Introduction: Diabetic ketoacidosis (DKA) is a life-threatening complication associated with significant morbidity and healthcare burden. Despite advances in diabetes care, recurrent DKA remains common, often reflecting gaps in treatment adherence and patient education. Identifying predictors of recurrence is crucial for risk stratification and targeted intervention. Methodology: We conducted a retrospective observational study of all adult DKA admissions to a tertiary centre between 2001 and 2025. Electronic medical records were reviewed for demographic data, biochemical parameters, precipitating factors, and clinical outcomes. DKA was defined using standard biochemical criteria. Recurrent DKA was defined as ≥2 admissions during the study period. Factors associated with recurrent DKA admissions were analyzed. Patients under the age of 18 years and those with missing vital information were excluded. Results: A total of 667 DKA admissions, comprising 566 patients, were identified, of which 101 admissions (15.1%) were recurrent, involving 65 patients. Among recurrent DKA episodes, the most common precipitating factors were infection (64.4%) and insulin omission (62.4%). After multivariate analyses, patients with type 1 diabetes mellitus (T1DM) were more likely to develop recurrent DKA compared to those with type 2 diabetes mellitus (aOR 4.16; 95% confidence interval [CI] 2.58–6.70; p <0.001). Insulin omission was strongly associated with recurrent DKA (aOR 2.29; 95% CI 1.46–3.60; p <0.001). In contrast, baseline glycated hemoglobin and chronic kidney disease were not significantly associated with recurrence. Diabetic counseling during the first DKA admission did not reduce recurrent DKA. There were no significant differences in mortality (3.9% vs 6.2%, p = 0.524) or critical care admission rates (40.6% vs 38.7%, p = 0.718) between recurrent and first DKA episodes. Conclusion Recurrent DKA accounts for a substantial proportion of DKA admissions and is strongly associated with insulin omission and T1DM. Our findings suggest that recurrent DKA is driven predominantly by behavioral and adherencerelated factors, indicating the need for multidisciplinary interventions beyond standard inpatient counseling.
Diabetic Ketoacidosis ; Retrospective Studies

Introduction: Diabetic ketoacidosis (DKA) in pregnancy is an uncommon yet life-threatening emergency, with disproportionate risks to both mother and fetus. Pregnancy-specific physiological changes predispose patients to rapid metabolic decompensation, often with atypical presentations. Despite this, local data remain limited. We describe the clinical profile, precipitating factors, and outcomes of DKA in pregnancy in a tertiary centre, with emphasis on potentially preventable triggers. Cases: Nine pregnant patients with DKA were identified from a retrospective review of all cases admitted for DKA from 2002 to 2025. Mean age was 31.67 ± 5.20 years; all were Malay. The majority had type 2 diabetes mellitus (55.6%), followed by type 1 diabetes (33.3%) and latent autoimmune diabetes in adults (11.1%). The mean period of amenorrhea was 19.67 ± 12.62 weeks. Infection was the leading precipitant (44.4%), with additional triggers including insulin omission (22.2%), hyperemesis gravidarum, preterm labor, steroid exposure, and perioperative fasting. Most diagnoses were made in the emergency department (55.6%). Biochemical parameters reflected significant severity (mean bicarbonate 7.89 ± 2.98 mmol/L; anion gap 25.00 ± 5.81), with 88.9% classified as severe DKA. Intensive Care Unit (ICU) care was required in 77.8% of cases. The majority (77.8%) were admitted to the ICU unit, with a median time to resolution of 13.00 ± 12.00 hours (interquartile range [IQR]), and the median hospital length of stay was 7.00 ± 5.00 days (IQR). Complications during treatment included hypokalemia (33.3%), acute kidney injury (22.2%), and hypoglycemia (11.1%). Rebound DKA occurred in one-third of patients. All patients were discharged clinically stable. Outcome data demonstrated pregnancy loss in three cases and one preterm birth. Conclusion DKA in pregnancy remains a severe and resource-intensive condition. This series highlights missed opportunities in prevention, with modifiable precipitants such as infection and insulin omission commonly identified. The high severity at presentation suggests delays in recognition. Early detection, optimized metabolic care, and targeted preventive strategies are crucial to improving maternal and fetal outcomes.
Female ; Pregnancy ; Diabetic Ketoacidosis

Background/Aims: Liver transplantation (LT) following total hepatectomy is a life-saving treatment for hepatocellular carcinoma (HCC). The HCC recurrence after LT hinders the effectiveness of the procedure. The objective of this study is to develop a pre-operative risk stratification model based on a liquid biopsy. Methods: We conducted a comprehensive multi-omics study of 260 HCC patients from three centers, including clinical data, low-coverage whole-genome sequencing of cell-free DNA (cfDNA) from plasma, as well as whole-exome, single-nucleus RNA, and spatial transcriptomics from matched tumor and non-tumor tissues. Results: We identified cfDNA-derived copy number alteration (CNA) signatures associated with post-transplant recurrence. By integrating cfDNA-derived CNA profiles with single-cell transcriptomic data, we traced recurrence-associated cfDNA to a distinct subpopulation of malignant cells within the primary tumor. These cells were embedded in a pro-metastatic microenvironment of specialized endothelial subtypes and cancer-associated fibroblasts. Notably, most recurrence-associated lesions were detectable in cfDNA prior to liver transplantation (LT). Building on these insights, we developed the ZJU Criteria based on CNA fragments and tumor markers, a pre-LT risk prediction tool that integrates conventional clinical factors with cfDNA-derived CNA signatures, and validated it using internal and independent external cohorts. Conclusion Our findings suggest that post-transplant recurrence commonly originates from advanced subclones that emerge late during tumor evolution. The ZJU Criteria provides an accurate, non-invasive strategy that significantly improves pre-LT risk stratification and clinical decision-making for patients with HCC.

The initiation of adaptive immune responses relies on the precise recognition and interpretation of antigenic information. In this process, the specific binding of T cell receptors (TCRs) to peptide-major histocompatibility complex (pMHC) molecules represents one of the key molecular events in the initiation of adaptive immune responses. Accordingly, the structural features of TCR-pMHC complexes provide a fundamental basis for dissecting antigen recognition mechanisms and support rational vaccine design, therapeutic target discovery in TCR-based immunotherapy, and TCR identification and optimization. However, experimental determination of TCR-pMHC structures remains costly, time-consuming, and limited in coverage, making computational approaches essential for rapidly obtaining reliable structural information. Computational methods for predicting the structures of TCR-pMHC complexes have advanced rapidly in recent years, driven by progress in deep learning-based modeling frameworks and the increasing availability of structural and sequence resources. Despite these developments, most existing tools do not adequately distinguish the key structural and biophysical differences between MHC class I (MHC-I) and MHC class II (MHC-II) complexes during model construction. As a consequence, their predictive performance differs substantially between class I and class II complexes. In general, structural predictions for class I complexes outperform those for class II complexes. This discrepancy may be related to several fundamental differences between the two systems, including the architecture of the peptide-binding groove, the distribution of peptide lengths, and the properties of peptide flanking residues (PFRs). Compared with MHC-I molecules, MHC-II molecules usually bind longer antigenic peptides, which typically range from 13 to 25 amino acids in length. PFRs at both termini of these peptides participate in regulating the overall conformation of TCR-pMHC class II complexes and exert a pronounced effect on the geometric and physicochemical characteristics of the TCR-pMHC binding interface. Furthermore, within the TCR recognition interface, the complementarity-determining regions (CDRs) consist of segments that differ markedly in conformational behavior. They commonly include regions that are relatively rigid and structurally stable, together with highly flexible segments exhibiting substantial conformational plasticity. These rigidity-flexibility features constitute an essential structural basis enabling TCRs to recognize diverse peptide-MHC ligands and to accommodate conformational heterogeneity at the interface. However, many current modeling tools, in an effort to enforce global conformational stability or reduce structural noise, tend to over-constrain intrinsically flexible regions. Such oversimplification may lead to inappropriate rigidification of flexible CDR loops, resulting in local structural distortions, compromised interface geometry, or even complete modeling failure for specific complexes. Against this background, the review approaches the field from the perspective of computational differences between MHC-I and MHC-II complexes. We first systematically organize and summarize available resources related to TCRs and pMHCs, including structural datasets, sequence databases, prediction tools, and benchmarking studies. We then focus on five representative tools capable of predicting both class I and class II complexes—AlphaFold2, AlphaFold3, TCRmodel2, tFold-TCR, and TCR-pHLA_ModellerS. After excluding structures present in the training sets of these tools, we constructed a benchmark dataset comprising 25 class I and 10 class II TCR-pMHC complexes in the bound state and conducted a systematic evaluation using this dataset. We first employ widely used general evaluation metrics, including All-Atom Root Mean Square Deviation (All-Atom RMSD), Backbone RMSD, Template Modeling score (TM-score), and DockQ, to assess the global conformational accuracy and interface modeling quality of class I and class II complexes. For class II complexes, we propose for the first time a peptide flanking residue deviation index, including the PFRs-Deviation Index (PFRs-DI), N-PFR-Deviation Index (N-PFR-DI), and C-PFR-Deviation Index (C-PFR-DI), to quantitatively characterize conformational deviations in PFRs. In addition, we propose the CDR conformational consistency index (CCC) designed to qualitatively evaluate the ability of prediction tools to capture TCR CDR conformational flexibility. These metrics collectively assess a tool’s ability to model both overall conformation and critical functional regions, thereby addressing the limitations of existing evaluation criteria that overemphasize global structure while inadequately capturing modeling quality in key functional areas. This establishes a unified analytical framework for MHC-I and MHC-II complexes to guide data resource selection, modeling strategy formulation, and evaluation system development. The framework further advances computational modeling and provides crucial support for multi-scale analysis of TCR-pMHC recognition mechanisms and their biological functions.

Radar-based non-contact health monitoring technology (RBNHMT) has emerged as a transformative paradigm in continuous health sensing, enabling non-invasive and continuous monitoring of physiological parameters and behavioral patterns by transmitting electromagnetic waves, analyzing the reflected signals, and detecting subtle bodily movements—ranging from millimeter-scale chest wall displacements due to respiration to micro-scale vibrations associated with cardiac activity—ultimately transforming them into quantifiable health data. Distinguished by its non-contact operation, inherent privacy preservation, and adaptability to diverse scenarios, RBNHMT exhibits stronger resistance to environmental interference than conventional contact-based monitoring, and has solidified its position as a prominent and dynamic research focus in the field of non-contact health monitoring. Currently, significant and multifaceted progress has been made across several key areas. In human activity recognition (HAR), systems leveraging micro-Doppler signatures or point cloud sequences achieve high-precision detection of gait, gestures, and fall events, with state-of-the-art deep learning-based models achieving accuracy rates exceeding 99% in controlled experimental settings. For vital sign and sleep monitoring, it not only tracks respiratory and heart rates continuously but also extracts clinically relevant metrics such as heart rate variability (HRV) for autonomic nervous system assessment and estimates blood pressure through indirect methods like pulse transit time analysis, while maintaining robustness in dynamic settings through advanced motion compensation algorithms. In sleep monitoring, it further enables sleep posture classification and apnea event detection. In emotion and stress recognition, it provides a non-intrusive approach for psychological assessment by analyzing autonomic-response physiological signal patterns or behavioral features. Furthermore, its applications in auxiliary medical diagnosis have expanded to promising interdisciplinary areas such as non-contact heart sound auscultation, radar-based screening for obstructive sleep apnea (OSA), and emerging research into breast cancer detection using microwave and millimeter-wave imaging techniques. However, several challenges impede its practical deployment. Signal quality is significantly compromised by multipath interference in complex indoor environments and clutter from static objects, and by motion artifacts in dynamic scenarios where gross body movements obscure the subtle physiological signals. Algorithmically, separating signals from multiple targets in close proximity and calibrating for substantial individual physiological differences, such as body habitus, baseline vital signs, remain difficult and limit generalizability. Hardware design also faces the challenge of balancing power consumption, cost, integration, and performance, often requiring trade-offs that constrain miniaturization, battery life, or measurement sensitivity. Future advancement, therefore, requires collaborative and targeted innovation across multiple dimensions. Algorithmically, developing adaptive signal processing models based on emerging paradigms such as few-shot learning (for user-specific calibration with minimal data) and reinforcement learning (for dynamic noise suppression) is essential. At the hardware level, highly integrated radar SoCs with embedded processing capabilities and advanced packaging technologies are crucial for achieving the dual goals of device miniaturization and cost reduction without sacrificing performance. At the system level, fusing radar data with complementary modalities such as infrared and acoustic sensing can create a synergistic, multi-modal framework that significantly enhances perceptual robustness and reliability in complex, real-world environments. This review provides a comprehensive synthesis that systematically summarizes the relevant theoretical foundations and application progress, and offers an in-depth analysis of the current technical bottlenecks. It aims to provide a clear development path and a foundational academic reference for the in-depth integration and practical application of RBNHMT in critical scenarios including rehabilitation engineering, smart elderly care, in-vehicle health monitoring, and beyond, thereby offering innovative technical support for the vision of universal, proactive, and personalized health management.

Objective: To establish an accurate and rapid typing method for Rh typing of samples from patients who have received recent blood transfusions by utilizing the difference in osmotic fragility between fresh and old red blood cells. Methods: A lysing solution suitable for destroying old RBCs was prepared. Sixty-one samples collected in our hospital in 2024 with Rh typing of double groups were treated with the lysing solution to remove the old allogeneic red blood cells while preserving the patient's own fresh red blood cells, followed by repeat Rh typing tests. Results: For 61 samples with Rh typing in double groups, 41 were accurately detected identified through the red blood cell lysis method, yielding an identification rate of 67.21%. No significant difference was observed compared to the detection rate of the commonly used capillary centrifugation modified method (χ =0.103, P>0.05). Conclusion: The red blood cell lysis method provides a novel and rapid experimental approach for clinical use in processing Rh-typed samples that are of double groups, thereby offering a basis for Rh compatibility blood transfusion.

Chronic post-surgical pain (CPSP) is a common long-term complication following lung surgery. Its high incidence significantly impacts patients’ quality of life and functional recovery, and imposes a substantial socioeconomic burden. This consensus aims to systematically establish a standardized integrated Chinese and Western medicine diagnostic and treatment framework for chronic post-lung surgery pain (CPLSP). Based on the latest domestic and international evidence-based medical research and multidisciplinary clinical experience, the working group comprehensively elaborates on core issues regarding CPLSP, including its definition, epidemiology, pathogenesis, clinical assessment, Western medical treatment, traditional Chinese medicine (TCM) treatment, and integrated strategies. The consensus emphasizes a patient-centered approach, adhering to the principles of multimodality, individualization, and stepwise management, highlighting the synergistic advantages of integrating Chinese and Western medicine throughout the entire perioperative management cycle encompassing "perioperative anti-inflammation, acute analgesia, and chronic rehabilitation." Through systematic literature retrieval and evidence integration, a total of 9 core recommendations were established to provide scientifically sound and clinically practical guidance.

ObjectiveTo conduct a theoretical study on the medical-rehabilitation integration. MethodsStarting from the background, objectives and content of the medical-rehabilitation integration, this study analyzed its innovative points from the dimensions of conceptual innovation, organizational innovation, model innovation and technological innovation. Results and ConclusionThe medical-rehabilitation integration is an innovation in medical services that takes conceptual innovation as the forerunner, organizational innovation as the foundation, model innovation as the carrier and technological innovation as the core.

OBJECTIVE To investigate the effects and mechanism of the Yishen paidu formula on renal fibrosis in rats with chronic renal failure （CRF） through the reactive oxygen species （ROS）/thioredoxin-interacting protein （TXNIP）/NOD-like receptor thermal protein domain associated protein 3 （NLRP3） pathway. METHODS Rats were randomly divided into control group， model group， Yishen paidu formula low-dose （Yishen paidu formula-L） group， Yishen paidu formula high-dose （Yishen paidu formula- H） group， Yishen paidu formula-H+pcDNA-NC group， and Yishen paidu formula-H+ pcDNA-TXNIP group， with 10 rats in each group. Except for control group， all other rats were fed a diet containing 0.5% adenine to establish a CRF model； the rats were then administered corresponding drugs or normal saline intragastrically or via tail vein， once daily， for 8 consecutive weeks. After the last administration， the levels of serum creatinine （Scr）， blood urea nitrogen （BUN）， ROS， superoxide dismutase （SOD）， malondialdehyde （MDA）， tumor necrosis factor-α （TNF-α）， interleukin （IL）-6， and IL-1β were measured in each group. Pathological changes in renal tissue were observed， and the protein expression levels of Collagen Ⅲ， α-smooth muscle actin （α-SMA）， transforming growth factor-β1 （TGF-β1）， TXNIP and NLRP3 in renal tissue were detected. RESULTS Compared with model group， the renal histopathological damage and fibrosis of rats in Yishen paidu formula-L group and Yishen paidu formula-H group were significantly alleviated. The levels of Scr， BUN， ROS， MDA， TNF- α， IL-6 and IL-1β， and the protein expressions of Collagen Ⅲ， α-SMA， TGF-β1， TXNIP and NLRP3 were significantly decreased， while SOD levels were significantly increased （P＜0.05）. Moreover， the changes were more pronounced in the Yishen paidu formula-H group （P＜0.05）. Compared with Yishen paidu formula-H+pcDNA-NC group， above indexes of rats in Yishen paidu formula-H+pcDNA-TXNIP group were reversed significantly （P＜0.05）. CONCLUSIONS Yishen paidu formula can inhibit renal fibrosis in CRF rats by suppressing the ROS/TXNIP/NLRP3 pathway.

The initiation of adaptive immune responses relies on the precise recognition and interpretation of antigenic information. In this process, the specific binding of T cell receptors (TCRs) to peptide-major histocompatibility complex (pMHC) molecules represents one of the key molecular events in the initiation of adaptive immune responses. Accordingly, the structural features of TCR-pMHC complexes provide a fundamental basis for dissecting antigen recognition mechanisms and support rational vaccine design, therapeutic target discovery in TCR-based immunotherapy, and TCR identification and optimization. However, experimental determination of TCR-pMHC structures remains costly, time-consuming, and limited in coverage, making computational approaches essential for rapidly obtaining reliable structural information. Computational methods for predicting the structures of TCR-pMHC complexes have advanced rapidly in recent years, driven by progress in deep learning-based modeling frameworks and the increasing availability of structural and sequence resources. Despite these developments, most existing tools do not adequately distinguish the key structural and biophysical differences between MHC class I (MHC-I) and MHC class II (MHC-II) complexes during model construction. As a consequence, their predictive performance differs substantially between class I and class II complexes. In general, structural predictions for class I complexes outperform those for class II complexes. This discrepancy may be related to several fundamental differences between the two systems, including the architecture of the peptide-binding groove, the distribution of peptide lengths, and the properties of peptide flanking residues (PFRs). Compared with MHC-I molecules, MHC-II molecules usually bind longer antigenic peptides, which typically range from 13 to 25 amino acids in length. PFRs at both termini of these peptides participate in regulating the overall conformation of TCR-pMHC class II complexes and exert a pronounced effect on the geometric and physicochemical characteristics of the TCR-pMHC binding interface. Furthermore, within the TCR recognition interface, the complementarity-determining regions (CDRs) consist of segments that differ markedly in conformational behavior. They commonly include regions that are relatively rigid and structurally stable, together with highly flexible segments exhibiting substantial conformational plasticity. These rigidity-flexibility features constitute an essential structural basis enabling TCRs to recognize diverse peptide-MHC ligands and to accommodate conformational heterogeneity at the interface. However, many current modeling tools, in an effort to enforce global conformational stability or reduce structural noise, tend to over-constrain intrinsically flexible regions. Such oversimplification may lead to inappropriate rigidification of flexible CDR loops, resulting in local structural distortions, compromised interface geometry, or even complete modeling failure for specific complexes. Against this background, the review approaches the field from the perspective of computational differences between MHC-I and MHC-II complexes. We first systematically organize and summarize available resources related to TCRs and pMHCs, including structural datasets, sequence databases, prediction tools, and benchmarking studies. We then focus on five representative tools capable of predicting both class I and class II complexes—AlphaFold2, AlphaFold3, TCRmodel2, tFold-TCR, and TCR-pHLA_ModellerS. After excluding structures present in the training sets of these tools, we constructed a benchmark dataset comprising 25 class I and 10 class II TCR-pMHC complexes in the bound state and conducted a systematic evaluation using this dataset. We first employ widely used general evaluation metrics, including All-Atom Root Mean Square Deviation (All-Atom RMSD), Backbone RMSD, Template Modeling score (TM-score), and DockQ, to assess the global conformational accuracy and interface modeling quality of class I and class II complexes. For class II complexes, we propose for the first time a peptide flanking residue deviation index, including the PFRs-Deviation Index (PFRs-DI), N-PFR-Deviation Index (N-PFR-DI), and C-PFR-Deviation Index (C-PFR-DI), to quantitatively characterize conformational deviations in PFRs. In addition, we propose the CDR conformational consistency index (CCC) designed to qualitatively evaluate the ability of prediction tools to capture TCR CDR conformational flexibility. These metrics collectively assess a tool’s ability to model both overall conformation and critical functional regions, thereby addressing the limitations of existing evaluation criteria that overemphasize global structure while inadequately capturing modeling quality in key functional areas. This establishes a unified analytical framework for MHC-I and MHC-II complexes to guide data resource selection, modeling strategy formulation, and evaluation system development. The framework further advances computational modeling and provides crucial support for multi-scale analysis of TCR-pMHC recognition mechanisms and their biological functions.