Search Results (3,132)

To address the challenge of microwave radiometer modeling in regions lacking radiosonde data, this study proposes a neural network retrieval method based on high-resolution the Final Reanalysis (FNL) reanalysis data and validates its feasibility. A microwave radiometer brightness temperature–profiles retrieval model was developed by the Back Propagation (BP) neural network, based on FNL reanalysis data from Qingdao, China. The model’s accuracy was evaluated by comparing retrieval results with synchronous radiosonde data, with an analysis of seasonal variations. Results indicate that the Root Mean Square Error (RMSE) of temperature profiles are 1.15 °C in the near-surface layer (0–2 km) and 2.05 °C in the mid-to-upper layers (>2 km). The comprehensive RMSE for relative humidity, water vapor density, and Integrated Water Vaper (IWV) are 17.27%, 0.96 g/m³, and 1.37 mm, respectively. Overall, the errors are relatively small, and the retrieval results exhibit strong spatiotemporal consistency with radiosonde data. The error increases most rapidly within the lower atmosphere (<2 km), with distinct seasonal differences observed. Temperature and relative humidity retrieval accuracies peak in summer, whereas water vapor density and IWV retrievals perform best in winter and worst in summer. This study confirms that reanalysis data–based modeling effectively addresses the issue of limited radiosonde coverage. This method is applicable to atmospheric remote sensing in regions lacking radiosonde data, such as oceans and plateaus. It provides a feasible solution to the regional limitations of microwave radiometer applications and expands the potential uses of reanalysis data. Full article

The emergence of large language models (LLMs) has redefined the potential of artificial intelligence in clinical domains. In this context, retrieval-augmented generation (RAG) systems provide a promising approach to enhance traceability, timeliness, and accuracy in tasks such as biomedical question answering (QA). This article presents a narrative and thematic review of the evolution of these technologies in maternal health, structured across five axes: technical foundations of RAG, advancements in biomedical LLMs, conversational agents in healthcare, clinical validation frameworks, and specific applications in obstetric telehealth. Through a systematic search in scientific databases covering the period from 2022 to 2025, 148 relevant studies were identified. Notable developments include architectures such as BiomedRAG and MedGraphRAG, which integrate semantic retrieval with controlled generation, achieving up to 18% improvement in accuracy compared to pure generative models. The review also highlights domain-specific models like PMC-LLaMA and Med-PaLM 2, while addressing persistent challenges in bias mitigation, hallucination reduction, and clinical validation. In the maternal care context, the review outlines applications in prenatal monitoring, the automatic generation of clinically validated QA pairs, and low-resource deployment using techniques such as QLoRA. The article concludes with a proposed research agenda emphasizing federated evaluation, participatory co-design with patients and healthcare professionals, and the ethical design of adaptable systems for diverse clinical settings. Full article

With the development of remote sensing technologies, the volume of remote sensing data is growing dramatically, making efficient management and retrieval of large-scale remote sensing images increasingly important. Recently, deep hashing for content-based remote sensing image retrieval (CBRSIR) has attracted significant attention due to its computational efficiency and high retrieval accuracy. Although great advancements have been achieved, the imbalance between easy and difficult image pairs during training often limits the model’s ability to capture complex similarities and degrades retrieval performance. Additionally, distinguishing images with the same Hamming distance but different categories remains a challenge during the retrieval phase. In this paper, we propose a novel deep relevance hashing (DRH) for remote sensing image retrieval, which consists of a global hash learning model (GHLM) and a local hash re-ranking model (LHRM). The goal of GHLM is to extract global features from RS images and generate compact hash codes for initial ranking. To achieve this, GHLM employs a deep convolutional neural network to extract discriminative representations. A weighted pairwise similarity loss is introduced to emphasize difficult image pairs and reduce the impact of easy ones during training. The LHRM predicts relevance scores for images that share the same Hamming distance with the query to reduce confusion in the retrieval stage. Specifically, we represent the retrieval list as a relevance matrix and employ a lightweight CNN model to learn the relevance scores of image pairs and refine the list. Experimental results on three benchmark datasets demonstrate that the proposed DRH method outperforms other deep hashing approaches, confirming its effectiveness in CBRSIR. Full article

High dimensional, small sample ribonucleic acid sequencing (RNA-seq) data pose a major challenge for reliable classification due to the curse of dimensionality and the risk of overfitting. This study addresses that challenge for prostate cancer by coupling machine learning (ML) based feature selection with a hybrid ensemble classifier. RNA-seq datasets retrieved from the University of California Santa Cruz (UCSC) Xena platform have been pre-processed, and the number of features has been reduced to 30 through systematic feature selection. Three complementary learners have been combined into a majority voting ensemble to improve robustness and generalization. Performance has been assessed using evaluation metric criteria and the area under the curve. The results have been corroborated on liver, lung, and thyroid cancer datasets from the literature. The proposed hybrid ensemble method has been achieved 97.82% accuracy with Lasso feature selection. Compared with conventional single model approaches, the proposed hybrid model has yielded accurate and reliable predictions. Additionally, biologically meaningful information has been presented using explainable artificial intelligence techniques regarding the importance of genes. These findings suggest that the joint use of feature selection and hybrid ensembles provides a practical and interpretable framework for classifying high dimensional genomic profiles under limited sample sizes. Full article

Coastal lagoons are fragile and dynamic ecosystems that are particularly vulnerable to climate change and anthropogenic pressures such as urbanization and eutrophication. These vulnerabilities highlight the need for frequent and spatially extensive monitoring of water quality (WQ). While satellite remote sensing offers a valuable tool to support this effort, the optical complexity and shallow depths of lagoons pose major challenges for retrieving water column biogeochemical parameters such as chlorophyll-a ([chl-a]) and suspended particulate matter ([SPM]) concentrations. In this study, we develop and evaluate a robust satellite-based processing chain using Sentinel-2 MSI imagery over two French Mediterranean lagoon systems (Berre and Thau), supported by extensive in situ radiometric and biogeochemical datasets. Our approach includes the following: (i) a comparative assessment of six atmospheric correction (AC) processors, (ii) the development of an Optically Shallow Water Probability Algorithm (OSWPA), a new semi-empirical algorithm to estimate the probability of bottom contamination (BC), and (iii) the evaluation of several [chl-a] and [SPM] inversion algorithms. Results show that the Sen2Cor AC processor combined with a near-infrared similarity correction (NIR-SC) yields relative errors below 30% across all bands for retrieving remote-sensing reflectance R_rs(λ). OSWPA provides a spatially continuous and physically consistent alternative to binary BC masks. A new [chl-a] algorithm based on a near-infrared/blue R_rs ratio improves the retrieval accuracy while the 705 nm band appears to be the most suitable for retrieving [SPM] in optically shallow lagoons. This processing chain enables high-resolution WQ monitoring of two coastal lagoon systems and supports future large-scale assessments of ecological trends under increasing climate and anthropogenic stress. Full article

Under the context of “Energy Conservation and Emission Reduction,” low-sulfur fuel has become widely adopted in maritime operations, posing significant challenges to cylinder lubrication systems. Traditional oil injection strategies, heavily reliant on manual experience, suffer from instability and high costs. To address this, a lightweight image retrieval model for cylinder lubrication is proposed, leveraging deep learning and computer vision to support oiling decisions based on visual features. The model comprises three components: a backbone network, a feature enhancement module, and a similarity retrieval module. Specifically, EfficientNetB0 serves as the backbone for efficient feature extraction under low computational overhead. MobileViT Blocks are integrated to combine local feature perception of Convolutional Neural Networks (CNNs) with the global modeling capacity of Transformers. To further improve receptive field and multi-scale representation, Receptive Field Blocks (RFB) are introduced between the components. Additionally, the Convolutional Block Attention Module (CBAM) attention mechanism enhances focus on salient regions, improving feature discrimination. A high-quality image dataset was constructed using WINNING’s large bulk carriers under various sea conditions. The experimental results demonstrate that the EfficientNetB0 + RFB + MobileViT + CBAM model achieves excellent performance with minimal computational cost: 99.71% Precision, 99.69% Recall, and 99.70% F1-score—improvements of 11.81%, 15.36%, and 13.62%, respectively, over the baseline EfficientNetB0. With only a 0.3 GFLOP and 8.3 MB increase in model size, the approach balances accuracy and inference efficiency. The model also demonstrates good robustness and application stability in real-world ship testing, with potential for further adoption in the field of intelligent ship maintenance. Full article

This paper presents an optimal control for autonomous vehicles navigating in intersection scenarios. The proposed controller is based on solving a Quadratic Programming optimization technique to provide a feasible control signal respecting actuator constraints. The proposed controller was implemented in a scale-sized vehicle and is executed using only on-board perception and computing systems to retrieve the state dynamics, i.e., an inertial measurement unit and a monocular camera, to compute the estimated states through intelligent computer vision algorithms. The stability of the error signals of the closed-loop system was proved both mathematically and experimentally, using standard performance indices for ten trials. The proposed technique was compared against LQR and MPC strategies, showing 67% greater accuracy than the LQR approach and 53.9% greater accuracy than the MPC technique, while turning during the intersection. Moreover, the proposed QP controller showed significantly greater efficiency by reducing the control effort by 63.3% compared to the LQR, and by a substantial 78.4% compared to the MPC. These successful results proved that the proposed controller is an effective alternative for autonomously navigating within intersection scenarios. Full article

Accurate estimation of the forest canopy height is crucial in monitoring the global carbon cycle and evaluating progress toward carbon neutrality goals. The Global Ecosystem Dynamics Investigation (GEDI) mission provides an important data source for canopy height estimation at a global scale. However, the non-zero half-width of the transmitted laser pulses (NHWTLP) and the influence of terrain slope can cause waveform broadening and overlap between canopy returns and ground returns in GEDI waveforms, thereby reducing the estimation accuracy. To address these limitations, we propose a canopy height retrieval method that combines the deconvolution of GEDI’s transmitted waveforms with terrain slope constraints on the ground response function. The method consists of two main components. The first is performing deconvolution on GEDI’s effective return waveforms using their corresponding transmitted waveforms to obtain the true ground response function within each GEDI footprint, thereby mitigating waveform broadening and overlap induced by NHWTLP. This process includes constructing a convolution convergence function for GEDI waveforms, denoising GEDI waveform data, transforming one-dimensional ground response functions into two dimensions, and applying amplitude difference regularization between the convolved and observed waveforms. The second is incorporating terrain slope parameters derived from a digital terrain model (DTM) as constraints in the canopy height estimation model to alleviate waveform broadening and overlap in ground response functions caused by topographic effects. The proposed approach enhances the precision of forest canopy height estimation from GEDI data, particularly in areas with complex terrain. The results demonstrate that, under various conditions—including GEDI full-power beams and coverage beams, different terrain slopes, varying canopy closures, and multiple study areas—the retrieved height (rh) model constructed from ground response functions derived via the inverse deconvolution of the transmitted waveforms (IDTW) outperforms the RH (the official height from GEDI L2A) model constructed using RH parameters from GEDI L2A data files in forest canopy height estimation. Specifically, without incorporating terrain slope, the rh model for canopy height estimation using full-power beams achieved a coefficient of determination (R²) of 0.58 and a root mean square error (RMSE) of 5.23 m, compared to the RH model, which had an R² of 0.58 and an RMSE of 5.54 m. After incorporating terrain slope, the rh_g model for full-power beams in canopy height estimation yielded an R² of 0.61 and an RMSE of 5.21 m, while the RH_g model attained an R² of 0.60 and an RMSE of 5.45 m. These findings indicate that the proposed method effectively mitigates waveform broadening and overlap in GEDI waveforms, thereby enhancing the precision of forest canopy height estimation, particularly in areas with complex terrain. This approach provides robust technical support for global-scale forest resource assessment and contributes to the accurate monitoring of carbon dynamics. Full article

Spatializing crop yield statistical data with administrative divisions as the basic unit helps reveal the spatial distribution characteristics of crop yield and provides necessary spatial information to support field management and government decision-making. However, owing to an insufficient understanding of the factors affecting yield, accurately depicting its spatial differences remains challenging. Taking Hailun city, Heilongjiang Province, as an example, this study proposes a yield downscaling method based on the standardized deviation from the mean of the comprehensive crop condition index (CCCI) during key phenological periods of the growing season. First, Sentinel-2 remote sensing data were used to retrieve crop condition parameters during key phenological periods, and the CCCI was constructed using the correlation between crop condition parameters in key phenological periods and statistical yield as the weight. Subsequently, regression analysis and the entropy weight method were applied to determine the spatiotemporal dynamic weights of the CCCI during key phenological stages and to calculate the standardized deviation from the mean. By combining these two components, the comprehensive spatial difference index of the crop growth condition (CSDICGC) was derived, which offered a new way to characterize the discrepancies between the pixel-level yield and statistical yield, thereby downscaling the yield statistical data from the administrative unit to the pixel scale. The results indicated that this method achieved a regional accuracy close to 100%, with a strong fit at the pixel scale. Pixel-level accuracy validation against ground-truth maize yield data resulted in an R² of 0.82 and a mean relative error (MRE) of 4.75%. The novelty of this study was characterized by the integration of multistage crop condition parameters with dynamic spatiotemporal weighting to overcome the limitations of single-index methods. The crop yield statistical data downscaling spatialization method proposed in this paper is simple and efficient and has the potential to be popularized and applied over relatively large regions. Full article

As a staple cereal worldwide, winter wheat plays a pivotal role in food security. Leaf chlorophyll serves as a direct indicator of photosynthetic performance and nitrogen nutrition, making it critical for precision management and yield gains. Consequently, rapid, nondestructive, and high-accuracy remote-sensing retrievals are urgently needed to underpin field operations and precision fertilization. In this study, canopy hyperspectral reflectance together with destructive chlorophyll assays were systematically acquired from Yangling field trials conducted during 2018–2020. Three families of spectral indices were devised: classical empirical indices; two-dimensional optimal spectral indices (2D OSI) selected by correlation-matrix screening; and novel three-dimensional optimal spectral indices (3D OSI). The main contribution lies in devising novel 3D OSIs that combine three spectral bands and demonstrating how their fusion with classic two-band indices can improve chlorophyll quantification. Correlation analysis showed that most empirical vegetation indices were significantly associated with chlorophyll (p < 0.05), with the new double difference index (NDDI) giving the strongest relationship (R = 0.637). Within the optimal-index sets, the difference three-dimensional spectral index (DTSI; 680, 807, and 1822 nm) achieved a correlation coefficient of 0.703 (p < 0.05). Among all multi-input fusion schemes, fusing empirical indices with 3D OSI and training with RF delivered the best validation performance (R² = 0.816, RMSE = 0.307 mg g⁻¹, MRE = 11.472%), and external data further corroborated its feasibility. Altogether, integrating 3D spectral indices with classical vegetation indices and deploying RF enabled accurate, nondestructive estimation of winter wheat chlorophyll, offering a new hyperspectral pathway for monitoring crop physiological status and advancing precision agricultural management and fertilization, can guide in-season fertilization to optimize nitrogen use, thereby advancing precision agriculture. Full article

Electronic health records (EHRs) are typically stored in relational databases, making them difficult to query for nontechnical users, especially under privacy constraints. We evaluate two practical clinical NLP workflows, natural language to SQL (NL2SQL) for EHR querying and retrieval-augmented generation for clinical question answering (RAG-QA), with a focus on privacy-preserving deployment. We benchmark nine large language models, spanning open-weight options (DeepSeek V3/V3.1, Llama-3.3-70B, Qwen2.5-32B, Mixtral-8 × 22B, BioMistral-7B, and GPT-OSS-20B) and proprietary APIs (GPT-4o and GPT-5). The models were chosen to represent a diverse cross-section spanning sparse MoE, dense general-purpose, domain-adapted, and proprietary LLMs. On MIMICSQL (27,000 generations; nine models × three runs), the best NL2SQL execution accuracy (EX) is 66.1% (GPT-4o), followed by 64.6% (GPT-5). Among open-weight models, DeepSeek V3.1 reaches 59.8% EX, while DeepSeek V3 reaches 58.8%, with Llama-3.3-70B at 54.5% and BioMistral-7B achieving only 11.8%, underscoring a persistent gap relative to general-domain benchmarks. We introduce SQL-EC, a deterministic SQL error-classification framework with adjudication, revealing string mismatches as the dominant failure (86.3%), followed by query-join misinterpretations (49.7%), while incorrect aggregation-function usage accounts for only 6.7%. This highlights lexical/ontology grounding as the key bottleneck for NL2SQL in the biomedical domain. For RAG-QA, evaluated on 100 synthetic patient records across 20 questions (54,000 reference–generation pairs; three runs), BLEU and ROUGE-L fluctuate more strongly across models, whereas BERTScore remains high on most, with DeepSeek V3.1 and GPT-4o among the top performers; pairwise t-tests confirm that significant differences were observed among the LLMs. Cost–performance analysis based on measured token usage shows per-query costs ranging from USD 0.000285 (GPT-OSS-20B) to USD 0.005918 (GPT-4o); DeepSeek V3.1 offers the best open-weight cost–accuracy trade-off, and GPT-5 provides a balanced API alternative. Overall, the privacy-conscious RAG-QA attains strong semantic fidelity, whereas the clinical NL2SQL remains brittle under lexical variation. SQL-EC pinpoints actionable failure modes, motivating ontology-aware normalization and schema-linked prompting for robust clinical querying. Full article

Structured-light 3D reconstruction is an active measurement technique that extracts spatial geometric information of objects by projecting fringe patterns and analyzing their distortions. It has been widely applied in industrial inspection, cultural heritage digitization, virtual reality, and other related fields. This review presents a comprehensive analysis of mainstream fringe-based reconstruction methods, including Fringe Projection Profilometry (FPP) for diffuse surfaces and Phase Measuring Deflectometry (PMD) for specular surfaces. While existing reviews typically focus on individual techniques or specific applications, they often lack a systematic comparison between these two major approaches. In particular, the influence of different projection schemes such as Digital Light Processing (DLP) and MEMS scanning mirror–based laser scanning on system performance has not yet been fully clarified. To fill this gap, the review analyzes and compares FPP and PMD with respect to measurement principles, system implementation, calibration and modeling strategies, error control mechanisms, and integration with deep learning methods. Special focus is placed on the potential of MEMS projection technology in achieving lightweight and high-dynamic-range measurement scenarios, as well as the emerging role of deep learning in enhancing phase retrieval and 3D reconstruction accuracy. This review concludes by identifying key technical challenges and offering insights into future research directions in system modeling, intelligent reconstruction, and comprehensive performance evaluation. Full article

This study presents a dual-layer GraphRAG framework for petroleum engineering question answering, in which instance-level facts and domain-level concepts are explicitly separated and integrated into retrieval-augmented generation. To evaluate the framework, a benchmark of 40 expert-constructed Q&A pairs was developed, covering factual, definitional, and explanatory queries derived from a real offshore oilfield dataset. Results show that the dual-layer graph consistently outperforms a single-layer baseline. Answer accuracy improves from 0.65 to 0.70, faithfulness from 0.54 to 0.61, and context relevance from 0.69 to 0.72, confirming that the system retrieves factual parameters more reliably and provides conceptually grounded explanations. Gains in evidence recall and coverage are more modest, highlighting areas for further optimization. A case study illustrates the framework’s ability to expand petroleum terminology (e.g., “sandstone → clastic rock”), producing responses that are not only quantitatively more reliable but also qualitatively more informative. The dual-layer design effectively addresses the semantic consistency gap in petroleum QA, offering practical value for reservoir evaluation, lithology interpretation, and technical decision support. These findings demonstrate the potential of GraphRAG to enhance knowledge management and intelligent services in petroleum engineering. Full article

Large-scale medical visual question answering (MedVQA) datasets are critical for training and deploying vision–language models (VLMs) in radiology. Ideally, such datasets should be automatically constructed from routine radiology reports and their corresponding images. However, no existing method directly links free-text findings to the most relevant 2D slices in volumetric computed tomography (CT) scans. To address this gap, a contrastive language–image pre-training (CLIP)-based key slice selection framework is proposed, which matches each sentence to its most informative CT slice via text–image similarity. This experiment demonstrates that models pre-trained in the medical domain already achieve competitive slice retrieval accuracy and that fine-tuning them on a small dual-supervised dataset that imparts both lesion- and organ-level awareness yields further gains. In particular, the best-performing model (fine-tuned BiomedCLIP) achieved a Top-1 accuracy of 51.7% for lesion-aware slice retrieval, representing a 20-point improvement over baseline CLIP, and was accepted by radiologists in 56.3% of cases. By automating the report-to-slice alignment, the proposed method facilitates scalable, clinically realistic construction of MedVQA resources. Full article

Significant uncertainties persist across different Leaf Area Index (LAI) products due to multiple factors; therefore, the accuracy assessment of the global LAI products is an indispensable step before their application. In this study, comprehensive validation of multi-scale LAI products including Sentinel-2, Landsat-8/9, and MCD15A3H was implemented utilizing fine-resolution LAI maps which were based on UAV images and field-measured LAI data. The validation results demonstrated a consistent, systematic underestimation across all the LAI products within the study area, the RMSE of these products ranged from 0.56 to 1.63, and the coarse-resolution MCD15A3H LAI product demonstrated highest accuracy (RMSE = 0.56, R² = 0.69). The Sentinel-2 products exhibited intermediate accuracy among all those products (RMSE: 1.16–1.36). The Landsat-8/9 LAI product showed markedly lower accuracy relative to Sentinel-2; its RMSE (1.63) exceeded that of Sentinel-2 10 m LAI and 20 m LAI by 40.52% and 21.64%, respectively. In addition, all these LAI products showed consistent seasonal variation patterns with the reference LAI maps. Moreover, Sentinel-2 10 m LAI products showed serious underestimation for all vegetation types, with forests providing the highest RMSE = 0.89. This study serves as a valuable reference for the application of multi-scale LAI products in heterogeneous terrain and provides directions for the improvement of fine-resolution LAI retrieval algorithms. Full article