Objective: To fabricate a hydrogel loaded with inositol hexaphosphate-zinc and preliminarily evaluate its performance in self-mineralization and osteoinduction, thereby providing a theoretical basis for the development of bone regeneration materials. Methods: The hydrogel framework (designated DF0) was formed by copolymerizing methacryloyloxyethyltrimethylammonium chloride and four-armed poly(ethylene glycol) acrylate, followed by sequentially loading inositol hexaphosphate anions via electrostatic interaction and zinc ions via chelation. The hydrogel loaded only with inositol hexaphosphate anions was named DF1, while the co-loaded hydrogel was named DF2. The self-mineralization efficacy of the DF0 , DF1 and DF2 hydrogels was characterized using scanning electron microscopy, transmission electron microscopy (TEM), energy dispersive spectroscopy (EDS), and selected area electron diffraction (SAED). The biocompatibility was assessed via live/dead cell staining and a CCK-8 assay. The osteoinductive capacity of the DF0 , DF1 and DF2 hydrogels on MC3T3-E1 cells was assessed via alkaline phosphatase (ALP) and Alizarin Red S (ARS) staining. In the aforementioned cell experiments, cells cultured in standard medium served as the control group Results: The DF0, DF1, and DF2 hydrogels were successfully synthesized. Notably, DF1 and DF2 exhibited distinct self-mineralization within 6 days. Results from TEM, EDS, and SAED confirmed that the mineralization products were amorphous calcium phosphate in group DF1, and amorphous calciumzinc phosphate in group DF2. Biocompatibility tests revealed that none of the hydrogels (DF0, DF1, and DF2) adversely affected cell viability or proliferation. In osteogenic induction experiments, both ALP and ARS staining were intensified in the DF1 and DF2 groups, with the most profound staining observed in the DF2 group. Conclusion The developed inositol hexaphosphate-zinc hydrogel (DF2) demonstrates the dual capacity to generate calcium-phosphate compounds through self-mineralization while exhibiting excellent osteoinductive properties. This biocompatible, dual-promoting osteogenic hydrogel presents a novel strategy for bone regeneration.

Objective To establish an analytical strategy covering multi-dataset processing, recursive feature elimination (RFE) screening and multi-model evaluation based on multiple machine learning algorithms, so as to screen radiation-sensitive genes and verify the feasibility of the evaluation strategy. Methods Qualified radiation transcriptome datasets were retrieved from public gene expression databases. Following standardized data preprocessing and feature preselection, 13 machine learning algorithms were adopted to construct models. The performance of each model was compared and validated in independent datasets. Results A total of 38 eligible datasets were included. Sixteen differentially expressed genes unreported in existing literature were screened out, among which ugcrhl, pdcl3, mct4, h2-g2 and fam120aos were correlated with radiation phenotypes. Ensemble learning algorithms including random forest and gradient boosting exhibited the optimal comprehensive performance. Independent dataset verification confirmed that the screened genes overlapped with known radiation-sensitive genes, and the model performance was consistent with the findings. Conclusion The machine learning strategy constructed in this study can effectively explore potential radiation-sensitive genes, and provides methodological support for subsequent relevant studies.

Objective To establish an analytical strategy covering multi-dataset processing, recursive feature elimination (RFE) screening and multi-model evaluation based on multiple machine learning algorithms, so as to screen radiation-sensitive genes and verify the feasibility of the evaluation strategy. Methods Qualified radiation transcriptome datasets were retrieved from public gene expression databases. Following standardized data preprocessing and feature preselection, 13 machine learning algorithms were adopted to construct models. The performance of each model was compared and validated in independent datasets. Results A total of 38 eligible datasets were included. Sixteen differentially expressed genes unreported in existing literature were screened out, among which ugcrhl, pdcl3, mct4, h2-g2 and fam120aos were correlated with radiation phenotypes. Ensemble learning algorithms including random forest and gradient boosting exhibited the optimal comprehensive performance. Independent dataset verification confirmed that the screened genes overlapped with known radiation-sensitive genes, and the model performance was consistent with the findings. Conclusion The machine learning strategy constructed in this study can effectively explore potential radiation-sensitive genes, and provides methodological support for subsequent relevant studies.

Objective To establish an analytical strategy covering multi-dataset processing, recursive feature elimination (RFE) screening and multi-model evaluation based on multiple machine learning algorithms, so as to screen radiation-sensitive genes and verify the feasibility of the evaluation strategy. Methods Qualified radiation transcriptome datasets were retrieved from public gene expression databases. Following standardized data preprocessing and feature preselection, 13 machine learning algorithms were adopted to construct models. The performance of each model was compared and validated in independent datasets. Results A total of 38 eligible datasets were included. Sixteen differentially expressed genes unreported in existing literature were screened out, among which ugcrhl, pdcl3, mct4, h2-g2 and fam120aos were correlated with radiation phenotypes. Ensemble learning algorithms including random forest and gradient boosting exhibited the optimal comprehensive performance. Independent dataset verification confirmed that the screened genes overlapped with known radiation-sensitive genes, and the model performance was consistent with the findings. Conclusion The machine learning strategy constructed in this study can effectively explore potential radiation-sensitive genes, and provides methodological support for subsequent relevant studies.

AIM: To explore the application effect of eyelid margin incision combined with floating bridge technique in the treatment of ptosis patients, and to analyze its effects on the margin reflex distance 1(MRD1)and the aesthetic effect of the eyes. METHODS:Prospective study. This study enrolled patients with blepharoptosis treated at the Department of Ophthalmology of the hospital between January 2023 and July 2025, who were assigned to either an observation group or a control group. The patients in the control group received levator aponeurosis shortening surgery through conventional double-eyelid incision, while the patients in the observation group underwent levator aponeurosis shortening surgery using the eyelid margin incision combined with floating bridge technique. Subsequently, the surgical-related indicators,the correction effect of ptosis, the aesthetic effect, the ocular function indicators, the postoperative complication occurrence rate, and recurrence rate were compared and analyzed between two groups.RESULTS:A total of 100 patients(100 eyes)with blepharoptosis were enrolled in this study, with 50 patients(50 eyes)in the observation group and 50 patients(50 eyes)in the control group. The observation group had a mean age of 27.33±5.32 y, including 20 males and 30 females. The control group had a mean age of 27.64±4.66 y, including 24 males and 26 females. Compared with the control group, the operative time of the observation group was longer, the postoperative swelling resolution time and natural recovery time were shorter,and the correction rate was higher(P<0.05). Three months after the surgery, the VSS score and POSAS score of the observation group were lower than those of the control group, while the scores for eyelid curvature, symmetry of eyelid height, and the PFH, MRD1, palpebral fissure size, and upper eyelid levator muscle strength were all higher than those of the control group(all P<0.05). However, there was no significant difference in postoperative complication occurrence rate and recurrence rate between two groups(all P>0.05).CONCLUSION:The combined use of eyelid margin incision and floating bridge technique for treating ptosis results in slightly longer operation time, but it can shorten the postoperative recovery period, effectively ameliorate the MRD1 and ocular function of patients, and enhance the correction effect and aesthetic quality of the eyes. In addition, it has a low incidence of complications and guaranteed safety.

Objective To investigate predictive value of soluble growth stimulation-expressed gene 2 protein (sST2), cardiac troponin T (cTnT), and creatine kinase-myocardial band (CK-MB) for steroid resistance and prognosis in patients with immune checkpoint inhibitor-associated myocarditis (ICIAM). Methods A total of 28 patients with ICIAM were prospectively enrolled from March 2023 to February 2024 at Zhongshan Hospital, Fudan University. All participants were divided into the steroid-resistant ICIAM (srICIAM) group and the non-srICIAM group. Baseline serum levels of sST2, cTnT and CK-MB were detected prior to treatment initiation. ROC curve analysis was applied to evaluate the predictive value of these biomarkers, and Kaplan-Meier survival analysis was performed to explore the relationship between sST2 and all-cause mortality. Results Baseline sST2, cTnT, and CK-MB levels were significantly higher in the srICIAM group (n＝15) than in the non-srICIAM group (n＝13, P＜0.05). Combination of cTnT and sST2 improved predictive performance for srICIAM (AUC＝0.857, sensitivity 0.714, specificity 1.000). After 6 months of follow-up, the high sST2 group had a significantly higher all-cause mortality than the low sST2 group (P＝0.016). Conclusions The combination of sST2 and cTnT could effectively predict srICIAM, and elevated sST2 is associated with increased higher all-cause mortality in ICIAM patients, providing a basis for early risk stratification and treatment optimization.

Yttrium-90 selective internal radiation therapy (⁹⁰Y-SIRT) is a treatment technique that delivers radioactive microspheres precisely to the arterial vascular bed of neoplasms, utilizing beta radiation to administer a high local dose of radiation to the neoplasm tissues. This technology has demonstrated significant efficacy in patients with unresectable pirmary liver cancers and liver metastases. This article systematically reviews the development history and clinical application status of ⁹⁰Y-SIRT in the treatment of liver cancer, and looks forward to future development directions.

Cardiovascular disease (CVD), chronic kidney disease (CKD), and metabolic disorders are the 3 major chronic diseases threatening human health, which are closely related and often coexist, significantly increasing the difficulty of disease management. In response, the American Heart Association (AHA) proposed a novel disease concept of “cardiovascular-kidney-metabolic (CKM) syndrome” in October 2023, which has triggered widespread concern about the co-treatment of heart and kidney diseases and the prevention and treatment of metabolic disorders around the world. This review posits that effectively managing CKM syndrome requires a new and multidimensional paradigm for diagnosis and risk prediction that integrates biological insights, advanced technology and social determinants of health (SDoH). We argue that the core pathological driver is a “metabolic toxic environment”, fueled by adipose tissue dysfunction and characterized by a vicious cycle of systemic inflammation and oxidative stress, which forms a common pathway to multi-organ injury. The at-risk population is defined not only by biological characteristics but also significantly impacted by adverse SDoH, which can elevate the risk of advanced CKM by a factor of 1.18 to 3.50, underscoring the critical need for equity in screening and care strategies. This review systematically charts the progression of diagnostic technologies. In diagnostics, we highlight a crucial shift from single-marker assessments to comprehensive multi-marker panels. The synergistic application of traditional biomarkers like NT-proBNP (reflecting cardiac stress) and UACR (indicating kidney damage) with emerging indicators such as systemic immune-inflammation index (SII) and Klotho protein facilitates a holistic evaluation of multi-organ health. Furthermore, this paper explores the pivotal role of non-invasive monitoring technologies in detecting subclinical disease. Techniques like multi-wavelength photoplethysmography (PPG) and impedance cardiography (ICG) provide a real-time window into microcirculatory and hemodynamic status, enabling the identification of early, often asymptomatic, functional abnormalities that precede overt organ failure. In imaging, progress is marked by a move towards precise, quantitative evaluation, exemplified by artificial intelligence-powered quantitative computed tomography (AI-QCT). By integrating AI-QCT with clinical risk factors, the predictive accuracy for cardiovascular events within 6 months significantly improves, with the area under the curve (AUC) increasing from 0.637 to 0.688, demonstrating its potential for reclassifying risk in CKM stage 3. In the domain of risk prediction, we trace the evolution from traditional statistical tools to next-generation models. The new PREVENT equation represents a major advancement by incorporating key kidney function markers (eGFR, UACR), which can enhance the detection rate of CKD in primary care by 20%-30%. However, we contend that the future lies in dynamic, machine learning-based models. Algorithms such as XGBoost have achieved an AUC of 0.82 for predicting 365-day cardiovascular events, while deep learning models like KFDeep have demonstrated exceptional performance in predicting kidney failure risk with an AUC of 0.946. Unlike static calculators, these AI-driven tools can process complex, multimodal data and continuously update risk profiles, paving the way for truly personalized and proactive medicine. In conclusion, this review advocates for a paradigm shift toward a holistic and technologically advanced framework for CKM management. Future efforts must focus on the deep integration of multimodal data, the development of novel AI-driven biomarkers, the implementation of refined SDoH-informed interventions, and the promotion of interdisciplinary collaboration to construct an efficient, equitable, and effective system for CKM screening and intervention.

Objective: To explore the relationship between nuclear factor-kappa B subunit 1 (NFKB1) and LIM-homeobox gene 2 (LHX2) polymorphisms and esophageal cancer susceptibility, so as to provide the reference for the prevention and treatment of esophageal cancer. Methods: A total of 100 patients with primary esophageal cancer diagnosed at the Affiliated Tumor Hospital of Xinjiang Medical University from 2019 to 2023 were selected as the case group, and 100 healthy individuals undergoing physical examination during the same period of time were selected as the control group. Demographic information, disease history and lifestyle data were collected through questionnaire surveys. The single nucleotide polymorphisms at the rs28362491 and rs4648068 loci of NFKB1 gene as well as rs10760310 and rs10121751 loci of LHX2 gene were detected using multiplex high-temperature ligase detection reaction technology. The relationship between these loci and esophageal cancer susceptibility were analyzed using a multivariable conditional logistic regression, linkage disequilibrium and haplotype analysis. The impact of the interaction between the above-mentioned loci and environmental factors on esophageal cancer susceptibility using the generalized multifactor dimensionality reduction (GMDR) method. Results: The case group comprised 73 males and 27 females, with a mean age of (64.02±8.90) years. The control group included 73 males and 27 females, with a mean age of (64.54±9.43) years. The genotype distributions of rs28362491, rs4648068, rs10760310 and rs10121751, loci in both groups conformed to Hardy-Weinberg equilibrium (all P>0.05). Multivariable conditional logistic regression analysis showed that rs10760310 and rs10121751 loci of LHX2 gene were associated with the esophageal cancer susceptibility (both P<0.05). The overdominant model of rs10760310 loci of LHX2 gene had the lowest Akaike information criterion value (OR=0.22, 95%CI: 0.10-0.47). GAA haplotypes at rs4648068, rs10760310 and rs10121751 loci were associated with a lower risk of esophageal cancer susceptibility (OR=0.26, 95%CI: 0.13-0.50). GMDR analysis revealed a statistically significant interaction between rs10760310 loci and smoking on esophageal cancer susceptibility (P<0.05, cross-validation consistency coefficient: 10/10). Conclusion The rs10760310 and rs10121751 loci polymorphisms of LHX2 gene may be associated with esophageal cancer susceptibility, and there is an interaction between rs10760310 loci and smoking on the esophageal cancer susceptibility.