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.

Objective:To explore the key genes and its biological functions of aldosterone producing adenoma (APA) using bioinformatics analysis.Methods:Differentially expressed genes of APA were identified from two training datasets GSE60042 and GSE64957 in GEO database. Function and pathway enrichment analyses for differentially expressed genes were performed and transcriptional regulation network among these genes were determined. Hub genes were extracted by node analysis from the protein-protein interaction (PPI) network. The expression of key genes was verified by a testing dataset GSE8514. Receiver operating characteristic(ROC) curve analysis was applied to assess the diagnostic efficiency of key genes in APA. The biofunction of each key gene were determined by gene set enrichment analysis (GSEA).Results:A total of 68 differentially expressed genes, including 33 up-regulated genes and 35 down-regulated genes, were detected from the training datasets. These genes were mainly enriched in aldosterone biosynthetic process, calcium signaling pathway, serotonin receptor signaling pathway, transcriptional activator activity, and regulation of transcription. JUN and VDR were at the center of the transcriptional factor-gene network. Furthermore, we identified nine Hub genes from the PPI network. In testing dataset, CYP11B2 and VDR showed the higher expression, while JUN, NFKBIZ, EGR3, and KLF6 showed lower expression in APA (all P<0.05), and the value of area under ROC curve analysis was 0.936, 0.833, 0.953, 0.854, 0.868, and 0.929, respectively. GSEA indicated the alter of key genes in APA led to up-regulation of the steroid biosynthesis, cell adhesion molecules, immune cells signaling pathway, and complement and coagulation cascades [all normalized enrichment score (NES)>1.5, P<0.05], but down-regulation of the DNA replication, ribosome, and autophagy (all NES<-1.5, P<0.05). Conclusion:Results of bioinformatics indicate that JUN and VDR are key transcriptional factors, and CYP11B2, NFKBIZ, EGR3, and KLF6 are the key genes for APA, which are involved in the steroid biosynthesis, cell adhesion molecules, immune cells signaling pathway in APA.

Objective: To investigate the characteristic and prognostic significance of leukemia stem cells associated antigens expressions including CD34, CD38, CD123, CD96 and TIM-3 in t (8;21) AML. Methods: Bone marrow samples of 47 t (8;21) AML patients were collected at diagnosis from October 2015 to April 2018 in Peking University Peoples’ Hospital, then flow cytometry method was performed to detect the expression frequencies of CD34, CD38, CD123, CD96 and TIM-3 to analyze the relationship between leukemia stem cells associated antigens expressions and relapse. Results: Of 47 t (8;21) AML patients tested, the median percentages of CD34⁺CD38^-, CD34⁺ CD38^-CD123⁺, CD34⁺CD38^- CD96⁺ and CD34⁺ CD38^- TIM-3⁺ cells among nucleated cells were 2.37%, 0.24%, 0.27% and 0.06%, respectively. All the frequencies of CD34⁺CD38^-, CD34⁺CD38^-CD123⁺, CD34⁺CD38^-CD96⁺ and CD34⁺ CD38^-TIM-3⁺ cells had no impact on the achievement of CR after the first course of induction. All higher frequencies of CD34⁺CD38^-, CD34⁺CD38^-CD123⁺, CD34⁺CD38^-CD96⁺ cells were related to higher 2-year CIR rate. Whereas, the frequency of CD34⁺ CD38^- TIM-3⁺ cells had no impact on CIR rate. Both high frequency of CD34⁺ CD38^- cells and the high level of minimal residual diseases (patients with <3-log reduction in the RUNX1-RUNX1T1 transcript level after the second consolidation therapy) were independent poor prognostic factors of CIR[P=0.025, HR=6.9 (95%CI 1.3-37.4) ; P=0.031, HR=11.1 (95%CI 1.2-99.2) ]. Conclusion Different leukemia stem cells associated antigens had distinct prognostic significance in t (8;21) AML. High frequencies of CD34⁺ CD38^-, CD34⁺ CD38^- CD123⁺ and CD34⁺CD38^-CD96⁺ cells at diagnosis predicted relapse in patients with t (8;21) AML.

Objective To explore the value of digital mammography in diagnosis of the negative palpable breast cancer in rural ar-eas.Methods 6 754 women randomly selected among 13 700 female outpatients (range,30-83 years)in rural areas underwent dig-ital mammography no matter they had breast palpable masses or not.Results 72 cases with breast cancer were diagnosed by digital mammography (72/6 754).Clinical palpation showed positive in 59 cases (59/72)and negative in 13 cases (13/72)with aged 35-77 years.Among 13 cases with negative palpable breast cancer,6 cases were carcinomas in situ,3 cases were intraductal carcinomas,and 4 cases were infiltrating ductal carcinomas.Conclusion Digital mammography may help in detection of the negative palpable breast cancer in early time.