Search Results (240)

Non-invasive fetal electrocardiogram (FECG) extraction from maternal abdominal ECG (AECG) is crucial for prenatal monitoring but remains challenging due to strong interference from maternal ECG (MECG), baseline drift, and noise. We propose an FECG extraction method based on minimal channel attention (MCA) and flow matching (FM), learning a deterministic mapping from AECG to FECG via a probabilistic path. To balance the preservation of physiological signals and separation of interference, we employ bridge variance scheduling for the diffusion process. Target matching loss is introduced to regress the FECG directly, enhancing training stability and waveform fidelity. For feature selection, a minimal channel attention module with global average pooling and a single linear layer is embedded after feature extraction, capturing cross-channel dependencies with minimal parameters. Enhanced residual connections are incorporated to retain underlying features and optimize gradient flow in deep networks. Experiments on two public datasets (ADDB and BDDB) with a leave-one-out cross-validation strategy show that our method achieves average Pearson correlation coefficients (PCCs) of 0.94 ± 0.050 on ADDB and 0.91 ± 0.122 on BDDB, demonstrating robust performance across diverse real-world recording conditions. The method balances high accuracy with efficient feature extraction, offering a reliable solution for non-invasive fetal heart health monitoring. Full article

Bulk RNA-sequencing (bulk RNA-seq) averages gene expression across cell mixtures, obscuring single-cell heterogeneity and spatial architectures essential for understanding pathological processes. We developed HistoMap, a deep learning-based framework for single-cell spatial deconvolution. The model employs a two-stage pipeline: first, reconstructing high-fidelity single-cell profiles from bulk data using a β-variational autoencoder, and second, utilizing a Histological Vision Transformer (H-ViT) to map these cells to tissue coordinates via dual guidance from transcriptomic references and H&E-stained morphological constraints. HistoMap demonstrated superior performance across diverse human tissues, achieving a Pearson Correlation Coefficient (PCC) of 0.800 on external validation. Application to 14 colorectal cancer cases revealed a Macro_SPP1-mediated desmoplastic barrier. SPP1+ macrophages act as spatial hubs at the invasive front, forming a physical “sequestration belt” that functionally excludes cytotoxic T cells from the tumor core. HistoMap successfully bridges bulk RNA-seq and spatial single-cell architectures. Our findings provide a molecular rationale for immune checkpoint blockade resistance and identify the SPP1-fibroblast axis as a pivotal target for therapeutic sensitization. Full article

Non-contact body-temperature prediction from facial thermography is affected by pose, occlusion, missing measurements, and inter-subject variation. This study proposes a graph–tabular latent-representation fusion framework for predicting body temperature from thermal facial landmark profiles. A Pearson correlation coefficient (PCC)-guided landmark graph models landmark-to-landmark thermal dependencies. At the same time, the same landmark-temperature signal is retained as a tabular representation to preserve global temperature-pattern interactions. The graph and tabular branches are encoded independently, fused at the latent level, and trained for target-landmark temperature regression with auxiliary reconstruction losses. Experiments were conducted on TFD68 under complete, missing completely at random (MCAR), and structured missing not at random (MNAR) conditions. The structured MNAR simulation combines 3D head-pose visibility modelling, accessory-driven occlusion, validation against real TFD68 occlusion annotations, and graph-construction sensitivity analyses. Results show that selected fused configurations improve over strong stand-alone graph and tabular baselines, particularly under MNAR-imputed evaluation, with the best selected configuration reducing prediction error by approximately $6\%$. Statistical testing further confirms significant improvements in most MNAR fused–baseline comparisons. Accuracy–efficiency analysis shows that fusion improves robustness at the cost of additional inference time, providing a flexible design space for thermal landmark-based body-temperature prediction. Full article

This paper proposes a machine learning-based framework for predicting transient nitrogen oxide (NOx) emissions from non-road heavy-duty diesel engines under NRTC conditions. The model combines the PSO and XGBoost algorithms to predict transient NOx emissions from non-road heavy-duty diesel engines. Experimental results were obtained through engine bench tests. Pearson’s correlation coefficient (PCC), Spearman’s correlation coefficient (SCC), and SHAP analysis were used to evaluate the relationship between engine operating parameters and NOx emissions, thereby improving the interpretability of features and the reliability of input parameter selection. The hyperparameters of the XGBoost model were optimized using the PSO algorithm. The optimized model exhibits good predictive performance. On the training set, the model’s R² coefficient reached 0.9989, MAE was 1.16 ppm, and RMSE was 1.57 ppm. On the test set, the R² value was 0.9662, and the MAE and RMSE were 6.33 ppm and 8.97 ppm, respectively. Furthermore, the PSO-XGBoost model was compared and analyzed with six traditional models. Finally, the generalization of the PSO-XGBoost model and its underlying mechanisms were discussed. The results indicate that the proposed PSO-XGBoost framework can effectively capture nonlinear relationships among diesel engine operating parameters and achieve accurate transient NOx prediction for non-road heavy-duty diesel engines. Full article

Accurate maize yield prediction across diverse environments is pivotal for modern breeding programs. While machine learning (ML) excels at capturing non-linear environmental effects, Genomic Best Linear Unbiased Prediction (GBLUP) remains a benchmark for modeling polygenic small-effect contributions. However, principled integration of these paradigms—while explicitly accounting for genotype-by-environment interaction (G×E)—remains a formidable challenge. We propose a two-step framework evaluated on the Genomes to Fields (G2F) 2022 dataset. In Step 1, ML models are employed to fit environmental main effects; in Step 2, genomic residuals are modeled via additive-dominance relationship matrices, augmented by an explicit low-rank G×E matrix. Candidate interaction markers were screened through plasticity-based genome-wide association studies (GWAS) across six phenotypic stability metrics and used to construct a low-rank candidate G×E representation, with a cross-validation-selected scaling parameter applied to control the contribution of the predicted G×E component. TwoStep_G×E_alpha0.33, achieved a within–environment Pearson correlation coefficient (PCC) of 0.376, outperformed both GBLUP and the competition-winning model (PCC = 0.357) in within-environment ranking. Furthermore, environment-adaptive selection yielded a genetic gain of 0.454 Mg ha⁻¹, representing a 34.7% improvement over GBLUP. Overall, the proposed framework provides a practical approach for environment-specific yield prediction and adaptive selection in maize breeding. Full article

Respiratory rate is one of the most important vital signs. It affects ventilation which relates to oxygen inhalation and carbon dioxide elimination. Currently, only a handful of prototypes are available for estimating the respiratory rate under the condition that users remain completely still. This research focuses on the development of a respiratory rate monitoring system that can detect human respiratory signals using force sensitive resistors (FSRs). The FSR sensors measure the forces from respiratory motion and then signal processing techniques are employed to minimize background noise and artifacts. Respiratory data are processed by a microcontroller and transmitted via Bluetooth to a mobile device for further processing and visualization. The system performance was evaluated in three stages. Firstly, for the proof by simulation, a mean absolute error (MAE), root mean square error (RMSE), and Pearson correlation coefficient (PCC) of 0.26, 0.37 breaths per minute (bpm), and 0.9998 are achieved, respectively, even when the noise level is very high, i.e., power signal-to-noise ratio is 0.25 or −6.02 decibel. Secondly, for the test on a robot, the MAEs are 0.25, 0.53, and 0.75 bpm; the RMSEs are 0.28, 0.64, and 0.92 bpm; the PCCs are approximately 1, 0.9993, and 0.9986, respectively, under sitting, walking, and jogging conditions. The system is further deployed on 14 human subjects yielding MAEs of 0.51, 1.24, and 1.92 bpm; RMSEs of 0.65, 1.63, and 2.22 bpm; and PCCs of 0.9893, 0.9831, and 0.9655, for human sitting, walking, and jogging, respectively. In the future, this respiratory rate monitoring system could be applied to patients, elderly individuals, or the general population who experience movement or locomotion during monitoring. Full article

Spatial transcriptomics (ST) has emerged as an essential technology for interpreting the molecular profiles underlying pathological tissue morphology. Most existing ST analyses are limited to 2D sections, which ignore the complex structural and molecular heterogeneity of biological tissues in 3D space and may cause diagnostic oversights. Since acquiring complete 3D ST volumes is resource-intensive, recent 3D imputation paradigms provide a cost-effective alternative by integrating 3D whole-slide images (WSIs) with sparse 2D ST references (e.g., a single slide). Despite this methodological advancement, effectively modeling complex cross-layer spatial dependencies remains challenging. Current mainstream solutions predominantly adopt standard Transformers for cross-scale feature aggregation, which may bring computational overhead and higher overfitting risk while having limited explicit mechanisms for hierarchical anatomical guidance. To address these limitations, we propose a Hierarchical MambaOut-based Spatial Imputation Graph Network (HM-ASIGN) for anatomy-aware 3D spatial transcriptomics imputation. Our architecture leverages MambaOut’s dynamic gated 1D convolutions as a parameter-efficient alternative to dense global self-attention. This design captures the depth-wise evolution of pathological features while reducing over-parameterization. Inspired by the macro-to-micro diagnostic reasoning of clinical pathologists, HM-ASIGN introduces a multi-scale recursive guidance mechanism. It constructs a top-down information flow by extracting global anatomical priors at macroscopic scales and injecting them as contextual anchors into regional and spot-level features in a cascaded manner. This helps ensure that fine-grained molecular predictions are properly constrained by global morphological structures. Evaluation experiments on multiple public breast cancer datasets demonstrate that HM-ASIGN achieves competitive reference-level performance against existing baselines, reaching a Pearson Correlation Coefficient (PCC) of 0.772. Specifically, when evaluated against the foundational ASIGN framework, it improves predictive accuracy while reducing the total parameter count by approximately 33.3% and improving inference throughput. Our results suggest that HM-ASIGN provides a computationally efficient approach for 3D spatial molecular mapping. Full article

Internal combustion engines fueled by fossil fuels are major contributors to greenhouse gas (GHG) emissions. This study analyzes and predicts GHG emissions from hydrogen-enriched compressed natural gas (HCNG)-fueled spark-ignition (SI) engines. Experiments were conducted under stoichiometric conditions, and emissions before and after the three-way catalytic converter (TWC) were analyzed by varying hydrogen fraction (0–50%), EGR ratio (0–23%), engine speed (900 rpm–1500 rpm), engine load (25–75%), and spark timing (8–49 °CA bTDC). Before the TWC, increasing the hydrogen fraction from HCNG0% to HCNG40% at 1500 rpm, 50% load, and 23% EGR reduced total GHG emissions from 184.3 to 65.17 g/kWh. Similarly, for HCNG20% at 900 rpm and 30% load, the TWC reduced the CO₂-equivalent emissions of CO, CH₄, and NOx from 18.531, 8.149, and 9.057 g${\mathrm{C}\mathrm{O}}_{2eq}$/kWh to 7.013, 1.626, and 0.429 g${\mathrm{C}\mathrm{O}}_{2eq}$/kWh, respectively. Pearson correlation analysis revealed strong linear relationships between operating parameters and GHG emissions. Furthermore, emissions were predicted using four Gaussian process regression (GPR) models: Squared, Exponential, Matern 5/2, and Rational. Among these, the Exponential GPR demonstrated the highest predictive accuracy, achieving RMSE values of 0.098, 0.022, and 0.035, with corresponding R² values of 0.999, 0.807, and 0.996 for CO, CH₄, and NOx, respectively. The findings of this study offer valuable insights into GHG emissions and support the development of cleaner, more efficient HCNG engines. Full article

Accurate measurement of the 2D soot spatial distribution is vital for optimizing combustion efficiency and reducing pollutant emissions. While 1D laser extinction (LE) is robust and cost-effective, it provides only line-of-sight integrated information, lacking the spatial resolution required to resolve complex soot topologies. We propose Soot-MixFormer, a hybrid deep learning model designed for the high-fidelity inversion of 2D soot distributions from 1D extinction data. The architecture integrates CNN-based local feature extraction with Transformer-based global dependency modeling. Key innovations include a dynamic decoupled generation head and a Dual-Axial Gated Refinement (DAGR) module coupled with a physical hard constraint layer to ensure mass conservation and physical consistency. Experimental results demonstrate that Soot-MixFormer significantly outperforms baseline MLP and CNN models, achieving a Structural Similarity Index (SSIM) of 0.800 and a Pearson Correlation Coefficient (PCC) of 0.915, and a highly suppressed Root Mean Square Error (RMSE) representing less than 10% relative error in high-concentration zones. Furthermore, the model exhibits exceptional robustness, maintaining a cosine similarity above 0.72 even under 10% simulated measurement noise. The model is highly efficient, with only 0.97 M parameters and a real-time inference speed of ~246 FPS. This study provides a novel, low-cost diagnostic paradigm for real-time, high-accuracy monitoring of soot fields in industrial combustion environments, effectively bridging the gap between simple 1D sensing and complex 2D spatial reconstruction. Full article

Conventional contact-based monitoring of heart rate (HR) presents challenges such as patient discomfort, skin irritation, and poor long-term adherence, motivating the development of contactless, video-based sensing systems. This study proposes a robust hybrid framework combining advanced signal processing with machine learning to enhance HR estimation accuracy from facial video. The methodology integrates a two-stage geometric stabilization pipeline with dense facial tessellation to mitigate motion. Eulerian Video Magnification (EVM) amplifies subtle color variations, followed by chrominance-based Region of Interest (ROI) filtering. Signal recovery utilizes a sliding-window Principal Component Analysis (PCA) for local coherence, followed by Second-Order Blind Identification (SOBI), with a Light Gradient Boosting Machine (LightGBM) classifier employed to automatically identify physiological sources. Validated on the challenging COHFACE dataset, the approach achieves a Mean Absolute Error (MAE) of 1.50 bpm, a Root Mean Square Error (RMSE) of 3.07 bpm, and a Pearson Correlation Coefficient (PCC) of 0.97 on the test set. The method demonstrates robustness across diverse lighting conditions, outperforming traditional algorithms and achieving parity with state-of-the-art deep learning models, while offering an interpretable solution for contactless health monitoring. Full article

Growing concerns about global climate change and its negative consequences for communities have put immense pressure on the building industry, which is one of the primary sources of greenhouse gas emissions. Due to the environmental issues associated with the manufacture of sustainable construction materials, alkali-activated concrete (AAC) has emerged as a competitive alternative to cement. To predict the compressive strength (CS) of AAC, four machine learning (ML) models, namely, Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Random Forest (RF), and Extreme Gradient Boosting (XGBoost), were employed in this study using 193 data points. The input variables include Precursor “P” (kg/m³), Blast Furnace Slag “BFS ratio”, Sodium hydroxide “Na” (kg/m³), silicate modulus “Ms”, water content “W” (kg/m³), fine aggregate “FA” (kg/m³), coarse aggregate “A” (kg/m³), and curing time “CT” (day), with CS (MPa) as the output variable. The dataset was checked for stationarity and then normalized to decrease data redundancy and increase integrity. Furthermore, three model combinations were developed based on the relationship between the input and target variables. The XGB-M3 model outperformed all other models with a high degree of accuracy, according to the study’s findings. Specifically, the Pearson correlation coefficient (PCC) was 0.9577, and the mean absolute percentage error (MAPE) was 14.95% during the calibration phase. SHAP, an explainable AI approach that provides interpretable insights into complex AI systems by assigning feature importance to model predictions, was employed. Results suggest the higher predictions from the XGB-M3 and RF-M3 models were largely driven by curing time (CT). Full article

Spaceborne microwave imagers are vital for monitoring global precipitation due to their wide swath and high sensitivity. This study proposes a deep learning approach that integrates a U-Net with a multi-task learning (MTL) framework. The model was separately trained over land and ocean using GPM Microwave Imager (GMI) brightness temperatures, with collocated precipitation rates and types from the Dual-frequency Precipitation Radar (DPR) as labels. This combines the accuracy of radars with the coverage of imagers to produce high-precision, wide-swath precipitation estimates. In the MTL setup, near-surface precipitation rate retrieval is the main task, and precipitation type classification is the auxiliary task. A composite loss (weighted MSE and quantile regression) is used for the main task, and weighted cross-entropy for the auxiliary task. Residual blocks and an attention mechanism are incorporated to improve physical representation and generalization, thereby significantly enhancing the model’s capability to retrieve heavy precipitation. The model was trained on 2015–2024 GPM data and evaluated on an independent six-month 2025 GMI dataset. Compared to a standard U-Net, the MTL model achieved significant gains: Pearson Correlation Coefficient (PCC) increased by 9.7% (ocean) and 13.7% (land), and Critical Success Index (CSI) by 10.7% (ocean) and 10.8% (land). The method was also applied to the FY-3G Microwave Radiation Imager (MWRI-RM). In case studies, it outperformed the official product, achieving average increases of 20.1% in PCC and 15.7% in CSI, respectively. Validation against FY-3G Precipitation Measurement Radar (June–August 2024) yielded over ocean PCC = 0.757, RMSE = 1.588 mm h⁻¹, MAE = 0.355 mm h⁻¹; over land PCC = 0.691, RMSE = 2.007 mm h⁻¹, MAE = 0.692 mm h⁻¹. The study demonstrates that the MTL-enhanced U-Net significantly improves the accuracy of spaceborne microwave imager rainfall retrieval and shows robust practical applicability. Full article

Reliable sensor fault detection is critical for the safe and efficient operation of complex industrial systems, such as thermal power plants. However, traditional data-driven methods and standard deep learning models often struggle to detect incipient gradual drift faults under severe environmental noise, primarily because they ignore the inherent physical correlations among multivariate sensor signals. To address this challenge, this paper proposes a novel Physics-Coupled Deep Long Short-Term Memory Autoencoder (PC-Deep-LSTM-AE). Specifically, we integrate a deep LSTM architecture with an explicit non-linear information compression bottleneck and layer normalization to enhance robust feature extraction in high-noise environments. Furthermore, we innovatively introduce a Physics-Coupling Loss (PCC Loss) that jointly optimizes the mean squared reconstruction error and the Pearson correlation coefficient, forcing the model to strictly preserve the dynamic physical relationships among multivariable signals. Extensive experiments were conducted on a real-world thermal power plant dataset with severe noise injection. The results demonstrate that the proposed PC-Deep-LSTM-AE achieves an outstanding F1-score of over 0.98, significantly outperforming mainstream baseline models, including Vanilla LSTM-AE, GRU-AE, Bi-LSTM-AE, and CNN-AE. The proposed method exhibits exceptional robustness and high interpretability for root-cause analysis, highlighting its immense potential for real-world industrial deployment. Full article

Background/Objectives: Contrast-enhanced CT (CECT) is essential for mediastinal and lymph node assessment but is often limited in patients with renal dysfunction, prior severe contrast reactions, or pediatric populations. Deep learning approaches, such as generative adversarial networks (GANs), allow the generation of synthetic CECT (sCECT) from non-contrast CT (NCCT) without contrast injection. Materials and Methods: A GAN-based model was trained using 400 CECT scans acquired between March and July 2024. The model was tested in 20 patients with lymphoma or metastatic lymphadenopathy diagnosed between January and July 2025, using only NCCT scans. Quantitative evaluation compared sCECT with CECT using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), and Pearson Correlation Coefficient (PCC). Two radiologists performed qualitative assessment, and Signal-to-Noise Ratio (SNR)/Contrast-to-Noise Ratio (CNR) values were measured for thoracic structures. Results: Compared with NCCT, sCECT demonstrated slightly lower MAE (20.87 ± 8.84 vs. 21.26 ± 9.26) and RMSE (45.22 ± 14.22 vs. 45.94 ± 15.07), and marginally higher PSNR (15.44 ± 2.70 vs. 15.38 ± 3.02), indicating modest improvements in pixel-wise similarity. SSIM values were comparable (0.610 ± 0.09 vs. 0.63 ± 0.10), while PCC decreased (0.61 ± 0.09 vs. 0.77 ± 0.15). All differences were statistically significant (p < 0.001). Despite these mixed quantitative results, sCECT was qualitatively rated significantly higher by radiologists, with improved visualization of mediastinal structures. SNR and CNR analyses further supported enhanced contrast depiction in sCECT compared with NCCT. Conclusions: The GAN-based model successfully generated sCECT from NCCT with modest quantitative similarity gains but clear qualitative improvement, particularly for mediastinal lymph node evaluation. Although synthetic enhancement represents a learned intensity transformation rather than true iodine-based attenuation, sCECT may serve as a valuable adjunct in patients with contraindications to iodinated contrast. Full article

In this paper, high-strength high-nitrogen steel Cr₁₈Mn₁₅ was fabricated using centrifugal casting. High-temperature tensile tests were subsequently performed on the centrifugally cast material. Based on the dynamic material model (DMM), power dissipation and instability maps were constructed for the steel. The results revealed that the optimal processing conditions for Cr₁₈Mn₁₅ are within a temperature range of 940 °C to 980 °C and a strain rate range of 0.001 s⁻¹ to 0.01 s⁻¹. Flow instability was observed primarily under high strain rate conditions (1 s⁻¹) at a lower temperature of 900 °C. Four constitutive equation models were established based on the experimental results, and the prediction accuracy was assessed by calculating their average absolute relative errors (AAREs) and correlation coefficients (r). It was found that the Modified-JC constitutive model could simultaneously take care of both accuracy and simulation convergence with an AARE of 17.823 and Pearson’s correlation coefficient (PCC) of 0.968. For the practical application of Cr₁₈Mn₁₅ high-nitrogen steel, a three-layer composite tube forming and a continuous rolling equipment were developed. The rolling and spreading process was simulated using finite elements, and the stress field, strain field, and temperature field in the spreading process were analyzed to determine the following optimum process parameters of the alloy: a temperature of 950 °C, a processing line speed of 1 m/s, and a preheating temperature of 200 °C. Full article