This paper systematically expounds the core challenges and development vision facing transfusion medicine and discipline construction in China under the strategic goals of achieving "Chinese modernization" and "Healthy China". This article puts forward for the first time the "blood transfusion view" of Chinese blood transfusion practitioners and elaborates on key challenges in the field, including the disciplinary system, talent pool, service balance, technical bottlenecks and standardization in emerging fields. To address the challenges, it is urgently necessary to establish the disciplinary status of transfusion medicine, improve talent cultivation, enhance the integration of scientific research and technology, optimize transfusion assessment, promote autologous blood transfusion, advance the localization of evidence-based medicine, and expand the management functions of biological therapy. The ultimate goal is to build a modern blood transfusion system led by science and technology and with Chinese characteristics to safeguard people's health.

[Objective] To investigate the current knowledge status of transfusion medicine among pediatricians/postgraduates and the reliability of large language models (LLMs) for assisted learning, and to assess changes in pediatricians' transfusion medicine knowledge before and after the implementation of the "Pediatric Transfusion Guideline" (hereafter referred to as the "Guideline"). [Methods] In January 2022 (prior to the implementation of the "Guideline"), a questionnaire was developed based on the "Guideline" content and distributed to pediatricians. Subsequently, in July 2025 (after the implementation of the "Guideline"), the "Pediatric Transfusion Medicine Knowledge Questionnaire" was designed based on the content of the January 2022 questionnaire. This questionnaire survey was conducted on pediatricians/postgraduates and LLMs. We analyzed the level of transfusion medicine knowledge among pediatricians/postgraduates and the reliability of LLMs for assisted learning, and compared the accuracy of pediatricians' responses before and after "Guideline" implementation. [Results] The survey results after the implementation of the "Guidelines" revealed that pediatricians/postgraduates achieved response accuracy rates exceeding 80% on the topic of "Patient Blood Management". However, response accuracy rates were below 30% for topics including "Types and Indications of Blood Components/Products" and "E-valuation of Transfusion Efficacy". The pediatricians' accuracy rates for related questions before and after the implementation of the "Guidelines" were 14.7%-68.9% and 3%-38%, respectively, and the comparison of accuracy rates for each question showed significant differences (P<0.001). The accuracy rates of the LLMs on the questionnaire were all below 90%. Among them, Doubao (81.1%) and Kimi (86.4%) achieved relatively higher accuracy rates, while Tencent Yuanbao (Hunyuan) had the lowest accuracy rate at only 59.5%. [Conclusion] The implementation of the "Guideline" may have improved pediatricians' knowledge level of pediatric transfusion medicine. However, their knowledge level of pediatric transfusion remains low, and LLMs cannot yet provide absolutely reliable guidance. Systematic training in pediatric transfusion medicine is urgently needed.

Polarimetric synthetic aperture radar (PolSAR) image classification is a fundamental remote sensing interpretation, which remains challenging because polarimetric responses vary strongly with target orientation and radar incidence angles. These variations introduce complex backscattering properties that are difficult to model. To address this issue, we present the rotation-aware multigranularity polarimetric spatial network (RAMPSNet), which adopts a hierarchical and modular design to extract rotation-consistent features from both the PolSAR data rotation domain and multigranularity polarimetric spatial perspective. RAMPSNet consists of three core modules. Specifically, the Rotation-Aware Module (RAM) explicitly models dependencies along the rotation dimension by combining a learnable rotation positional embedding, a rotation-aware self-attention mechanism, and a lightweight feedforward network. These components jointly capture global and local rotational relationships and allow RAM to extract stable and rotation-consistent polarimetric features. The multigranularity feature extraction block (MGFEB) splits the input along the channel dimension into three parallel branches: a global frequency-aware branch, a local multikernel convolution branch, and an identity mapping branch. This design enables MGFEB to extract complementary structural, textural, and fine-grained polarimetric spatial features for richer semantic information learning with low computational overhead. Finally, the Channel-Spatial Interaction Module sequentially models channel and spatial dependencies to enhance discriminative features, suppress redundancy, and ensure structural coherence and produces discriminative final representations. Experiments on three widely used PolSAR datasets show that RAMPSNet consistently achieves higher accuracy than state-of-the-art methods, as well as fewer parameters and lower computational cost.

[Objective] Through analyzing the current handling capabilities for complex cases in blood transfusion compatibility detection and the application status of artificial intelligence-assisted (AI-assisted) technologies, this study explores the establishment of an effective AI-augmented protocol for managing challenging blood transfusion compatibility detection. [Methods] This research systematically analyzes, designs, and explores an AI-augmented operational workflow for resolving challenging blood transfusion compatibility detection cases. Through three representative case studies, we evaluate its effectiveness, accuracy, and efficiency in addressing real-world diagnostic challenges. [Results] The AI-augmented operational model demonstrates significant efficacy in resolving complex blood transfusion compatibility challenges, including complex blood typing, antibody specificity identification, challenging cross-matching, and transfusion strategy formulation. [Conclusion] AI-augmented technologies demonstrate immense potential in resolving complex blood transfusion compatibility detections. By enabling intelligent, automated, precise, and standardized solutions, they significantly enhance diagnostic accuracy and operational efficiency, which is critical for ensuring transfusion safety and advancing personalized transfusion medicine. This study delineates both the advantages and existing limitations of AI implementation, explores future developmental trajectories, and provides a theoretical framework and practical implementation pathways for deeper integration of AI in transfusion medicine.

To guide clinical blood transfusion practices for pediatric patients, the National Health Commission has issued the health standard "Guideline for pediatric transfusion" (WS/T 795-2022). Blood transfusion is one of the most commonly used supportive treatments for children with hematological diseases. This guideline provides guidance and recommendations for blood transfusions in children with aplastic anemia, thalassemia, autoimmune hemolytic anemia, glucose-6-phosphate dehydrogenase deficiency, acute leukemia, myelodysplastic syndromes, immune thrombocytopenic purpura, and thrombotic thrombocytopenic purpura. This article presents the evidence and interpretation of the blood transfusion provisions for children with hematological diseases in the "Guideline for pediatric transfusion", aiming to assist in the understanding and implementing the blood transfusion section of this guideline.

Penile squamous cell carcinoma (PSCC) is a malignant reproductive tumor, and circRNAs are essential regulators in the progression of cancers. However, their specific roles in PSCC have not been adequately investigated, which are available to function via the competing endogenous RNA (ceRNA) network. We collected 5 normal and 6 PSCC tissue samples for bulk RNA sequencing and obtained differentially expressed circRNAs (DE-circRNAs) and differentially expressed mRNAs (DE-mRNAs). Then we conducted a correlation analysis between them, followed by the prediction and intersection of their target miRNAs and mRNAs by online databases. We constructed a ceRNA network of 11 circRNAs, 14 miRNAs, and 33 mRNAs. The enrichment analysis showed that the positively circRNA-related mRNAs are mainly involved in the positive regulation of leukocyte activation and cell adhesion. It was revealed that circ_CDR1as and its target gene GATA6 were significantly downregulated in PSCC, and their reduction in tumor samples was validated by an external dataset and qRT-PCR, indicating a potential regulatory relationship between them. Our study identified that circ_CDR1as may affect PSCC progression through the circRNA-miRNA-mRNA network, and GATA6 could be one of its possible targets for PSCC therapy strategies.

Based on systematic review and consensus meetings of international multidisciplinary experts, the Transfusion and Anemia Expert Initiative—Control/Avoidance of Bleeding (TAXI-CAB) project team developed management strategies for platelet and plasma transfusion in critically ill children. This consensus presents five expert consensus statements and two recommendations addressing two key questions: 1) What Laboratory Tests and Physiologic Triggers Should Guide the Decision to Administer a Platelet or Plasma Transfusion in Critically ill Children? 2) What Product Attributes Are Optimal to Guide Specific Product Selection? This consensus provides guidance for decision-making regarding plasma and platelet transfusion in critically ill children in two aspects: relevant laboratory testing indicators and additional special properties of blood components. This article explains the rationale behind the recommendations in this part of the guideline, aiming to emphasize the need for clinicians to develop transfusion strategies based on multidimensional assessment, while calling for enhanced interdisciplinary collaboration and evidence-based research to optimize blood management in critically ill children, reducing the risk of over-transfusion and improving treatment outcomes. Furthermore, there remains an urgent need for further research to explore laboratory indicators associated with bleeding risk to guide transfusion therapy.

This study aimed to identify the optimal prediction method and key preoperative variables for red blood cell (RBC) transfusion risk in patients undergoing mitral valve surgery. We conducted a retrospective study involving 1477 patients from eight large tertiary hospitals in China who underwent mitral valve surgery with cardiopulmonary bypass. From thirty collected preoperative variables, the Max-Relevance and Min-Redundancy (mRMR) method was used for feature selection, and various machine learning models were evaluated. Of the 1477 patients, 862 received RBC transfusions. The mRMR method identified ten significant preoperative variables. The LightGBM model demonstrated superior performance, achieving an area under the curve (AUC) of 0.935 in the training set and 0.734 in the validation set, with 74.2% accuracy in a prospective dataset. SHAP analysis revealed the ten most influential variables were hematocrit, RBC count, weight, body mass index, fibrinogen, hemoglobin, height, age, left ventricular dilation, and sex. In conclusion, LightGBM was identified as the optimal model for predicting RBC transfusion needs. The model’s high accuracy can assist clinicians in anticipating transfusions and improving blood management decisions.

Cropland is fundamental to food security, and monitoring cropland non-agriculturalization through satellite enforcement can effectively manage and protect cropland. However, existing research primarily focuses on optical imagery, and there are problems such as low data processing efficiency and long updating cycles, which make it difficult to meet the needs of large-scale rapid monitoring. To comprehensively and accurately obtain cropland change information, this paper proposes a method based on the Google Earth Engine (GEE) cloud platform, combining optical imagery and synthetic aperture radar (SAR) data for quick and accurate detection of cropland non-agriculturalization. The method uses existing land-use/land cover (LULC) products to quickly update cropland mapping, employs change vector analysis (CVA) for detecting non-agricultural changes in cropland, and introduces vegetation indices to remove pseudo-changes. Using Shanwei City, Guangdong Province, as a case study, the results show that (1) the cropland map generated in this study aligns well with the actual distribution of cropland, achieving an accuracy of 90.8%; (2) compared to using optical imagery alone, the combined optical and SAR data improves monitoring accuracy by 22.7%, with an overall accuracy of 73.65%; (3) in the past five years, cropland changes in Shanwei followed a pattern of an initial increase followed by a decrease. The research in this paper can provide technical reference for the rapid monitoring of cropland non-agriculturalization on a large scale, so as to promote cropland protection and rational utilization of cropland.

Limited labels and detailed changed land-cover interpretation requirements pose challenges for time-series PolSAR change monitoring research. Accurate labels and supervised models are difficult to reuse between massive unlabeled time-series PolSAR data due to the complex distribution shifts caused by different imaging parameters, scene changes, and random noises. Moreover, many related methods can only detect binary changes in PolSAR images and struggle to track the detailed land cover changes. In this study, an unsupervised cross-domain method based on limited-label transfer learning and a vision transformer (LLTL-ViT) is proposed for PolSAR land-cover change monitoring, which effectively alleviates the problem of difficult label reuse caused by domain shift in time-series SAR data, significantly improves the efficiency of label reuse, and provides a new paradigm for the transfer learning of time-series polarimetric SAR. Firstly, based on the polarimetric scattering characteristics and manifold-embedded distribution alignment transfer learning, LLTL-ViT transfers the limited labeled samples of source-domain PolSAR data to unlabeled target-domain PolSAR time-series for initial classification. Secondly, the accurate samples of target domains are further selected based on the initial transfer classification results, and the deep learning network ViT is applied to classify the time-series PolSAR images accurately. Thirdly, with the reliable secondary classification results of time-series PolSAR images, the detailed changes in land cover can be accurately tracked. Four groups of cross-domain change monitoring experiments were conducted on the Radarsat-2, Sentinel-1, and UAVSAR datasets, with about 10% labeled samples from the source-domain PolSAR. LLTL-ViT can reuse the samples between unlabeled target-domain time-series and leads to a change detection accuracy and specific land-cover change tracking accuracy of 85.22–96.36% and 72.18–88.06%, respectively.

To guide transfusion practice in critically ill children who often need plasma and platelet transfusions, the Transfusion and Anemia Expertise Initiative-Control/Avoidance of Bleeding (TAXI-CAB) developed Recommendations and Expert Consensus for Plasma and Platelet Transfusion Practice in Critically Ill Children. This guideline addresses 53 recommendations related to plasma and platelet transfusion in critically ill children with 8 kinds of diseases, laboratory testing, selection/treatment of plasma and platelet components, and research priorities. This paper introduces the specific methods and results of the recommendation formation of the guideline.

To guide clinical blood transfusion practices for pediatric patients, the National Health Commission has issued the health standard "Guideline for pediatric transfusion" (WS/T 795-2022). Blood transfusion for children undergoing hematopoietic stem cell transplantation is highly complex and challenging. This guideline provides recommendations on transfusion thresholds and the selection of blood components for these children. This article presents the evidence and interpretation of the transfusion provisions for children undergoing hematopoietic stem cell transplantation, with the aim of enhancing the understanding and implementation of the "Guideline for pediatric transfusion".

OBJECTIVES:Multiple myeloma (MM) is a hematologic malignancy characterized by clonal proliferation of plasma cells and remains incurable. Patients with primary refractory multiple myeloma (PRMM) show poor response to initial induction therapy. This study aims to develop a machine learning-based model to predict treatment response in newly diagnosed multiple myeloma (NDMM) patients, in order to optimize therapeutic strategies. METHODS:NDMM and post-treatment MM patients hospitalized in the Department of Hematology, Third Xiangya Hospital, Central South University, between August 2022 and July 2023 were enrolled. Post-treatment MM patients were categorized into PRMM patients and treatment-responsive MM (TRMM) patients based on therapeutic efficacy. Serum metabolites were detected and analyzed via metabolomics. Based on the metabolomics analysis results and combined with transcriptomic sequencing data of NDMM patients from databases, differentially expressed amino acid metabolism-related genes (AAMGs) among post-treatment NDMM patients with varying therapeutic outcomes were screened. Using bioinformatics analyses and machine learning algorithms, a predictive model for treatment response in NDMM was constructed and used to identify patients at risk for PRMM. RESULTS:A total of 61 patients were included: 22 NDMM, 23 TRMM, and 16 PRMM patients. Significant differences in metabolite levels were observed among the 3 groups, with differential metabolites mainly enriched in amino acid metabolism pathways. Follow-up data were available for 16 of the 22 NDMM patients, including 12 treatment responders (ND_TR group) and 4 with PRMM (ND_PR group). A total of 23 differential metabolites were identified between these 2 groups: 6 metabolites (e.g., tryptophan) were upregulated and 17 (e.g., citric acid) were downregulated in the ND_TR group. Transcriptomic data from 108 TRMM and 77 PRMM patients were analyzed to identify differentially expressed AAMGs, which were then used to construct a prediction model. The area under the receiver operating characteristic curve (AUC) for the model exceeded 0.8, and AUC values in 3 external validation cohorts were all above 0.7. CONCLUSIONS:This study delineated the metabolic alterations in MM patients with different treatment response, suggesting that dysregulated amino acid metabolism may be associated with poor treatment response in PRMM. By integrating metabolomics and transcriptomics, a machine learning-based predictive model was successfully established to forecast treatment response in NDMM patients.

Real-world study (RWS), based on multi-source data from real medical environments, is gradually becoming an important supplement to traditional randomized controlled trials, and its application in the field of transfusion medicine is becoming increasingly widespread. This article systematically reviews the definition and methodological system of RWS, examines its application cases in clinical blood transfusion research, and discusses the advantages, limitations, and future research directions of RWS, aiming to provide a reference for evidence-based research in blood transfusion medicine.

To guide clinical blood transfusion practices for pediatric patients, the National Health Commission has issued the health standard "Guideline for pediatric transfusion" (WS/T 795-2022). Blood transfusion is one of the most commonly used supportive treatments for children with hematological diseases. This guideline provides guidance and recommendations for blood transfusions in children with aplastic anemia, thalassemia, autoimmune hemolytic anemia, glucose-6-phosphate dehydrogenase deficiency, acute leukemia, myelodysplastic syndromes, immune thrombocytopenic purpura, and thrombotic thrombocytopenic purpura. This article presents the evidence and interpretation of the blood transfusion provisions for children with hematological diseases in the "Guideline for pediatric transfusion", aiming to assist in the understanding and implementing the blood transfusion section of this guideline.

Background: Ozone can enhance the expression of some growth factors (GFs) in platelet rich plasma (PRP), recent study showed oxygen-rich PRP (ozonized PRP) have better therapeutic effects on bone and joint diseases. PRP injection has been widely used in the treatment of facial rejuvenation, but the efficacy of sufficient oxygen-rich PRP in facial rejuvenation has not been studied. Objective: Firstly, we examined whether ozone treatment can increase the concentration of GFs of PRP in vitro. And then a variety of subjective and objective detection methods were used to evaluate the effect of sufficient(10-12 mL each time for the injection of face and neck) oxygen-rich (ozonized PRP) PRP injection in facial rejuvenation by follow-up for 6 months. At last, we investigated the satisfaction, side effects and pain score of the treatment through a questionnaire survey. Methods: The concentration of main GFs in PRP treated with different dose of ozone in vitro was measured by ELISA. Clinical picture, the collagen thickness of dermis by reflectance confocal microscope(RCM), skin conditions (including spots, ultraviolet (UV) spots, brown spots, red area, pores, wrinkles, texture and porphyrin) by VISIA were collected before treatment and each month follow-up visit after treatment until 6-month follow-up period was finished. Patients' satisfaction, side effects and pain score were collected at the end of follow-up period. Results: PRP treated by high-dose ozone (57 mg/mL, ozone/PRP volume ratio:1/1) in vitro showed a significant increase in endothelial growth factor (EGF) and transforming growth factor-b (TGF-b) compared to baseline(P < 0.05). Collagen thickness of forehead, cheek and neck improved significantly compare to the baseline until to the 6 months after treatment. Spots, UV spots, brown spots, red area and texture improved significantly compare to the baseline(P < 0.05). All of participants reported improvement and have a median pain score of 4.19. No serious adverse events were observed. Conclusions: Ozone treatment can increase the concentration of GFs such as EGF and TGF-b in PRP in vitro. Sufficient oxygen-rich PRP injection may be an effective and promising method to treat facial rejuvenation. (c) 2024, The Japanese Society for Regenerative Medicine. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/ 4.0/).

InSAR and optical techniques represent two principal approaches for the generation of large-scale Digital Elevation Models (DEMs). Due to the inherent limitations of each technology, a single data source is insufficient to produce high-quality DEM products. The increasing deployment of satellites has generated vast amounts of InSAR and optical DEM data, thereby providing opportunities to enhance the quality of final DEM products through the more effective utilization of the existing data. Previous research has established that complete DEMs generated by InSAR technology can be combined with optical DEMs to produce a fused DEM with enhanced accuracy and reduced noise. Traditional DEM fusion methods typically employ weighted averaging to compute the fusion results. Theoretically, if the weights are appropriately selected, the fusion outcome can be optimized. However, in practical scenarios, DEMs frequently lack prior information on weights, particularly precise weight data. To address this issue, this study adopts a fully connected artificial neural network for elevation fusion prediction. This approach represents an advancement over existing neural network models by integrating local elevation and terrain as input features and incorporating curvature as an additional terrain characteristic to enhance the representation of terrain features. We also investigate the impact of terrain factors and local terrain feature as training features on the fused elevation outputs. Finally, three representative study areas located in Oregon, USA, and Macao, China, were selected for empirical validation. The terrain data comprise InSAR DEM, AW3D30 DEM, and Lidar DEM. The results indicate that compared to traditional neural network methods, the proposed approach improves the Root-Mean-Squared Error (RMSE) ranges, from 5.0% to 12.3%, and the Normalized Median Absolute Deviation (NMAD) ranges, from 10.3% to 26.6%, in the test areas, thereby validating the effectiveness of the proposed method.

Landslides are the most serious geological disaster in our country, causing economic losses. Because they go undetected, a large number of landslides that have caused disasters are not in the catalogue. At present, Interferometric Synthetic Aperture Radar (InSAR) has been widely used in the identification of landslides. However, it is time-consuming, inefficient, etc., to survey landslides throughout our large country. In the context of massive SAR data, this problem is more obvious. Therefore, based on the current technique of using differential interferogram phase gradient stacking to avoid phase unwrapping errors, a landslide phase gradient dataset has been constructed. To validate the dataset’s effectiveness and applicability, deep learning methods were introduced, applying the dataset to four networks: U-Net, Attention-Unet, Bisenet v2, and Deeplab v3. The results indicate that the phase gradient dataset performs well across different models, with the Attention-Unet network demonstrating the best performance. Specifically, the precision, recall, and accuracy on the test dataset were 0.8771, 0.8712, and 0.9834, respectively, and the accuracy on the validation dataset was 0.8523. Finally, in this paper, the model is applied to landslide identification in Gansu Province, China, during 2022-2023, and a total of 1882 landslides are found. These landslides are mainly concentrated in the south of Gansu Province, where the terrain is relatively undulating. The results show that this method can quickly and accurately realize landslide automatic identification in a wide area and provide technical support for large-scale landslide disaster surveys.

Landslide susceptibility mapping (LSM) constitutes a valuable analytical instrument for estimating the likelihood of landslide occurrence, thereby furnishing a scientific foundation for the prevention of natural hazards, land-use planning, and economic development in landslide-prone areas. Existing LSM methods are predominantly data-driven, allowing for significantly enhanced monitoring accuracy. However, these methods often overlook the consideration of landslide mechanisms and uncertainties associated with non-landslide samples, resulting in lower model reliability. To effectively address this issue, a knowledge-guided landslide susceptibility assessment framework is proposed in this study to enhance the interpretability and monitoring accuracy of LSM. First, a landslide knowledge graph is constructed to model the relationships between landslide entities and summarize landslide susceptibility rules. Next, combining the obtained landslide rules with geographic similarity principles, high-confidence non-landslide samples are selected to optimize the quality of the samples. Subsequently, a Landslide Knowledge Fusion Cell (LKF-Cell) is utilized to couple landslide data with landslide knowledge, resulting in the acquisition of informative and semantically rich landslide event features. Finally, a precise and credible landslide susceptibility assessment model is built based on a convolutional neural network (CNN), and landslide susceptibility spatial distribution levels are mapped. The research findings indicate that the CNN-based model outperforms traditional machine learning algorithms in predicting landslide probability; in particular, the Area Under the Curve (AUC) of the model was improved by 3–6% after sample optimization, and the AUC value of the LKF-Cell method was 6–11% higher than the baseline method.

Abstract In the InSAR-DSM, which is the Digital Surface Model (DSM) acquired through Interferometric Synthetic Aperture Radar (InSAR), the presence of significant noise, voids, and elevation anomalies in water bodies is pervasive. These disparities arise from factors such as diminished backscattering due to electromagnetic wave absorption by water, surface-based specular reflection, and the impact of temporal decorrelation. To tackle this challenge, an automated elevation editing method for water bodies based on InSAR-DSM is introduced. Water bodies are automatically categorized as oceans, lakes, or rivers based on their morphological characteristics and location criteria. Distinct algorithms are subsequently deployed for elevation editing within each category. Results demonstrate that after editing water body elevations, all missing values within the original InSAR-DSM for water bodies are effectively addressed. Elevation anomalous pixels within water bodies are reduced by 68.5%, and the elevation standard deviation, which reflects elevation variability in water bodies, is decreased by 47.31%. This process effectively rectifies water body elevations within InSAR-DSM.