Search Results (25)

Graphical methods of correlation analysis, such as correlation n-ptychs or hotspots, focus on the identification of the strength and direction of functional relationships between sets of attributes in multidimensional datasets. Since these correlation structures only take into account values of the attributes, situations arise when the relationship is coincidental, meaning that there is no real-world causality between the values of the observed attributes but these values still exhibit significant correlation. This problem of correlation analysis as a whole motivates the need for semantic evaluation of significant relationships identified using its methods—a task that could potentially be time- and resource-intensive when conducted manually. However, modern results in the large language model area provide tools for the automatization of such tasks. Hence, this work focuses on the design and implementation of a novel large language model-based method for semantic evaluation of correlation structures embedded in a correlation graph, specifically correlation n-ptychs for $n\in \{3, 4, 5\}$ and correlation hotspots. In the method, the large language model is automatically prompted to assess the semantic nature of relationships in the set of correlation substructures of the dataset, identify their real-world relevance, and visualize the result in the form of a Semantic evaluation card. The proposed approach is evaluated using two benchmarking datasets focusing on the visualization method used in the model, large language model interaction with the correlation substructures, and comparative analysis with previously used tools in the area. Full article

Moroccan ICH is a rich domain that remains difficult to structure formally due to the heterogeneity of textual descriptions and the diversity of documented cultural practices. This article proposes a semi-automatic and adaptable methodological framework for constructing an ontology-driven knowledge graph from unstructured texts, applied to Moroccan ICH. The approach begins by classifying documents into the five predefined UNESCO categories using lexical, semantic, and hybrid methods, followed by intra-category semantic clustering to identify thematic substructures that inform ontological modeling. The results show that hybrid approaches achieve the best performance in document classification, while clustering requires an adaptive strategy for each category. Building on these stages, ICHOnto was generated as a CIDOC CRM-aligned ontology enriched with UNESCO categories, expert-validated subcategories, and entities and relations extracted from the texts. The resulting resource was evaluated through logical consistency, SHACL compliance, and functional assessment using Competency Questions. The evaluation confirms that ICHOnto provides a coherent, exploitable, and interoperable semantic resource for representing, organizing, and querying Moroccan ICH. Its modular structure also supports adaptation to other heritage corpora or domains based on unstructured textual data. Full article

Objective: To develop and validate an artificial intelligence (AI) system for automated lens opacities classification system III (LOCS III)-based grading of all three major cataract subtypes using anterior segment optical coherence tomography (AS-OCT). Methods: This is a single-center cross-sectional study. AS-OCT images were collected and manually graded by ophthalmologists according to LOCS III. The dataset was randomly split into training, validation, and test sets. We propose a novel multi-granularity mask-guided network (MMNet) that jointly performs lens substructure segmentation and severity grading. The model’s performance was assessed on an independent test set for automatic grading of cortical cataract (CC), nuclear cataract (NC), and posterior subcapsular cataract (PSC) and the grading performance of the proposed method against ophthalmologists was also evaluated. The model’s interpretability was assessed via attention heatmaps and feature visualization. Results: The proposed MMNet exhibited high agreement with ground truth conducted through gold standard. The proportions of predictions with an absolute error < 1.0 for three subtypes range from 83.02% to 89.94%. The model’s grading accuracy for cataract subtypes was between 82.20 ± 1.41% and 89.76 ± 1.31% among the three subtypes, the Area Under the Curve (AUC) was between 0.954 (95% CI, 0.952–0.969; p < 0.001) and 0.973 (95% CI, 0.964–0.985; p < 0.001). The MMNet shows a satisfactory mean absolute error (MAE) of 0.14 ± 0.35 in CC, 0.10 ± 0.30 in NC, and 0.17 ± 0.38 in PSC grading. It also achieved a fast grading speed of 0.0178 s/image against manual grading. Conclusions: The proposed AI model presented advanced performance on AS-OCT images in automated LOCS III-based cataract grading for CC and NC, and also showed feasibility in PSC assessment. Full article

The automatic generation of semantic Level of Detail (LOD) 2 models from TomoSAR point clouds is frequently compromised by elevation side-lobes, data sparsity, and inherent geometric distortions. In particular, the energy dispersion caused by side-lobes blurs vertical structures, making the extraction of floor details and accurate floor height estimation significantly challenging. To overcome these limitations, we present a refined reconstruction framework that tightly couples tomographic imaging mechanisms with building geometric priors. For fine-grained vertical reconstruction, we employ a geometry-constrained inverse projection strategy that concentrates scattered energy back onto the building façade to mitigate side-lobe interference. This is complemented by a Global Coherent Integration method, utilizing spectral analysis to robustly recover periodic floor patterns and estimate average floor heights. In the horizontal domain, we address the conflict between noise suppression and feature preservation through a separation-of-axes morphological strategy. Unlike traditional isotropic filtering, this approach processes orthogonal directions independently to bridge data gaps while strictly maintaining sharp building corners and recovering fine substructures. Validated on airborne Ku-band datasets, the proposed method demonstrates the capability to produce topologically complete and semantically rich urban models from sparse radar observations. Full article

Statistical Energy Analysis (SEA) requires a physically meaningful subsystem definition, whereas manual partitioning of complex shell structures is often time-consuming and strongly dependent on engineering experience. To address this issue, this study proposes an automatic initial subsystem partitioning framework for shell FE models based on explicit prior attributes available in the model definition. The method unifies four classes of physical discontinuities—geometric discontinuity, thickness discontinuity, material/property discontinuity, and topological discontinuity—within a single adjacency evaluation procedure. The shell FE mesh is represented through element adjacencies, and adjacencies crossing any identified physical discontinuity are removed so that the remaining connected components define the partitioned subsystems. In this way, the framework generates partitioning results with explicit boundaries and traceable origins without relying on posterior response-field analysis or manually prescribed subsystem boundaries. Because the procedure operates directly on existing large-scale shell FE models and does not require additional response-feature construction or complex pre-partitioning, it provides a lightweight, repeatable, and practically executable automation path for SEA-related front-end modeling. The resulting partitions are intended as physically explicit initial partitioning results that provide a reliable boundary basis for higher-level statistical modeling objectives. When a coarser subsystem representation is required for subsequent modeling, further aggregation may be introduced as an optional enhancement according to the modeling objective, rather than as a prerequisite for the validity of the present method. Full article

Direct collocation transcription is a dominant technique for solving complex optimal control problems, converting continuous dynamics into large-scale, sparse nonlinear programming problems. The computational efficiency of this approach is fundamentally limited by the evaluation of first- and second-order derivatives required by modern optimization algorithms. While general-purpose automatic differentiation tools exist, they often fail to fully exploit the repetitive substructure inherent in trajectory discretization. This paper presents a vectorized, sparse, second-order forward automatic differentiation framework specifically tailored for direct collocation methods. By explicitly distinguishing between scalar and vector nodes within the expression graph, the proposed method leverages the independence of mesh point evaluations to enable Single Instruction, Multiple Data (SIMD) execution and optimize memory access patterns. This structure-aware approach ensures linear time complexity with respect to the number of discretization nodes while maintaining the flexibility to handle complex dependencies. The methodology is implemented in the open-source software package pockit and is validated through three distinct engineering case studies: the aggressive stabilization of a nano-quadrotor, the powered descent guidance of a reusable launch vehicle, and a low-thrust heliocentric orbital transfer. These applications demonstrate the framework’s capability to deliver high-performance derivative computation for large-scale, nonlinear dynamical systems. Full article

Optimal sensor placement (OSP) for structural health monitoring of large-scale space grid structures must enable reliable identification of localized member deterioration with sparse instrumentation. Modal-based OSP criteria optimize observability of a healthy model but do not directly minimize uncertainty in substructure-level damage parameters. We partition the structure into substructures, simulate axial and biaxial bending stiffness-loss cases, and compute truncated modal flexibility. Each element is encoded by stacked end-node flexibility submatrices over $m=6$ modes. A multi-task, zero-anchored multi-layer perceptron is trained to regress three nonnegative damage parameters and classify damage presence using losses tailored for small-damage accuracy. Sensor sensitivities are obtained by automatic differentiation of the surrogate with respect to flexibility features and aggregated with scenario weights emphasizing critical bending and neighbor-substructure interference scenarios. A greedy D-optimal design then maximizes the log-determinant of a regularized Fisher information matrix under practical coverage constraints; substructure selections are merged into a globally feasible layout. On a representative space grid, the method improves task-oriented identifiability over EFI and MKE across budgets $K_{\mathrm{tot}}=30$–60 (higher-damage D-optimality, lower A-optimality trace, and reduced proxy variance indicators), while yielding lower modal log-determinants. These findings indicate risk-weighted, substructure-first task design as an alternative to purely modal criteria for substructure-level damage-parameter identification. Full article

We summarize the experience of studying 2D features in the galaxy distribution of galaxy cluster fields. For the detailed study of the inner structure of galaxy clusters, algorithms were developed for detecting various types of regular substructures inside such objects automatically. Substructures in galaxy clusters arise from interactions as well as the evolution of the cosmic web, but cannot be described according to the schemes of morphological classification, both classical and modern, because some regular substructures are not present. Our algorithms are based on numerical criteria that permit the determination of classical morphological types, connected with parameters such as the degree of concentration to the cluster center and/or to a straight line, on a statistically significant level. Other types of substructures can also be detected with corresponding algorithms. As a result, we can analyze intracluster features, such as crosses, semi-crosses, complex crosses, and compact dense chains. All algorithms are realized in the “Cluster Cartography” tool and can be used with data taken from different catalogs. The algorithms and their realization in program code must simplify, standardize, and speed up the analysis of 2D distributions of galaxies in clusters. It is possible in future to adapt the algorithms for the 3D case. The results of statistically valid morphological classification are useful for studies of the evolution of galaxy clusters. Full article

Background: Segmenting computed tomography (CT) is crucial in various clinical applications, such as tailoring personalized cardiac ablation for managing cardiac arrhythmias. Automating segmentation through machine learning (ML) is hindered by the necessity for large, labeled training data, which can be challenging to obtain. This article proposes a novel approach for automated, robust labeling using domain knowledge to achieve high-performance segmentation by ML from a small training set. The approach, the domain knowledge-encoding (DOKEN) algorithm, reduces the reliance on large training datasets by encoding cardiac geometry while automatically labeling the training set. The method was validated in a hold-out dataset of CT results from an atrial fibrillation (AF) ablation study. Methods: The DOKEN algorithm parses left atrial (LA) structures, extracts “anatomical knowledge” by leveraging digital LA models (available publicly), and then applies this knowledge to achieve high ML segmentation performance with a small number of training samples. The DOKEN-labeled training set was used to train a nnU-Net deep neural network (DNN) model for segmenting cardiac CT in N = 20 patients. Subsequently, the method was tested in a hold-out set with N = 100 patients (five times larger than training set) who underwent AF ablation. Results: The DOKEN algorithm integrated with the nn-Unet model achieved high segmentation performance with few training samples, with a training to test ratio of 1:5. The Dice score of the DOKEN-enhanced model was 96.7% (IQR: 95.3% to 97.7%), with a median error in surface distance of boundaries of 1.51 mm (IQR: 0.72 to 3.12) and a mean centroid–boundary distance of 1.16 mm (95% CI: −4.57 to 6.89), similar to expert results (r = 0.99; p < 0.001). In digital hearts, the novel DOKEN approach segmented the LA structures with a mean difference for the centroid–boundary distances of −0.27 mm (95% CI: −3.87 to 3.33; r = 0.99; p < 0.0001). Conclusions: The proposed novel domain knowledge-encoding algorithm was able to perform the segmentation of six substructures of the LA, reducing the need for large training data sets. The combination of domain knowledge encoding and a machine learning approach could reduce the dependence of ML on large training datasets and could potentially be applied to AF ablation procedures and extended in the future to other imaging, 3D printing, and data science applications. Full article

Capturing long-term dependency from historical behaviors is the key to the success of sequential recommendation; however, existing methods focus on extracting global sequential information while neglecting to obtain deep representations from subsequences. Previous research has revealed that the restricted inter-item transfer is fundamental to sequential modeling, and some potential substructures of sequences can help models learn more effective long-term dependency compared to the whole sequence. To automatically find better subsequences and perform efficient learning, we propose a sequential recommendation model with a gated recurrent unit and Transformers, abbreviated as GAT4Rec, which employs Transformers with shared parameters across layers to model users’ historical interaction sequences. The representation learned by the gated recurrent unit is used as the gating signal to identify the optimal substructure in user sequences. The fused representation of the subsequence and edge information is extracted by the encoding layer to make the corresponding recommendations. Experimental results on four well-known publicly available datasets demonstrate that our GAT4Rec model outperforms other recommendation models, achieving performance improvements of 5.77%, 1.35%, 11.58%, and 1.79% in the normalized discounted cumulative gain metric (NDCG@10), respectively. Full article

The automatic segmentation of cardiac computed tomography (CT) and magnetic resonance imaging (MRI) plays a pivotal role in the prevention and treatment of cardiovascular diseases. In this study, we propose an efficient network based on the multi-scale, multi-head self-attention (MSMHSA) mechanism. The incorporation of this mechanism enables us to achieve larger receptive fields, facilitating the accurate segmentation of whole heart structures in both CT and MRI images. Within this network, features extracted from the shallow feature extraction network undergo a MHSA mechanism that closely aligns with human vision, resulting in the extraction of contextual semantic information more comprehensively and accurately. To improve the precision of cardiac substructure segmentation across varying sizes, our proposed method introduces three MHSA networks at distinct scales. This approach allows for fine-tuning the accuracy of micro-object segmentation by adapting the size of the segmented images. The efficacy of our method is rigorously validated on the Multi-Modality Whole Heart Segmentation (MM-WHS) Challenge 2017 dataset, demonstrating competitive results and the accurate segmentation of seven cardiac substructures in both cardiac CT and MRI images. Through comparative experiments with advanced transformer-based models, our study provides compelling evidence that despite the remarkable achievements of transformer-based models, the fusion of CNN models and self-attention remains a simple yet highly effective approach for dual-modality whole heart segmentation. Full article

Movement and muscle control are crucial for the survival of all free-living organisms. This study aimed to explore differential patterns of cortical and subcortical activation across different stages of muscle control using functional magnetic resonance imaging (fMRI). An event-related design was employed. In each trial, participants (n = 10) were instructed to gently press a button with their right index finger, hold it naturally for several seconds, and then relax the finger. Neural activation in these temporally separated stages was analyzed using a General Linear Model. Our findings revealed that a widely distributed cortical network, including the supplementary motor area and insula, was implicated not only in the pressing stage, but also in the relaxation stage, while only parts of the network were involved in the steady holding stage. Moreover, supporting the direct/indirect pathway model of the subcortical basal ganglia, their substructures played distinct roles in different stages of muscle control. The caudate nucleus exhibited greater involvement in muscle contraction, whereas the putamen demonstrated a stronger association with muscle relaxation; both structures were implicated in the pressing stage. Furthermore, the subthalamic nucleus was exclusively engaged during the muscle relaxation stage. We conclude that even the control of simple muscle movements involves intricate automatic higher sensory–motor integration at a neural level, particularly when coordinating relative muscle movements, including both muscle contraction and muscle relaxation; the cortical and subcortical regions assume distinct yet coordinated roles across different stages of muscle control. Full article

Ceramic thickness and technicians’ manipulative variables are critical factors affecting the resultant shade of dental ceramo-metallic restorations. This study investigated the effect of the following variables on shade duplication of ceramo-metallic specimens: (a) ceramic thickness; (b) differences between several technicians (inter-technician variability); and (c) the ability of each technician to repeat the resultant shade (intra-technician variability). Ninety ceramo-metallic specimens were prepared and divided into three main groups (n = 30/gp) according to the different technicians who built up the veneering ceramic of the specimens. Each group was further subdivided into three subgroups (n = 10/subgroup) according to the thickness of the ceramic (1, 1.5, and 2 mm built over a 0.5 mm-thick metal substructure). Three different technicians were asked to follow the same protocol as regards the same ceramic batch (Shade 3M2, Vita VM13, Zahnfabrik, Germany), firing temperature, and number of firing cycles. Meanwhile, each technician followed his own protocol with regard to other ceramic manipulative variables. The duplicated shades of the specimens were investigated using the Vita Easyshade spectrophotometer by using the verify shade mode. Color difference (∆E) values were calculated between the target shade (3M2) and the duplicated shades of the specimens automatically by the Vita Easyshade spectrophotometer (Vita, Zahnfabrik, Germany). The effect of ceramic thickness and inter- and intra-technician variability on the duplication of the target shade was investigated. The results showed that the effect of ceramic thickness on the duplicated shades depended on inter-technician variability. High inter-technician variability (∆E = 2–6.4) was noticed in contrast to low intra-technician variability (∆E = 0.2–1.5). It could be concluded that proper shade-duplication of ceramo-metallic restoration was a cumulative technique intimately related to manipulative variables and ceramic thickness. Full article

Cardiac substructure segmentation is a prerequisite for cardiac diagnosis and treatment, providing a basis for accurate calculation, modeling, and analysis of the entire cardiac structure. CT (computed tomography) imaging can be used for a noninvasive qualitative and quantitative evaluation of the cardiac anatomy and function. Cardiac substructures have diverse grayscales, fuzzy boundaries, irregular shapes, and variable locations. We designed a deep learning-based framework to improve the accuracy of the automatic segmentation of cardiac substructures. This framework integrates cardiac anatomical knowledge; it uses prior knowledge of the location, shape, and scale of cardiac substructures and separately processes the structures of different scales. Through two successive segmentation steps with a coarse-to-fine cascaded network, the more easily segmented substructures were coarsely segmented first; then, the more difficult substructures were finely segmented. The coarse segmentation result was used as prior information and combined with the original image as the input for the model. Anatomical knowledge of the large-scale substructures was embedded into the fine segmentation network to guide and train the small-scale substructures, achieving efficient and accurate segmentation of ten cardiac substructures. Sixty cardiac CT images and ten substructures manually delineated by experienced radiologists were retrospectively collected; the model was evaluated using the DSC (Dice similarity coefficient), Recall, Precision, and the Hausdorff distance. Compared with current mainstream segmentation models, our approach demonstrated significantly higher segmentation accuracy, with accurate segmentation of ten substructures of different shapes and sizes, indicating that the segmentation framework fused with prior anatomical knowledge has superior segmentation performance and can better segment small targets in multi-target segmentation tasks. Full article

Purpose: In the past decade, there has been a rapid increase in the development of automatic cardiac segmentation methods. However, the automatic quality control (QC) of these segmentation methods has received less attention. This study aims to address this gap by developing an automatic pipeline that incorporates DL-based cardiac segmentation and radiomics-based quality control. Methods: In the DL-based localization and segmentation part, the entire heart was first located and cropped. Then, the cropped images were further utilized for the segmentation of the right ventricle cavity (RVC), myocardium (MYO), and left ventricle cavity (LVC). As for the radiomics-based QC part, a training radiomics dataset was created with segmentation tasks of various quality. This dataset was used for feature extraction, selection, and QC model development. The model performance was then evaluated using both internal and external testing datasets. Results: In the internal testing dataset, the segmentation model demonstrated a great performance with a dice similarity coefficient (DSC) of 0.954 for whole heart segmentations. Images were then appropriately cropped to 160 × 160 pixels. The models also performed well for cardiac substructure segmentations. The DSC values were 0.863, 0.872, and 0.940 for RVC, MYO, and LVC for 2D masks and 0.928, 0.886, and 0.962 for RVC, MYO, and LVC for 3D masks with an attention-UNet. After feature selection with the radiomics dataset, we developed a series of models to predict the automatic segmentation quality and its DSC value for the RVC, MYO, and LVC structures. The mean absolute values for our best prediction models were 0.060, 0.032, and 0.021 for 2D segmentations and 0.027, 0.017, and 0.011 for 3D segmentations, respectively. Additionally, the radiomics-based classification models demonstrated a high negative detection rate of >0.85 in all 2D groups. In the external dataset, models showed similar results. Conclusions: We developed a pipeline including cardiac substructure segmentation and QC at both the slice (2D) and subject (3D) levels. Our results demonstrate that the radiomics method possesses great potential for the automatic QC of cardiac segmentation. Full article