Objective: To investigate the factors influencing intraoperative blood transfusion in patients with hip fractures and to develop a machine learning (ML) model for predicting this risk. Methods: A total of 424 patients with hip fractures who underwent surgical treatment between November 2022 and March 2025 in our hospital were selected. Key feature variables of intraoperative blood transfusion risk were identified using the Boruta algorithm. Four different ML algorithms—support vector machine (SVM), linear discriminant analysis (LDA), mixed discriminant analysis (MDA), and extreme gradient boosting (XGBoost)—were used to develop predictive models for intraoperative blood transfusion risk. The predictive performance of the four ML models were evaluated using accuracy, precision, receiver operating characteristic (ROC) curves, precision-recall curves (PRC), precision-recall gain curves (PRGC), and F1 scores. Shapley additive interpretation (SHAP) was used to interpret the final model. Results: Among the 424 patients, 77(18.2%) received intraoperative blood transfusion. The Boruta algorithm identified albumin (ALB), activated partial thromboplastin time (APTT), types of anesthesia, types of fracture, and hemoglobin (Hb) as key feature variables for predicting intraoperative blood transfusion risk. In model evaluation, the SVM model outperforms the other three models across multiple metrics, including the area under the receiver operating characteristic curve (AUC), recall, recall gain, accuracy, precision, F1 score, and the area under the precision-recall curve (PRC-AUC). The SVM model, interpreted and visualized based on SHAP values, effectively predicted intraoperative blood transfusion risk in patients with hip fracture. A visual online application was developed based on the SVM model (https://pbo-nomogram.shinyapps.io/blood/). Conclusion: Preoperative low ALB and Hb levels, prolonged APTT, general anesthesia, and intertrochanteric fractures are risk factors for intraoperative blood transfusion in hip fracture patients. The risk prediction model for intraoperative blood transfusion constructed based on the SVM algorithm has optimal performance, which provides new ideas and methods for the clinical early identification of hip fracture patients with high transfusion risk and the implementation of targeted interventions.

Persistent left superior vena cava is a rare venous variant that is often combined with cardiovascular malformations. In thoracic surgery, especially mediastinal tumor resection, neglect of this variant may make the surgery difficult and risky, and careful preoperative imaging interpretation and adequate preoperative evaluation play an important role in the perioperative safety of the patient. In this paper, we reported a case of a 17-year-old female patient with a persistent left superior vena cava combined with mediastinal tumors. She was successfully discharged 5 days after thoracoscopic surgery, and after 3 years of postoperative follow-up, no tumor recurrence was observed.

Objective To assess the prognostic value of serum lysine hydroxylase 3(PLOD3)and cytokeratin 19 fragment(CYFRA21-1)in predicting the efficacy of three-dimensional brachytherapy in patients with pulmonary metastases.Methods A total of 102 patients with lung metastases who underwent three-dimensional brachytherapy at our hospital were selected as the lung metastasis group from August 2021 to August 2023.During the same period,a control group consisting of 60 healthy individuals who underwent physical examinations was selected.The lung metastasis group was further divided into an effective group(n=66)and an ineffective group(n=36)based on therapeutic outcomes.Enzyme-linked immunosorbent assay was employed to measure the levels of serum PLOD3 and CYFRA21-1.Multiple logistic regression analysis was conducted to identify factors influencing the efficacy of three-dimensional brachytherapy in the lung metastasis group.Receiver operating characteristic(ROC)curves were used to assess the predictive value of serum PLOD3 and CYFRA21-1 for treatment efficacy.Results The levels of serum PLOD3 and CYFRA21-1 in the case group were(34.47±6.17)μg/L,(27.85±5.14)μg/L,respectively,which exhibited significantly higher values compared to those observed in the control group(7.26±2.21)μg/L,(9.31±2.46)μg/L(P<0.05).A positive correlation was found between serum PLOD3 and CYFRA21-1 in the Lung metastasis group(r=0.667,P=0.000).Logistic regression analysis revealed that multiple metastatic lesions along with elevated serum levels of PLOD3 and CYFRA21-1 were identified as risk factors for the efficacy of three-dimensional brachytherapy in patients with lung metastases(P<0.05).Furthermore,when combined with three-dimensional brachytherapy,the area under the curve(AUC)for serum PLOD3 and CYFRA21-1 was calcu-lated as 0.868,demonstrating a superior performance compared to individual measurements of either serum PLOD3 alone(AUC=0.815)or CYFRA21-1 alone(P<0.05).Conclusions The levels of serum PLOD3 and CYFRA21-1 are elevated in patients with lung metastases,exhibiting a significant correlation with the efficacy of three-dimensional brachytherapy.The combined utilization of these two biomarkers demonstrates a robust predictive value for treatment efficacy in patients suffering from lung metastases.

Objective:To explore the changes in carcinoembryonic antigen(CEA)during the immune process of programmed death-1(PD-1)inhibitor Sintilimab in human epithelial growth factor receptor 2(HER2)negative advanced gastric cancer patients and its relationship with prognosis.Methods:HER2 negative late stage gastric cancer patients(88 cases)who were treated in North Anhui Coal Power Group General Hospital from May 2020 to April 2022 were selected as study subjects,all of whom received PD-1 inhibitor Sintilimab treatment;according to the prognosis,they were divided into death group(36 cases)and survival group(52 cases).Followed up was conducted every 6 months to collect tumor marker levels before and after treatment.Analyzed relationship between tumor markers(CEA,CA199,CA125)levels and clinical staging,lymph node metastasis and prognosis.Kaplan-Meier sur-vival curve was used to analyze survival time of patients with negative and positive tumor markers.Multivariate Cox regression model and stepwise regression analysis were used to identify risk factors affecting prognosis.Spearman was used for correlation analysis.Results:After two cycles of treatment,72 cases(81.82%)had disease control and 16 cases(18.18%)had progression.Compared with patients before treatment,positive rate of serum tumor markers in patients after treatment was significantly reduced(P<0.05).Positive rates of CEA and CA199 were significantly correlated with clinical staging(P<0.05).When predicting patient death,sensitivity of CEA level was the highest(48.57%),while CA125 had the highest specificity(95.62%)and the lowest sensitivity(25.71%).Kaplan-Meier sur-vival curve analysis showed that the survival time of patients with positive tumor markers were significantly shorter than that of negative patients(P<0.05).Clinical staging,serum CEA and CA199 levels were independent factors for predicting prognosis(P<0.05).Spear-man correlation analysis results showed that the multiple increases in CEA(r=-0.512,P=0.005)and CA199(r=-0.467,P=0.011)were negatively correlated with patient survival time.Conclusion:After treatment with PD-1 inhibitor Sintilimab,serum tumor markers levels in HER2 negative advanced gastric cancer patients have been significantly reduced on average.High serum levels of CEA,CA199 and CA125 can all predict poor prognosis,and clinical staging,serum CEA and CA199 levels are independent factors in pre-dicting prognosis.The higher the increase in CEA and CA199,the shorter the patient's survival time.

Objective:To design a morning quality assurance method for pencil-beam scanning proton system to achieve integrated measurement of multiple parameters.Methods:A functionally partitioned morning check phantom was designed and manufactured, which was fixed to a specific position on the treatment bed with a 3D-printed clip, along with the two-dimensional matrix ionization chamber, for consistency checks of proton field and beam-related parameters. Additionally, a groove for an imaging phantom was reserved on one side of the clip for the functional check of the onboard imaging guidance system. The sensitivity and specificity of the aforementioned morning check method were tested, demonstrating its effectiveness. The morning check data from a rotating beam treatment room at the Hefei Ion Medical Center over a continuous period of 7 months (126 d) were analyzed. The output, field flatness, symmetry, field size and the duration of morning check were observed.Results:The results showed that the changes in the output, field flatness, and symmetry were all within 1%, the change in the field size was within 0.5 mm, and the range variations for both 155 MeV and 240 MeV energy levels were within 1 mm. The changes in the spot size for the four energy levels of 100 MeV, 130 MeV, 160 MeV, and 190 MeV were all within 2%, and the spot position deviations were within 1.5 mm. The entire morning check process could be completed within 20 min.Conclusions:The morning check method designed and manufactured in this study specifically for pencil-beam scanning proton therapy can efficiently and integrally measure various proton system parameters and can be used as an implementation method for clinical proton therapy morning check.

This study aims to explore different construction methods for animal models of traditional Chinese medicine(TCM)syndromes and their advantages and disadvantages,to propose optimization strategies for existing problems in current construction methods,and to provide reference for constructing animal models of TCM syndromes that both preserve the essence of TCM syndromes and conform to modern scientific research standards.Using"traditional Chinese medicine","syndrome",and"animal model"as key words,articles related to animal models of TCM syndromes from CNKI,Wanfang,and VIP databases are searched and reviewed.Then the theoretical basis,technical characteristics,and existing problems of the main construction methods of current TCM syndrome animal models are systematically sorted out,and corresponding optimization measures are proposed for the existing problems.The construction methods of TCM syndrome animal models include TCM etiology and pathogenesis construction,modern medical etiology and pathology construction,and integration of TCM and Western medicine for diseases and syndromes.The TCM etiology and pathogenesis construction method is guided by a holistic perspective,constructing syndrome models by simulating external factors such as six pathogenic factors and emotional disorders.Although it conforms to TCM theoretical connotation and has simple operation and strong controllability,this method has problems such as low modeling success rate and poor etiology-syndrome fit.The modern medical etiology and pathology construction method is based on microscopic pathological mechanisms,adopting highly controllable technical means such as drug intervention and surgical modeling.Although it has the characteristics of clear objective indicators and excellent reproducibility,this method has defects such as deviation from the essence of TCM"syndrome"and insufficient safety.The integrated TCM-Western medicine disease-syndrome method shows significant complementarity in syndrome essence restoration degree and technical feasibility,achieves systematic integration of TCM basic theories and clinical syndrome differentiation thinking in methodology,and integrates the objective evaluation system of modern medicine,improving the clinical consistency between Western medicine pathological mechanisms and TCM syndrome evolution patterns.However,this method still faces common challenges such as ambiguous syndrome identification standards and distortion of disease progression simulation.The construction of TCM syndrome animal models faces challenges such as poor theoretical adaptability and poor technical standardization,but has irreplaceable value in verifying the efficacy of prescriptions and promoting the internationalization of TCM.In the future,the construction of TCM syndrome animal models should be optimized through measures such as optimizing animal selection,improving the theoretical basis of preparation methods,standardizing the setting of modeling factors,and clarifying the standard for modeling success.

Tracheal intubation is the most commonly used way to establish artificial airway in intensive care unit patients, and it is the premise of respiratory support and treatment, the success of extubation is an important basis to measure the prognosis of patients. Influenced by many factors, extubation of patients may fail, resulting in prolonged hospitalization, increased medical expenses and increased incidence of complications. This article reviews the literature at home and abroad, and summarizes the main mechanism, risk factors and prevention strategies of extubation failure in intensive care unit patients, aiming at improving the recognition ability of clinical medical staff for extubation failure and providing reference for clinical management and follow-up research.

Objective:To analyze the association between inflammation-related dietary patterns and the risk for cognitive impairment in older adults aged ≥65 years in longevity areas in China by using reduced rank regression (RRR) analysis.Methods:This study used cross-sectional data from the 2021 Healthy Aging and Biomarkers Cohort Study, including the information about study participants' demographic characteristics, lifestyles, daily life activities, and disease histories. Dietary intake was obtained by using a simplified food frequency questionnaire. Cognitive impairment was evaluated based on the Mini-Mental State Examination Scale combined with years of education. Fasting venous blood samples were collected to detect inflammatory markers, especially high-sensitivity C-reactive protein (hs-CRP) and the platelet-to-lymphocyte ratio (PLR). RRR analysis was used to obtain inflammation-related dietary patterns using hs-CRP and PLR as response variables. Multivariate logistic regression model was used to analyze the association between dietary pattern score and the risk for cognitive impairment. Restricted cubic spline was used to explore the dose response relationship, and mediation analysis was used to quantify the mediating effects of hs-CRP and PLR.Results:Two dietary patterns were identified with RRR. The primary pattern was characterized by higher intakes of flour, red meat, and dairy products, and lower intake of fresh vegetables, explaining 6.84% of the variance in food intake and 0.50% of the variance in inflammatory markers. Compared with the T1 group, the T3 group had significantly higher risk for cognitive impairment ( OR=1.242, 95% CI: 1.034-1.491). Each one standard deviation increase in the dietary pattern score was associated with an 8.7% increase in the risk for cognitive impairment ( OR=1.087, 95% CI: 1.008-1.172), with a significant linear trend (overall-model P<0.001, non-linear P=0.295). Mediation analysis indicated that hs-CRP mediated 6.2% of the association between the dietary pattern and the risk for cognitive impairment. Conclusion:The inflammation- related dietary pattern characterized by higher consumption of flour, red meat, and dairy products and lower consumption of fresh vegetables is associated with an increased risk for cognitive impairment in older adults, and hs-CRP partially mediates this association.

Objective:To report the preliminary clinical exploration result of spatially fractionated radiation therapy (SFPT) using the pencil-beam-scanning (PBS) technique at a single center for the treatment of patients with bulky tumors.Methods:Data on the clinical characteristics, tumor characteristics, and dosimetric parameters were retrospectively collected from patients with bulky tumors at the Radiation Oncology Department, Ion Medical Center, First Affiliated Hospital of University of Science and Technology of China (i.e., the Hefei Ion Medical Center) from April 2024 to December 2024. Three-dimensional lattice radiotherapy (LRT) was primarily utilized in the SFRT, with multi-field robust optimization performed using pencil beam scanning. SFRT target volumes (STVs) were defined as 1.0 cm-diameter spheres. Primary observation indicators included the remission rate of tumor-associated symptoms, followed by the local tumor control rate.Results:A total of 12 patients were enrolled in this study, with a median age of 61 years (28-85 years). The primary tumors included hepatocellular carcinoma (six patients), sarcomas (three patients), and lung cancer (two patients). Eight patients received concurrent systemic therapy. The SFRT plans showed a median gross tumor volume (GTV) of 429.63 cm 3 (120.60-2 053.30 cm 3), a median STV number of 8 (3-20), a single-fraction dose to STVs of 10 GyE, a median irradiation quantity of 6 (3-8), a median STV of 8 cm 3 (3-20 cm 3), a median STV proportion of 2.09% (0.62%-3.30%), a median GTV corresponding to a single STV of 52.91 cm 3 (30.25-159.82 cm 3), and a median peak-to-valley dose ratio of 3.37 (2.29-7.60). All patients received conventionally fractionated proton therapy (CFPT), with a median prescription dose of 50 GyE (40-60 GyE). Furthermore, these patients showed a median follow-up time of 174 d (133-235 d), a remission rate of tumor-associated symptoms of 75%, and a local control rate of 100%. Four patients experienced grade 1-2 treatment-related adverse events, suggesting high overall tolerability of the patients. Conclusions:SFRT represents a promising technique with high control rates and tolerability for bulky tumors, providing the possibility for quick symptom relief and the control of tumor progression.