediction, this paper proposes a new and more efficient deep learning hybrid model, KAN‐LSTM, based on the LSTM (long short‐term memory) and integrating the KAN (Kolmogorov–Arnold network). The hybrid architecture improves the learning process by replacing the original MLP (multi‐layer perceptron) with the KAN, overcoming the limitations of poor interpretability and fixed activation functions in LSTM. Prediction experiments conducted on multidimensional financial data in the stock market show that the KAN‐LSTM hybrid model outperforms the original LSTM in all evaluation metrics, demonstrating superior performance and more efficient prediction capabilities. Specifically, the MAE (mean absolute error) decreased by 2.43%, the RMSE (root mean squared error) decreased by 1.92%, the MAPE (mean absolute percentage error) decreased by 2.2%, and the increased by 19.08%.

Background: Hemorrhagic transformation (HT) is still a major negative outcome of mechanical thrombectomy (MT) in acute ischemic stroke (AIS), significantly affecting death rates and health complications. Despite advances in patient selection and procedural procedures, reliable prediction models for stratifying patients based on their risk of hypertension are still lacking. Artificial intelligence (AI) facilitates the utilization of complex, multidimensional data to aid in early forecasting and tailored stroke treatment. Objective: To develop and validate a machine learning model for predicting hemorrhagic transformation after MT using a nationally representative inpatient database, and to identify the most impactful clinical predictors. Methods: We used the National Inpatient Sample (2016-2020) to conduct a retrospective cohort study. Adult patients (≥18 years) hospitalized with AIS (ICD-10: I63.x) and undergoing MT were identified. HT was defined using verified ICD-10 codes (I61.x and I62.x). A machine learning pipeline using Extreme Gradient Boosting (XGBoost) was created. Demographics, hospital characteristics, stroke risk factors (atrial fibrillation, hypertension, diabetes), thrombolytic usage, and surrogate indicators for stroke severity were all taken into account. The data were divided into training (80%) and testing (20%) cohorts. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration plots, and SHAP (Shapley Additive Explanation) values, which stress feature relevance. Results: Among 45,321 AIS patients with MT, 3,746 (8.3%) developed HT. The AI model has an AUC of 0.87 (95% CI: 0.86-0.89), demonstrating excellent discriminatory performance. The use of thrombolytics, coagulopathy, chronic renal disease, advanced age, atrial fibrillation, and therapy at large metropolitan teaching institutions were all significant predictors of hypertension. SHAP analysis gives comprehensive insight into variable interactions, which improves the model's clinical interpretability. Conclusion: This research presents the first nationally representative AI-driven model for predicting hemorrhagic change after MT. The approach, which combines real-world data with machine learning, provides a therapeutically effective tool for pre-procedural risk classification and individualized post-thrombectomy treatment. Future inclusion into stroke treatment pathways might reduce complications and improve patient outcomes.

Due to the increasing intensity and complexity of network interactions in aviation data transmission systems, the need for developing methods to detect signs of unauthorized interference in aviation operations is significantly growing. The importance of this issue is due to the need to ensure control systems and affect the safety of aircraft flights. This article develops and presents a method for analyzing multidimensional combinations of network traffic features in aviation data transmission systems, based on a modified frequent-pattern FP-Growth algorithm adapted specifically for multidimensional data. A distinctive feature of the proposed approach is maintaining the contextual integrity of network event attributes, enabling the identification of hidden dependencies among various parameters of network events that are inaccessible to traditional one-dimensional frequent pattern analysis algorithms. A model for representing network events as multidimensional transactions is formulated, and an algorithm for constructing a multidimensional frequent-pattern tree and extracting stable combinations of features with a predefined frequency of occurrence is proposed. Experimental validation using real network traffic data confirmed the capability of detecting network attack patterns and previously unrecorded anomalous feature combinations. A quantitative evaluation of the proposed method’s performance was conducted, confirming its efficiency and suitability for processing substantial data volumes characteristic of aviation data transmission systems in real-time conditions. The developed method provides improved protection for aviation networks and timely identification of threats to aviation operations. The developed method can be applied to enhance the resilience of aviation data transmission systems for air traffic management and prioritize protective measures to ensure flight safety.

Multimodal prediction of catheter ablation outcomes in patients with persistent atrial fibrillation.

The aim of this study was to investigate, validate and apply Machine Learning (ML) algorithms to predict functional disability in elderly individuals using data from ELSI-Brazil. Furthermore, it sought to map the performance of the models and identify key multidimensional variables-encompassing sociodemographic and economic aspects, health status, behaviors, mental health and access to services-that could serve as early risk indicators and, based on the selected model, understand which characteristics favor or disfavor the screening of functional disability. Data from ELSI-Brazil (2015-2016) with 4502 participants were analyzed, after careful selection and pre-processing, which included imputing missing data, standardization and encoding via one-hot encoder. The selection of 49 predictor variables, from sociodemographic, health and behavioral domains, enabled the development of classification models. The SMOTE technique and tenfold cross-validation, associated with Bayesian optimization, were applied. The interpretability of the selected model was performed through SHAP analysis. The Ridge Classifier model showed robust performance, with a ROC-AUC of 0.785 (95% CI: 0.756-0.813), sensitivity of 0.703 and specificity of 0.723, in addition to a high negative predictive value (84.5%). SHAP analysis showed that variables such as depressive symptoms, concern about mobility and self-rated health status were decisive in classifying the functional disability risk. The results suggest that the use of ML techniques, integrated with multidimensional health data commonly collected in primary care settings, offers a promising tool for screening and early intervention in functional disability in the elderly. This approach may substantiate decision-making in clinical practice and health support policies aimed at active and healthy aging.

Rehabilitation, neuroplasticity, and machine learning: Approaching artificial intelligence for equitable health systems.

Background: Since each dimension of a tensor can store different types of genomics data, compared to matrix methods, utilizing tensor structure can provide a deeper understanding of multi-dimensional data while also facilitating the discovery of more useful information related to cancer. However, in reality, there are issues such as insufficient utilization of prior knowledge in multi-omics data and limitations in the recovery of low-tubal-rank tensors. Therefore, the method proposed in this article was developed. Objective: In this paper, we proposed a low transformed tubal rank tensor model (LTTRT) using a spatial-tubal constraint to accurately partition different types of cancer samples and provide reliable theoretical support for the identification, diagnosis, and treatment of cancer. Method: In the LTTRT method, the transformed tensor nuclear norm based on the transformed tensor singular value decomposition is characterized by the low-rank tensor, which can explore the global low-rank property of the tensor, resolving the challenge of the tensor nuclear norm-based method not achieving the lowest tubal rank. Additionally, the introduction of weighted total variation regularization is conducive to extracting more information from sequencing data in both spatial and tubal dimensions, exploring cross-correlation features of multiple genomic data, and addressing the problem of overlooking prior knowledge from various perspectives. In addition, the L1-norm is used to improve sparsity. A symmetric Gauss‒Seidel-based alternating direction method of multipliers (sGS-ADMM) is used to update the LTTRT model iteratively. Results: The experiments of sample clustering on multiple integrated cancer multi-omics datasets show that the proposed LTTRT method is better than existing methods. Experimental results validate the effectiveness of LTTRT in accurately partitioning different types of cancer samples. Conclusion: The LTTRT method achieves precise segmentation of different types of cancer samples.

While automated insulin delivery (AIDs) systems have significantly improved glycemic control for individuals with type 1 diabetes (T1D), there remains a need for identifying and acting upon complex physiologic and behavioral patterns which consistently lead to hypo- and hyperglycemia. Prior methods have lacked the ability to automatically identify and extract patterns across mixed-type multidimensional data (eg, insulin, glucose, activity) without instilling bias from stipulations on time-lagged coupling, pattern length, or pre-defining patterns. We introduce a new pattern-detection technique-Block-based Recurrence Quantification Analysis (BlockRQA)-and preliminary results using BlockRQA in an AID on both in silico and in an outpatient feasibility study. We first introduce the BlockRQA algorithm, which extends Recurrence Quantification Analysis for use in categorical and continuous time-series data, while maintaining interpretable patterns in the domain of interest, in contrast to prior state-of-the-art approaches which require embeddings. Next, we demonstrate the feasibility of utilizing these patterns and BlockRQA with an existing AID system (BlockRQA+AID) to identify and dose for patterns leading to hyperglycemia in individuals with T1D. We demonstrate how BlockRQA+AID can improve glucose outcomes in patterns leading to hyperglycemia in silico. And we show real-world results using BlockRQA+AID to reduce hyperglycemic events (>250 mg/dL) via an interim safety analysis of a small outpatient pilot study. For all cases, we show BlockRQA efficiently identifies, aggregates, and scores behavioral patterns which can be targeted for clinical intervention. The BlockRQA is a powerful pattern recognition tool that may be used to identify glucose outcome patterns to guide AID dosing.

Perspective on integrated multi-omics approaches and constraint-based modeling to explore metabolic functionality on the evolution of bacterial antibiotic resistance.

IntroductionAtrial fibrillation (AF), the most common sustained arrhythmia globally, necessitates effective strategies for stroke prevention. Although current risk stratification tools, such as the CHA₂DS₂-VASc score, are widely used to guide anticoagulation therapy, their limited predictive accuracy underscores the urgent need for more precise and reliable models. This study aims to establish a nationwide AF registry incorporating multi-dimensional data to identify novel risk factors and develop a more accurate stroke prediction model to improve risk stratification and guide anticoagulation therapy in patients with AF.Methods/analysisThe risk factors for stroke in patients with non-valvular AF in China (REFINE) registry is a nationwide, multicentre, observational registry integrating retrospective (n=20 000) and prospective (n=5000) cohorts. Demographics, lifestyle, medical history, physical examination, laboratory tests, ECG, echocardiography, contrast-enhanced CT scan and blood samples will be collected at baseline. Long-term follow-up will be performed to identify clinical events and treatment at the timepoint. We aim to use the multidimensional dataset to establish a more precise stroke risk predictive tool.Ethics and disseminationThe study is approved by the Ethics Committee of Fuwai Hospital, CAMS&PUMC (No. 2022–1845; No. 2024–2489) and registered at ClinicalTrials.gov, identifier NCT05598632. The results of this study will be disseminated through publications in peer-­reviewed journals and conference presentations.Trial registration numberNCT05598632.

Pyridaben exposure triggers osmotic and oxidative imbalance causing morphologic deformities in Daphnia magna.

Interstitial lung diseases (ILDs) are a heterogeneous group of diseases characterised by varying degrees of inflammation and fibrosis. Among these, fibrotic interstitial lung disease (FILD) is receiving increasing attention. Many questions remain about FILD, such as identifying which ILDs are likely to progress to FILD, the timing of such progression, early recognition methods and biomarkers for FILD recognition and progression, highlighting the urgent need for a large multicentre FILD cohort to advance relevant studies. The Fibrotic Interstitial Lung Disease Early Recognition and Strategic Therapy Study in China (FIRST) study is a prospective, multicentre cohort study of FILD conducted across 40 hospitals/centres in China, aiming to reveal clinical phenotype and further integrate multidimensional data to develop prediction models for fibrosis progression. The study will enrol more than 10 000 patients, using existing national ILD cohorts, and will collect comprehensive clinical, imaging, histological and biological data to support histopathological, imaging and biomarker-related studies. The primary outcome is fibrosis progression within 12 months, with secondary outcomes including the natural history, mortality, comorbidities and treatment conditions of FILD, using a standardised data collection and follow-up approach supported by an Electronic Data Capture system. The study's protocol has undergone a thorough review and received approval from the Ethics Committee of China-Japan Friendship Hospital and the other participating sites currently enrolling patients. The findings of the study will be shared with the broader scientific community through publication in peer-reviewed journals. NCT06655090; Pre-results.