Multidimensional Data Research Articles

AI-Aided Diagnosis For Neurodegenerative Diseases: Prospects And Challenges

Neurodegenerative disease (ND) represents a chronic disease characterized by loss of neuron function and death, such as Alzheimer's disease, Parkinson's disease, etc. It’s difficult and complex to diagnose ND, due to the requirement for the synthesis of multiple biomarker data, such as genes, proteins, images, etc. The advent of artificial intelligence (AI) technology, particularly through machine learning (ML) and deep learning (DL) algorithms, introduces novel methodologies and tools for ND diagnosis. These methodologies extract pertinent information and patterns from massive, multidimensional, and non-linear data, assisting doctors to render more precise, prompt, and objective assessments. This article provides an overview of the updated implementation status of AI in ND diagnosis, assesses the advantages and contributions of AI to ND diagnosis, as well as the existing limitations and challenges, and offers insights to the future development direction of AI in ND diagnosis.

MRMLREC: A Two-Stage Approach for Addressing Data Sparsity in MOOC Video Recommendation (Student Abstract)

With the abundance of learning resources available on massive open online courses (MOOCs) platforms, the issue of interactive data sparsity has emerged as a significant challenge.This paper introduces MRMLREC, an efficient MOOC video recommendation which consists of two main stages: multi-relational representation and multi-level recommendation, aiming to solve the problem of data sparsity. In the multi-relational representation stage, MRMLREC adopts a tripartite approach, constructing relational graphs based on temporal sequences, courses-videos relation, and knowledge concepts-video relation. These graphs are processed by a Graph Convolution Network (GCN) and two variant Graph Attention Networks (GAT) to derive representations. A variant of the Long Short-Term Memory Network (LSTM) then integrates these multi-dimensional data to enhance the overall representation. The multi-level recommendation stage introduces three prediction tasks at varying levels—courses, knowledge concepts, and videos—to mitigate data sparsity and improve the interpretability of video recommendations. Beam search (BS) is employed to identify top-β items at each level, refining the subsequent level's search space and enhancing recommendation efficiency. Additionally, an optional layer offers both personalization and diversification modes, ensuring variety in recommended videos and maintaining learner engagement. Comprehensive experiments demonstrate the effectiveness of MRMLREC on two real-world instances from Xuetang X.

Global change impacts on mineral nutritional quality of cereal grains: Coordinated datasets and analyses to advance a systems-based understanding

ContextGlobal nutritional health outcomes are directly reliant on agroecosystem nutrient outputs. Appropriately, there is concern surrounding the impacts of a changing climate not only on crop yields, but also on crop nutritional quality (e.g., mineral nutrient concentrations). Quantifying the impacts of elevated CO2 concentrations, elevated temperature, drought stress, edaphic factors, and agronomic management on crop yields and mineral nutrition is critical, yet a systems-level understanding of these interactive factors is poorly developed, limiting our ability to effectively target solutions. Empirical data for climate impacts on crop nutritional quality remain scarce, with much of the research emerging from valuable, but geographically limited, Free-air CO2 Enrichment (FACE) experiments, several of which suggest that human nutrition will be adversely impacted by e[CO2]. Specific concerns center on observed declines in grain protein, iron, and zinc concentrations due to already wide-spread human nutritional deficiencies in these nutrients. ObjectivesAs global change experiments expand to pursue questions regarding interactive climate impacts on crop yields and nutritional quality, it is imperative to interrogate the measurements, data standardization, and metadata needed for unifying synthesis. The data reported for shifts in crop nutritional quality are often incomplete, precluding the generalizability and comparability of results. MethodsWe frame this review around six inter-reliant methods, tools, and practices to support maximally useful experimental datasets to inform questions of global change impacts on crop nutrition and aid in detecting genotypic differences in mineral nutrient density. The bulk of the data and discussion centers on wheat (Triticum aestivum L.) due to the central role this crop plays in human nutrition and sustained biofortification efforts. ResultsTo permit experimental comparability and synthesis, datasets should (1) clearly delineate analytical methods and standards and (2) link mean nutrient concentrations with the covariate of yield. (3) Multi-year, multi-location data is required to identify genotypes with significant deviations in nutrient concentrations, with (4) data normalized for yield within appropriate analytical frameworks. (5) Inclusion of data on soil properties, weather, and abiotic and biotic stresses as well as (6) agronomic practices and nutrient management is essential for understanding global change impacts on nutritional outcomes. ConclusionsCoordinated, multi-dimensional data will permit the syntheses and meta-analyses needed to identify and quantify climate impacts on nutrition. ImplicationsThis work is essential to effectively target nutritional solutions, to develop modeling tools to support nutritional planning, and to identify areas where agronomic management and breeding can minimize climate impacts on nutritional outcomes.

Multi-Dimensional Fusion of Spectral and Polarimetric Images Followed by Pseudo-Color Algorithm Integration and Mapping in HSI Space

Spectral–polarization imaging technology plays a crucial role in remote sensing detection, enhancing target identification and tracking capabilities by capturing both spectral and polarization information reflected from object surfaces. However, the acquisition of multi-dimensional data often leads to extensive datasets that necessitate comprehensive analysis, thereby impeding the convenience and efficiency of remote sensing detection. To address this challenge, we propose a fusion algorithm based on spectral–polarization characteristics, incorporating principal component analysis (PCA) and energy weighting. This algorithm effectively consolidates multi-dimensional features within the scene into a single image, enhancing object details and enriching edge features. The robustness and universality of our proposed algorithm are demonstrated through experimentally obtained datasets and verified with publicly available datasets. Additionally, to meet the requirements of remote sensing tracking, we meticulously designed a pseudo-color mapping scheme consistent with human vision. This scheme maps polarization degree to color saturation, polarization angle to hue, and the fused image to intensity, resulting in a visual display aligned with human visual perception. We also discuss the application of this technique in processing data generated by the Channel-modulated static birefringent Fourier transform imaging spectropolarimeter (CSBFTIS). Experimental results demonstrate a significant enhancement in the information entropy and average gradient of the fused image compared to the optimal image before fusion, achieving maximum increases of 88% and 94%, respectively. This provides a solid foundation for target recognition and tracking in airborne remote sensing detection.

Discovering invariant spatial features in electron energy loss spectroscopy images on the mesoscopic and atomic levels

Over the last two decades, Electron Energy Loss Spectroscopy (EELS) imaging with a scanning transmission electron microscope has emerged as a technique of choice for visualizing complex chemical, electronic, plasmonic, and phononic phenomena in complex materials and structures. The availability of the EELS data necessitates the development of methods to analyze multidimensional data sets with complex spatial and energy structures. Traditionally, the analysis of these data sets has been based on analysis of individual spectra, one at a time, whereas the spatial structure and correlations between individual spatial pixels containing the relevant information of the physics of underpinning processes have generally been ignored and analyzed only via the visualization as 2D maps. Here, we develop a machine learning-based approach and workflows for the analysis of spatial structures in 3D EELS data sets using a combination of dimensionality reduction and multichannel rotationally invariant variational autoencoders. This approach is illustrated for the analysis of both the plasmonic phenomena in a system of nanowires and in the core excitations in functional oxides using low loss and core-loss EELS, respectively. The code developed in this manuscript is open sourced and freely available and provided as a Jupyter notebook for the interested reader.

Enabling privacy-preserving medication analysis in distributed EHR systems

The digitalization of the medical field has resulted in a vast amount of medical data stored in Electronic Health Record (EHR) systems. Among the various valuable applications of EHRs, medication efficacy analysis stands out as a powerful tool for drug development and enhancing patient treatment. This analysis relies on multi-dimensional medical data, encompassing various factors. Currently, medical research is increasingly inclined towards inter-institutional sharing of medical data to enhance the accuracy and reliability of analysis results. However, due to the sensitivity of personal medical data, privacy and security considerations must be taken into account when conducting medical analyses across institutions. In this paper, we present a novel privacy-preserving scheme for distributed medication efficacy analysis. Our work allows for the analysis of multi-dimensional medical data by leveraging the private set intersection protocol and the Paillier cryptosystem. To validate the feasibility of our scheme, we implement it using some real medical datasets. Experimental results demonstrate that our approach offers enhanced security and lower computational and communication overhead compared to the original approach, which solely processes single-dimensional data at a time.

Thrombosis risk in single- and double-heterozygous carriers of factor V Leiden and prothrombin G20210A in FinnGen and the UK Biobank

AbstractThe factor V Leiden (FVL; rs6025) and prothrombin G20210A (PTGM; rs1799963) polymorphisms are 2 of the most well-studied genetic risk factors for venous thromboembolism (VTE). However, double heterozygosity (DH) for FVL and PTGM remains poorly understood, with previous studies showing marked disagreement regarding thrombosis risk conferred by the DH genotype. Using multidimensional data from the UK Biobank (UKB) and FinnGen biorepositories, we evaluated the clinical impact of DH carrier status across 937 939 individuals. We found that 662 participants (0.07%) were DH carriers. After adjustment for age, sex, and ancestry, DH individuals experienced a markedly elevated risk of VTE compared with wild-type individuals (odds ratio [OR] = 5.24; 95% confidence interval [CI], 4.01-6.84; P = 4.8 × 10−34), which approximated the risk conferred by FVL homozygosity. A secondary analysis restricted to UKB participants (N = 445 144) found that effect size estimates for the DH genotype remained largely unchanged (OR = 4.53; 95% CI, 3.42-5.90; P < 1 × 10−16) after adjustment for commonly cited VTE risk factors, such as body mass index, blood type, and markers of inflammation. In contrast, the DH genotype was not associated with a significantly higher risk of any arterial thrombosis phenotype, including stroke, myocardial infarction, and peripheral artery disease. In summary, we leveraged population-scale genomic data sets to conduct, to our knowledge, the largest study to date on the DH genotype and were able to establish far more precise effect size estimates than previously possible. Our findings indicate that the DH genotype may occur as frequently as FVL homozygosity and may confer a similarly increased risk of VTE.

A new model construction based on the knowledge graph for mining elite polyphenotype genes in crops.

Identifying polyphenotype genes that simultaneously regulate important agronomic traits (e.g., plant height, yield, and disease resistance) is critical for developing novel high-quality crop varieties. Predicting the associations between genes and traits requires the organization and analysis of multi-dimensional scientific data. The existing methods for establishing the relationships between genomic data and phenotypic data can only elucidate the associations between genes and individual traits. However, there are relatively few methods for detecting elite polyphenotype genes. In this study, a knowledge graph for traits regulating-genes was constructed by collecting data from the PubMed database and eight other databases related to the staple food crops rice, maize, and wheat as well as the model plant Arabidopsis thaliana. On the basis of the knowledge graph, a model for predicting traits regulating-genes was constructed by combining the data attributes of the gene nodes and the topological relationship attributes of the gene nodes. Additionally, a scoring method for predicting the genes regulating specific traits was developed to screen for elite polyphenotype genes. A total of 125,591 nodes and 547,224 semantic relationships were included in the knowledge graph. The accuracy of the knowledge graph-based model for predicting traits regulating-genes was 0.89, the precision rate was 0.91, the recall rate was 0.96, and the F1 value was 0.94. Moreover, 4,447 polyphenotype genes for 31 trait combinations were identified, among which the rice polyphenotype gene IPA1 and the A. thaliana polyphenotype gene CUC2 were verified via a literature search. Furthermore, the wheat gene TraesCS5A02G275900 was revealed as a potential polyphenotype gene that will need to be further characterized. Meanwhile, the result of venn diagram analysis between the polyphenotype gene datasets (consists of genes that are predicted by our model) and the transcriptome gene datasets (consists of genes that were differential expression in response to disease, drought or salt) showed approximately 70% and 54% polyphenotype genes were identified in the transcriptome datasets of Arabidopsis and rice, respectively. The application of the model driven by knowledge graph for predicting traits regulating-genes represents a novel method for detecting elite polyphenotype genes.

Neural longitudinal mapping of multidimensional performance profiles of Latin American universities

Objective. To introduce an artificial intelligence method and to use it to analyze the evolution of the multidimensional performance profiles of the most prominent Latin American universities, according to the Times Higher Education Latin America University Rankings.&#x0D; Design/Methodology/Approach. To multidimensionally characterize universities' performance profiles, we use the rankings' sub-scores indicators, which quantify five dimensions of academic endeavor assessed by this ranking. To automatically compare and visually analyze performance profile's evolution our method uses an artificial neural network.&#x0D; Results/Discussion. The neurocomputational procedure allowed us to discover all the characteristic performance profiles of the 50 best ranked universities (20 institutional profiles in 2019), and to visualize, in a knowledge map, the universities' groups sharing similar profiles. Furthermore, the profile's evolution of this group of universities was analyzed, and visually displayed in a sequence of knowledge maps, covering the four years period 2016-2019. In general, these universities show a remarkable improvement in Teaching, Research and Citation scores during 2016-2019. The profiles diversity of the best ranked universities, and the predominance and homogenization process of Brazilian universities profiles are noteworthy.&#x0D; Conclusions. Performance profile characterization using multiple indicators is a matter of interest in diverse domains. However, visualization or comparison of multidimensional performance profiles is not an easy task for the human mind. Even more challenging is the visual analysis of multidimensional performance profiles evolution. The neuro-longitudinal technique introduced is a useful tool to analyze and visualize the evolution of multidimensional performance profiles.&#x0D; Originality/Value. The approach and techniques introduced in the paper have an important degree of generality and can be used to analyze other type of rankings or multidimensional data.

Random subsequence forests

The random forest classifier is widely used in different fields due to its accuracy and robustness. Since its invention, the random forest algorithm is naturally developed for multi-dimensional vectorial data since features can be directly sampled during the decision tree construction procedure. In the context of discrete sequence classification, an explicit feature set is not readily available and we need to employ a feature extraction algorithm before building the random forest. However, such a predefined feature subset may limit the diversity of decision trees since the set of candidate features is composed of all subsequences. As a result, the predictive accuracy of constructed random forest classifier may be reduced. To address this, we propose a new algorithm that is able to directly build a random forest by choosing features from the set of all subsequences adaptively. To improve the running efficiency of our algorithm, the count-suffix tree is utilized to facilitate the fast frequency counting of subsequences so as to accelerate the generation of each randomized decision tree. The experimental results on 15 real datasets show that our method can outperform those state-of-the-art classification algorithms in terms of the predictive accuracy. The source code of our method can be found at: https://github.com/JiaqiWang-dlut/RSForest.

Abstract P278: 1-year Outcomes of A1c Reduction, Weight Loss, and Lowered QRISK3 Scores in Type 2 Diabetes Remission: Insights From an RCCT Leveraging Whole-Body Digital Twin Technology

Background and Aims: The effects of digital twin (DT) technology for remission of T2D on cardiovascular risk reduction is unknown. We evaluated the effectiveness of the DT to improve A1c and body weight and QRISK3 (QResearch Cardiovascular Disease Risk Algorithm 3), which is a prediction algorithm to estimate the 10-year risk for developing cardiovascular disease (CVD), in patients enrolled to achieve remission of T2DM. DT platform uses AI and Internet of Things, to integrate multi-dimensional data to give precision nutrition and health recommendations via the mobile app and by coaches. Materials and Methods: We evaluated the data from prospective cohort of 204 participants who had been on DT for 1 year. Remission was defined as A1C levels less than 6.5% without medication for over 3 months. Outcomes included the change in HbA1c, body weight and change in QRISK3 scores. The DT uses a machine learning algorithm to integrate clinical and sensor data to predict personal glucose response. Patients were connected to continuous glucose monitoring (CGM) throughout the study and self-recorded dietary intake using the mobile app. Results: The study participants had a mean diabetes duration of 3.7±2.6 years and a mean age of 44±8.2 years. Based on the ADA criteria, at 1 year, 73% (n=149/204) continued to be under remission. The mean QRISK3 at baseline was 12.6 (±10.7, 95% CI 11.2 to 14) which significantly reduced to 9.5 (±7.9, 95% CI 8.5 to 10.6) at 360 days, p=0.0009. A1c (%) at baseline was 8.9 (±1.8, 95% CI 8.6 to 9.1, minimum 5.5, maximum 16.2) which reduced to 6 (±0.6, 95% CI 5.9 to 6, minimum 4.7, maximum 9.2), p&lt;0.0001. Body weight (kg) at baseline was 78.3 (±14.2, 95% CI 76.4 to 80.2) which reduced to 70.7 (±12.8, 95% CI 69 to 72.4), &lt;0.0001. There was a significant correlation between the reduction in body weight and QRISK3 (Pearson r 0.24 95% CI 0.1 to 0.3, p=0.0004). However, the correlation between the reduction in A1c and QRISK3 (Pearson r 0.04, 95% CI -0.09 to 0.18, p=0.53 ns) was comparable. At baseline, there were 53% of participants who had low risk score (&lt;10), 28% as intermediate risk (10-20), 19% as high risk (&gt;20%) that at 1 year shifted to 67% as low risk, 23% as intermediate risk and 10% as high-risk scores. Conclusion: Incorporating the Digital Twin (DT) technology, which blends AI and IoT for tailored health guidance, our RCCT indicated significant 1-year advancements in A1c, weight, and QRISK3 scores among T2DM patients aiming for remission. The transformative potential of DT is evident, particularly in shifting a majority of participants towards lower cardiovascular risk categories.

Abstract P468: Correlation Among A1c Improvement, Weight Reduction, and Decreased ASCVD Risk in Participants Enrolled for Type 2 Diabetes Remission: Outcomes at 1 Year From an RCCT Using Whole-Body Digital Twin Technology

Background and Aims: The effects of digital twin (DT) technology for remission of T2D on cardiovascular risk reduction is unknown. We evaluated the effectiveness of the DT to improve A1c and body weight and ASCVD risk, in patients enrolled to achieve remission of T2DM. DT platform uses AI and Internet of Things, to integrate multi-dimensional data to give precision nutrition and health recommendations via the mobile app and by coaches. Materials and Methods: Data from 208 participants who had been on DT for 1 year were analyzed. Remission was defined as A1C levels less than 6.5% without medication for over 3 months. Outcomes included the change in HbA1c, body weight and change in ASCVD risk scores. The DT uses a machine learning algorithm to integrate clinical and sensor data to predict personal glucose response. Patients were connected to continuous glucose monitoring (CGM) throughout the study and self-recorded dietary intake using the mobile app. Results: The study participants had a mean diabetes duration of 3.7±2.7 years and a mean age of 44±8.5 years. Based on the ADA criteria, at 1 year, 72.6% (n=151/208) continued to be under remission. The mean 10-year ASCVD risk (%) at baseline was 7.9 (±7.8, 95% CI 6.8 to 8.9) which reduced to 3.6 (±4, 95% CI 3.1 to 4.2) at 360 days, p&lt;0.0001. A1c (%) at baseline was 8.9 (±1.8, 95% CI 8.7 to 9.2, minimum 5.5, maximum 16.2) which reduced to 6 (±0.6, 95% CI 5.9 to 6.1, minimum 4.7, maximum 9.2), p&lt;0.0001. Body weight (kg) at baseline was 78 (±14.3, 95% CI 76.1 to 79.9) which reduced to 70.4 (±12.9, 95% CI 68.7 to 72.1), p&lt;0.0001. There was a significant correlation between the reduction in A1c and ASCVD (Pearson r 0.41, 95% CI 0.31 to 0.5, p&lt;0.0001). Similarly, there was a significant correlation between the reduction in body weight and ASCVD (Pearson r 0.25, 95% CI 0.12 to 0.3, &lt;0.0001). At baseline, there were 49% of participants who had low risk score (&lt;5%), 16% as borderline risk (5-7.5%), 27% as intermediate risk (7.5-19.9), 8% as high risk (&gt;20%) that at 1 year shifted to 77% as low risk, 11% as borderline, 11% as intermediate risk and 1% as high-risk scores. Conclusion: The implementation of digital twin (DT) technology, utilizing AI and Internet of Things, demonstrated significant potential in T2DM management in participants enrolled to achieve remission. In our study with 208 participants, DT effectively reduced A1c levels, body weight, and ASCVD risk scores over 360 days. The correlations between reductions in A1c, body weight, and ASCVD underscore the potential of DT as a transformative tool in diabetes and cardiovascular risk management.

Predicting Student Performance in Online Learning: A Multidimensional Time-Series Data Analysis Approach

As an emerging teaching method, online learning is becoming increasingly popular among learners. However, one of the major drawbacks of this learning style is the lack of effective communication and feedback, which can lead to a higher risk of students failing or dropping out. In response to this challenge, this paper proposes a student performance prediction model based on multidimensional time-series data analysis by considering multidimensional data such as students’ learning behaviors, assessment scores, and demographic information, which is able to extract the characteristics of students’ learning behaviors and capture the connection between multiple characteristics to better explore the impact of multiple factors on students’ performance. The model proposed in this paper helps teachers to individualize education for students at different levels of proficiency and identifies at-risk students as early as possible to help teachers intervene in a timely manner. In experiments on the Open University Learning Analytics Dataset (OULAD), the model achieved 74% accuracy and 73% F1 scores in a four-category prediction task and was able to achieve 99.08% accuracy and 99.08% F1 scores in an early risk prediction task. Compared with the benchmark model, both the multi-classification prediction ability and the early prediction ability, the model in this paper has a better performance.

Quantifying the Influence of Climate Variables on Vegetation Through Remote Sensing and Multi-dimensional Data Analysis

Quantifying the Influence of Climate Variables on Vegetation Through Remote Sensing and Multi-dimensional Data Analysis

Artificial intelligence and multimodal data fusion for smart healthcare: topic modeling and bibliometrics

Advancements in artificial intelligence (AI) have driven extensive research into developing diverse multimodal data analysis approaches for smart healthcare. There is a scarcity of large-scale analysis of literature in this field based on quantitative approaches. This study performed a bibliometric and topic modeling examination on 683 articles from 2002 to 2022, focusing on research topics and trends, journals, countries/regions, institutions, authors, and scientific collaborations. Results showed that, firstly, the number of articles has grown from 1 in 2002 to 220 in 2022, with a majority being published in interdisciplinary journals that link healthcare and medical research and information technology and AI. Secondly, the significant rise in the quantity of research articles can be attributed to the increasing contribution of scholars from non-English speaking countries/regions and the noteworthy contributions made by authors in the USA and India. Thirdly, researchers show a high interest in diverse research issues, especially, cross-modality magnetic resonance imaging (MRI) for brain tumor analysis, cancer prognosis through multi-dimensional data analysis, and AI-assisted diagnostics and personalization in healthcare, with each topic experiencing a significant increase in research interest. There is an emerging trend towards issues such as applying generative adversarial networks and contrastive learning for multimodal medical image fusion and synthesis and utilizing the combined spatiotemporal resolution of functional MRI and electroencephalogram in a data-centric manner. This study is valuable in enhancing researchers’ and practitioners’ understanding of the present focal points and upcoming trajectories in AI-powered smart healthcare based on multimodal data analysis.

The effect of social media marketing on voting intention; an application of multidimensional panel data

PurposeThis study examines the effect of social media marketing on voting intention applying a combination of fuzzy logic methodology and a multidimensional panel data model.Design/methodology/approachThe study adopts a multidimensional panel data method that includes several fixed effects. The dependent variable is a multifaceted construct that measures the participants’ intention to vote. The independent variables are electronic word of mouth (eWOM), customisation (CUS), entertainment (ENT), interaction (INT), trendiness (TRD), candidate’s perceived image (CPI), religious beliefs (RB), gender and age. The grouping variables that signify fixed effects are employment status, level of education, mostly used social media and religion. First, the significance of said fixed effects was tested through an ANOVA process. Then, the main model was estimated, including the significant grouping variables as fixed effects.FindingsEmployment status and level of education were significant fixed effects. Also, eWOM, ENT, INT, CPI, RB and gender significantly affected participants’ voting intention.Research limitations/implicationsBeing based on a questionnaire that asked participants about how they perceive different aspects of social media, the present study is limited to their perceptions. Therefore, further studies covering the voters’ behaviour in action could be efficient complements to the present study.Practical implicationsThe findings could guide the political parties into prioritizing the aspects of social media in forming an effective campaign resulting in being elected.Social implicationsThe findings have the potential to help the public in making better informed decisions when voting. Furthermore, the results of this study indicate applications for social media which are beyond leisure time fillers.Originality/valueFuzzy logic and multidimensional panel data estimates are this study’s novelty and originality. Structural equation modelling and crisp linguistic values have been used in previous studies on social media’s effect on voting intent. The former refines the data gathered from a questionnaire, and the latter considers the possibility of including different grouping factors to achieve a more efficient and less biased estimation.

A Secure and Fast Range Query Scheme for Encrypted Multi-Dimensional Data

In recent years, more and more data has been stored on the cloud to provide various services. These data often contain users' private information, which inevitably raises concerns about data security. Encryption before outsourcing is a direct solution to mitigate these concerns. However, traditional encryption schemes such as block encryption make basic data services hard to support. Therefore, this paper proposes a secure and fast range query scheme for encrypted multi-dimensional data, called SFRQ. The scheme constructs a secure index over the ciphertexts of multi-dimensional data, utilizing the R-tree index, Bloom filter, and 0-1 encoding techniques. This secure index enables the cloud to provide fast range query services over the ciphertexts of multi-dimensional data. The authors have evaluated SFRQ through extensive experiments, which demonstrate its high efficiency. Additionally, the security analysis shows that no external entity, including the cloud, can obtain additional information during the entire query process.

A multicenter high-quality data registry for advanced proton therapy approaches: the POWER registry

BackgroundPaucity and low evidence-level data on proton therapy (PT) represent one of the main issues for the establishment of solid indications in the PT setting. Aim of the present registry, the POWER registry, is to provide a tool for systematic, prospective, harmonized, and multidimensional high-quality data collection to promote knowledge in the field of PT with a particular focus on the use of hypofractionation.MethodsAll patients with any type of oncologic disease (benign and malignant disease) eligible for PT at the European Institute of Oncology (IEO), Milan, Italy, will be included in the present registry. Three levels of data collection will be implemented: Level (1) clinical research (patients outcome and toxicity, quality of life, and cost/effectiveness analysis); Level (2) radiological and radiobiological research (radiomic and dosiomic analysis, as well as biological modeling); Level (3) biological and translational research (biological biomarkers and genomic data analysis). Endpoints and outcome measures of hypofractionation schedules will be evaluated in terms of either Treatment Efficacy (tumor response rate, time to progression/percentages of survivors/median survival, clinical, biological, and radiological biomarkers changes, identified as surrogate endpoints of cancer survival/response to treatment) and Toxicity. The study protocol has been approved by the IEO Ethical Committee (IEO 1885). Other than patients treated at IEO, additional PT facilities (equipped with Proteus®ONE or Proteus®PLUS technologies by IBA, Ion Beam Applications, Louvain-la-Neuve, Belgium) are planned to join the registry data collection. Moreover, the registry will be also fully integrated into international PT data collection networks.

Data-driven integrated study on operational performance degradation detection and recovery control of VFDR

The Variable Flow Ducted Rocket (VFDR) serves as the preferred propulsion system for hypersonic aircraft. However, it is prone to health degradation issues due to component failures, aging, and changes in operational conditions, which significantly challenge flight safety. This study introduces a data-driven integrated control approach for modeling, diagnosing, and remedying performance deterioration. Initially, a method for open-set modeling of performance degradation is developed using the Linear Parameter-Varying (LPV) modeling technique and gap metric theory, tackling the hurdles of data collection, overlapping fault classes, and modeling of unidentified faults in the VFDR system; thus, providing vital data support for detecting performance degradation. Moreover, employing the LPV model for performance degradation, a hybrid deep neural network combining multi-scale CNN and LSTM is trained to diagnose system performance issues online using multi-dimensional time-domain sequential data. Subsequently, a smooth transition control method utilizing sliding mode compensation is applied to regain control of system performance based on diagnostic outcomes and a pre-established VFDR performance recuperation controller. Simulations of three types of performance deterioration scenarios caused by changes in actuators, sensors, and the system's operational environment are conducted to demonstrate the efficiency of the proposed approach. The simulation outcomes affirm that the proposed approach is capable of effectively modeling system performance deterioration, diagnosing faults, and controlling recovery, offering promising technical support for the secure operation of intricate closed-loop control systems with stringent precision and reliability requirements.

The Implementation Effect of China’s River and Lake Chief System

The river and lake chief system offers a valuable policy toolkit to mitigate the degradation of water ecology, thereby bolstering water resource management for sustainable water development in China. To evaluate the effects of implementation and improve policy, this study took Beijing as a typical case and conducted a quantitative assessment using multidimensional data. The findings suggest that while the river and lake chief system in Beijing is effective and has significantly contributed to the ecological management of rivers and lakes, there are also notable regional disparities and urban–rural divergences. In addition, human activities are the main sources of environmental pollution in rivers and lakes, which should be the focus of the river and lake chief system. The river and lake chief system needs to embed more public participation and cooperative governance. This research aids in better understanding China’s river and lake chief system for both researchers and practitioners, facilitating the advancement of the knowledge body of global water policy and governance.