课程文献中心 更多>>

摘要： Coronary artery disease (CAD) remains the leading cause of mortality in the United States, and despite decades of innovation in secondary prevention, recurrent cardiovascular events continue to occur, amassing an enormous clinical and economic bur...

关键词： Approximate computing   Configurable multiplier   Dynamic voltage and frequency scaling (DVFS)   Energy-efficient hardware design   Multi-objective optimization  

摘要： Approximate multipliers can provide significant energy savings for compute-intensive applications with inherent error tolerance. However, existing approximate multipliers often lack the runtime configurability required by dynamic voltage and frequency scaling (DVFS) systems, either due to their fixed architectures or their power-centric optimization that overlooks timing adaptability. This paper presents TSE-CAM, a multi-level configurable approximate multiplier developed through a two-stage structural evolution framework. TSE-CAM provides three delay–accuracy configurations with distinct critical-path delays, enabling configurable operating points for DVFS-oriented MAC arrays. The proposed two-stage evolution framework first employs multi-objective optimization to identify a high-accuracy configuration, then derives reduced-delay configurations through structural degradation guided by masking, pruning, and substitution strategies. Experimental results show that TSE-CAM reduces critical-path delay by 36.6% and power consumption by 28.6%, while maintaining the MRED of 0.80% and the NMED of $5.92\times 10^{-4}$. In VGG-16 inference, all configurations preserve accuracy drops within 2%. In the DVFS-oriented MAC-array evaluation, the low-delay mode achieves up to $1.42\times$ throughput and 40.7% energy saving.

关键词： Sensor fusion   SLAM   LiDAR odometry  

摘要： High-altitude downward-looking LiDAR odometry remains challenging due to point cloud sparsity, scale variation, and geometric degeneration. This paper presents NDT-CAM, a visual-assisted LiDAR odometry framework for high-altitude UAV surveying and mapping. The proposed method integrates segmented motion compensation, a scale perceptron, and visual constraints within an NDT-based registration framework. Segmented motion compensation is used to alleviate motion distortion under dynamic flight, while the scale perceptron adaptively adjusts the voxel size and local search range according to the spatial extent of the current scene. To improve robustness in geometrically weak regions, visual feature information is incorporated as an auxiliary consistency constraint into the iterative optimization. Experiments on both public and private datasets demonstrate that the proposed method achieves robust and competitive localization performance across different flight altitudes, LiDAR types, and platform settings. The method maintains real-time performance over an operating range covering altitudes from 0 m to 130 m and speeds from 0 m/s to 12 m/s. These results suggest that adaptive, visual-assisted NDT provides a practical solution for high-altitude UAV mapping under weak GNSS/IMU dependence. We open-source the code and dataset on https://***/huashu996/NDT-CAM.

关键词： Geometric feature extraction   Classification   Feature recognition   Sheet metal manufacturing   Process planning   Manufacturability  

摘要：

• •
  Geometric feature dataset of 7856 Sheet metal CAD models for multi-class classification.
• •
  Industrial dataset of 7000 sheet metal parts with manufacturable features and groups.
• •
  Random Forest and GRU model for CAD classification and process sequence generation.
• •
  Comparative evaluation of learning architectures to justify model selection.

关键词： Computer-aided design   CAD generative modeling   Image-guided modeling   Multimodal learning   Interpretable modeling  

摘要：

• •
  This paper introduces an interpretable framework for CAD generation and editing.
• •
  We proposes a modular architecture with CoT for human-centric design.
• •
  We build a new dataset, Mind-CAD EditBench, for instruction-driven CAD editing.

关键词： Breast cancer   Organoid   Organ-on-chip   Microfluidics   Drug testing   MCF-7   2D   Two-Dimensional   3D   Three-Dimensional   OoC   Organ-on-Chip   Breast Cancer Cell Line   PDMS   Polydimethylsiloxane   WHO   World Health Organization   FEA   Finite Element Analysis   CAD   Computer Aided Drawing   Re   Reynolds number   Pe   Peclet number   N-S   Navier-Stokes   N-P   Nernst-Planck  

摘要：

• •
  A low-cost Organ-on-Chip (OoC) device was developed for breast cancer organoids under controlled flow conditions.
• •
  MCF-7-derived organoids successfully maintained their structure over 15 days within the chip.
• •
  The moderate flow rate is considered as the most effective.
• •
  Dynamic flow improves drug efficacy.
• •
  The OoC platform mimics the breast tumor environment, offering potential for preclinical cancer research.

关键词： Abnormality   Artificial intelligence   Boosting   CAD   Deciduous   Deep learning   Dental   DNN models   Ensemble learning   Furcation detection   Image processing   Intraoral periapical radiographs   Mandible   Object detection   Radiographs   Segmentation  

摘要： In dentistry, Artificial Intelligence (AI) plays a crucial role in transforming the process of disease detection and diagnosis. Among the numerous applications, AI is particularly valuable in the early detection of dental problems, like furcation involvement. This problem is especially common in children’s teeth, can penetrate deep into the tooth structure and potentially affect the developing permanent teeth. If not treated promptly, it may lead to serious problems, including tooth root bifurcation and eventual tooth loss. To tackle this problem, the present research introduces FD-CAD (Furcation Detection Computer-Aided Diagnosis), a state-of-the-art Computer-Aided Diagnosis system developed to detect furcation involvement in radiographs of deciduous mandibular molars. This FD-CAD system leverages advanced Deep learning (DL) models optimizing the Intraoral periapical radiographs (IOPA), including You Look Only Once (YOLO) models, Faster region-based convolutional neural network (Faster-RCNN), Fully convolutional one-stage object detector (FCOS), and RetinaNet. A comprehensive performance evaluation of these models was conducted, and Faster-RCNN emerged as the top contender. Furthermore, the diagnostic capability of the system was significantly enhanced using a domain-adapted boosting-based approach, Deciduous detection boosting (DDB). Moreover, this enhancement led to a significant improvement in performance, achieving an Average precision (AP) of 96.62% at an IoU threshold of 0.5 (AP50) and 95.39% at an Intersection over Union (IoU) of 0.7 (AP70). Additionally, these results demonstrate the reliability and robustness of the proposed method in detecting furcation involvement in deciduous teeth. In conclusion, these findings highlight the potential of DL in transforming early dental diagnosis, particularly in pediatric dentistry, and set the stage for more effective and preventive dental care.

关键词： CAM   Co-training   Dual-branch   Pseudo-label   Weakly supervised semantic segmentation  

摘要： Weakly Supervised Semantic Segmentation (WSSS) remains highly challenging because image-level supervision typically produces class activation maps (CAMs) that are incomplete or noisy when used as pixel-level pseudo-labels. Despite the architectural efficiency of one-stage approaches, they are often hindered by tight encoder-label coupling: CAMs and segmentation predictions are derived from the same encoder and optimized jointly, leading to the propagation and reinforcement of initial CAM inaccuracies by the segmentation outputs. To circumvent this limitation, we propose SynerNet, a one-stage dual-branch framework that explicitly mandates complementary yet synergistic objectives: one branch generates broad pseudo-labels to enhance coverage, while the other produces precise pseudo-labels to sharpen localization. With such pseudo-labels, the segmentation network yields simultaneously comprehensive and accurate predictions. The broad branch (B-CAM) leverages global attention to expand foreground coverage by guiding ambiguous regions toward likely foreground, whereas the precise branch (P-CAM) emphasizes fine localization by encouraging unreliable pixels toward the background. Through cross-supervision, the two branches effectively decouple the optimization process, alleviating the risk of error reinforcement inherent in direct coupling. To further integrate their strengths, we introduce a confidence matrix derived from multi-scale ViT features, in which pixels consistently classified across layers are treated as high-confidence, while inconsistent ones are marked as uncertain. This enables a confidence-guided fusion strategy that directly adopts reliable predictions and adaptively blends uncertain regions using contributions from both branches. Such a complementary design mitigates error reinforcement and promotes mutually beneficial learning, enabling the network to generate high-fidelity pseudo-labels in a fully end-t

关键词： AD   Alzheimer’s disease   AML   Acute myeloid leukemia   ARD   Age-related disease   CAD   Coronary artery disease   CH   Clonal hematopoiesis   CHD   Coronary heart disease   CHIP   Clonal hematopoiesis of indeterminate potential   CHIPI   Clonal hematopoiesis in predisposed individuals   CKD   Chronic kidney disease   COPD   Chronic obstructive pulmonary disease   CVD   Cardiovascular disease   HF   Heart failure   HSC   Hematopoietic stem cell   Hs-CRP   High-sensitivity C-reactive protein   HSPC   Hematopoietic stem progenitor cell   IL   Interleukin   MDS   Myelodysplastic syndrome   MPN   Myeloproliferative neoplasm   NLRP3   NLR family pyrin domain containing 3   TNF-α   Tumor necrosis factor-alpha   T2D   Type 2 diabetes   VAF   Variant allele frequency   Clonal hematopoiesis of undetermined potential (CHIP)   inflammaging   age-related diseases (ARDs)   Omics techniques   ARD biomarker and target   artificial intelligence  

摘要： Clonal hematopoiesis of undetermined potential (CHIP), initially discovered as a mere hematological curiosity, now represents a clinically significant factor that is changing our view of ageing, inflammaging and the risk of age-related diseases (ARDs). In line with this, new meta-analyses link the presence of CHIP to higher all-cause mortality from ARDs and to an increase in adverse events in the presence of larger clones. Mechanistically, CHIP induces inflammaging, but a bidirectional relationship has also been highlighted. In both cases, the pro-inflammatory pathways primarily involved are IL-1β/IL-6 signaling, activation of the NLR family pyrin domain containing 3 (NLRP3) inflammasome, and Toll-Like Receptors (TLRs). However, these pathways only partially explain the role of CHIP in the complex pathophysiology of ARDs. Despite this, it is proposing it as an ARDs biomarker and pharmacological target for their prevention. However, the clinical risk of CHIP triggering the onset and severe progression of ARDs depends on the mutation type, clone size, co-mutations, and host and environmental factors, and current high-risk models require adjustments to overcome some limitations. Omics technique can likely be integrated into CHIP mutation sequencing to predict and monitor ARDs risk. Consequently, pragmatic screening strategies currently remain limited. Furthermore, although potential therapeutic targets exist, translating the findings into effective interventions is currently challenging. Ethical, patient-centered clinical trial designs with validated endpoints are essential to avoid over medicalization and ensure safety, feasibility, and generalizability, as is the use of artificial intelligence. An overview of these aspects is reported in such review.

关键词： Cellular beams   Inverse design   Surrogate modeling   Physics-informed machine learning   Structural design automation  

摘要：

• •
  Physics-informed surrogate from 14,094 FE models enables fast cellular beam design.
• •
  Inverse design framework identifies feasible beam geometries from target resistance.
• •
  Code rules, fabrication limits and robustness are integrated in a unified workflow.
• •
  Structural performance and sustainability metrics are assessed separately for trade-offs.