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.

ObjectiveTo observe the analgesic efficacy and safety of Qihuang acupuncture theory combined with opioid analgesics in patients with moderate to severe cancer pain due to lung cancer. MethodsPatients with moderate to severe cancer pain from lung cancer were randomly divided into Qihuang acupuncture group and control group， with 33 cases in each group. The control group was treated with long-acting opioid analgesics at maintenance doses and supplementary analgesic medications as needed. In case of breakthrough pain， short-acting opioids were used for rescue. The Qihuang acupuncture group received Qihuang acupuncture treatment in addition to the treatment used in the control group， administered once every other day， with 3 sessions constituting one treatment course. The treatment duration for both groups was 5 days. The primary outcome was the change in pain intensity， measured using the numerical rating scale （NRS） before and after treatment， and the NRS change rate was calculated. Secondary endpoints included the daily NRS change rate， the Eastern Cooperative Oncology Group （ECOG） Performance Status （PS） score， the European Organization for Research and Treatment of Cancer Quality of Life Core Questionnaire （EORTC QLQ-C30） score， and the 24-hour equivalent hydrocodone sustained-release tablet dose. Laboratory tests， including routine blood， urine， stool， liver function， and kidney function， were performed before and after treatment. Adverse events were recorded throughout the trial. ResultsAll patients completed the trial， and both groups showed a decrease in average NRS scores and PS scores after treatment， with the Qihuang acupuncture group showing lower average NRS scores and PS scores than the control group （P＜0.05 or P＜0.01）. After treatment， the NRS change rate in the Qihuang acupuncture group was （0.42±0.17）， significantly higher than that in the control group （0.14±0.27， P＜0.01）. The daily NRS change rate during treatment was also higher in the Qihuang acupuncture group compared to the control group （P＜0.01）. The Qihuang acupuncture group showed an increase in overall health status and functional scores in the EORTC QLQ-C30， and a decrease in symptom scores for fatigue， nausea and vomiting， pain， dyspnea， insomnia， appetite loss， constipation， and financial difficulties. In contrast， overall health status and constipation scores in the control group increased， while scores of fatigue， nausea and vomiting， pain， and appetite loss decreased （P＜0.05 or P＜0.01）. After treatment， the 24-hour equivalent hydrocodone sustained-release tablet dose did not show significant difference in the Qihuang acupuncture group （P＞0.05）， while the control group showed a significant increase in the 24-hour dose （P＜0.01）. No significant abnormalities were observed in laboratory tests before and after treatment in either group. During the study， the incidence of nausea and vomiting as well as constipation in the Qihuang acupuncture group was both 3.03% （1/33）， while the incidence in the control group was 27.27% （9/33） and 36.36% （12/33）， respectively， with the Qihuang acupuncture group showing significantly lower incidence （P＜0.01）. No serious adverse reactions were observed in either group. ConclusionQihuang acupuncture therapy combined with opioid analgesics is more effective than using opioids alone in relieving pain in patients with moderate to severe cancer pain due to lung cancer. It can improve the patients' physical condition and quality of life， reduce the dose of opioid analgesics， and has good safety.

Listeria brainstem encephalitis with myelitis is extremely rare in clinical practice.Since the clinical manifestations are non-specific,MRI is helpful for diagnosis.Positive cerebrospinal fluid culture is considered the gold standard for diagnosis.This article reports a case of an immunocompetent individual with listeria brainstem encephalitis with myelitis,aiming to enhance the awareness of this condition.
Humans ; Brain Stem/pathology* ; Diagnostic Errors ; Encephalitis/complications* ; Encephalomyelitis, Acute Disseminated/diagnosis* ; Listeriosis/complications* ; Myelitis/complications*

To explore the value of the diaphragm thickness fraction (TF) and diaphragm mobility (DM) measured by ultrasound for predicting ventilator withdrawal success in patients with acute respiratory distress syndrome (ARDS) after cardiac surgery. Methods: This study included 246 patients undergoing the spontaneous breathing trial. Diaphragmatic function was evaluated by ultrasound, including the diaphragm thickness at the end of calm breathing (thickness of the diaphragm at functional residual capacity [TdiFRC]) and the maximum diaphragm thickness at the end of inspiration (thickness of the diaphragm at full vital capacity [TdiFVC]); TF=(TdiFVC–TdiFRC)/TdiFRC×100%. DM, the oxygenation index (the ratio of the partial pressure of arterial oxygen to the fraction of inspired oxygen), and the rapid shallow breathing index (RSBI) were measured. Results: Successful liberation from mechanical ventilation was observed in 209 patients. There were no significant differences in the TdiFRC (0.3±0.1 cm vs. 0.3±0.1 cm) or TdiFVC (0.3±0.1 cm vs. 0.2±0.1 cm) between the ventilator withdrawal success group and the ventilator withdrawal failure group (P>0.05). The TF was greater in the ventilator withdrawal success group than in the ventilator withdrawal failure group (40.8%±15.8% vs. 37.7%±9.2%, P<0.01). DM in the ventilator withdrawal success group was greater than that in the ventilator withdrawal failure group (1.5±0.5 cm vs. 1.2±0.4 cm, P=0.040). The RSBI was lower in the ventilator withdrawal success group than in the ventilator withdrawal failure group (74.3±25.6 breaths·min–1·L –1 vs. 89.9±34.5 breaths·min–1·L –1, P<0.01). Conclusions: Diaphragmatic ultrasound can be used to predict the success of ventilator withdrawal in patients with ARDS.

Objective: To summarize the treatment methods and efficacy of a patient with pure red cell aplasia (PRCA) after allogeneic hematopoietic stem cell transplantation (allo-HSCT), and to accumulate relevant case data. Methods: The clinical treatment and laboratory test data of a patient with PRCA after allo-HSCT in our hospital were retrospectively collected. The therapeutic strategy, monitoring parameters, and treatment outcomes were summarized. Results: Upon suspicion of post-transplant PRCA, the patient was promptly treated with intravenous injection of human immunoglobulin and three sessions of plasma exchange. The titer of blood group antibodies in the patient decreased, and the hemolytic symptoms were relieved. Over one year post-transplantation, the patient exhibited a sustained impairment of erythropoiesis, necessitating continued red blood cell transfusions. After treatment with intravenous daratumumab (400 mg twice weekly for 4 weeks), the pateint's hemoglobin (Hb) and reticulocyte (Ret) levels normalized rapidly, the ABO blood type converted from the recipient to the donor type, and the titer of IgM blood group antibodies returned to normal. The patient was successfully weaned off red blood cell transfusions, indicating the clinical efficacy of the treatment. Conclusion: Daratumumab shows effectiveness in the treatment of refractory PRCA after allo-HSCT in the case. It is essential to monitor Hb, Ret and the titer of blood group antibodies during treatment. Nevertheless, the interference of daratumumab with the titer of blood group antibodies should be considered.

Objective:To evaluate corrective intra-articular osteotomy in the treatment of malreduced tibial pilon fractures in young patients.Methods:A retrospective study was conducted to analyze the 23 patients who had been treated for malreduced tibial pilon fractures by corrective intra-articular osteotomy from January 2013 to December 2022 at Department of Orthopedics, Orthopedic Research Institute, West China Hospital. They were 21 males and 2 females with a median age of 42.9(28.6, 48.1) years. The median interval between intra-articular osteotomy and initial injury or previous surgery was 97 (51, 166) d. The clinical efficacy was assessed by the ankle-hindfoot score of American Orthopaedic Foot and Ankle Society (AOFAS), MOS item short form health survey (SF-36), visual analog scale (VAS) pain score, ankle range of motion (ROM), modified Kellgren-Lawrence arthritis grading, and complications.Results:The median follow-up period was 35.4 (26.2, 72.2) months. At the last follow-up, the AOFAS ankle-hindfoot score, SF-36, VAS pain score, and ankle ROM for all patients were, respectively, (80.7±12.1) points, (72.2±13.9) points, 2.0(0, 3.0) points and 23.0°(18.0°, 30.0°), and the modified Kellgren-Lawrence arthritis grading revealed stage Ⅱ in 17 cases, stage Ⅲ in 5 cases, and stage Ⅳ in 1 case. For the 12 patients with sufficient preoperative data, the AOFAS ankle-hindfoot score [(80.9±10.1) points], SF-36 [(72.9±12.0) points], VAS pain score [2.5 (0.5, 3.0) points], and ankle ROM (26.5°±7.9°) at the last follow-up were significantly better than the preoperative values [(42.2±16.0) points, (36.0±8.8) points, 6.0 (5.5, 6.5) points, and 21.3°±7.0°] (all P<0.05). One patient who experienced significant pain postoperatively planned to receive ankle fusion surgery. No patients experienced deep infection, fracture non-union, or failure of internal fixation. Conclusion:Corrective intra-articular osteotomy may be a viable alternative treatment of malreduced tibial pilon fractures in young patients, because it can effectively improve short-term clinical and radiographic outcomes.

Objective:To explore the value of nomogram model based on multimodal ultrasound features for predicting the malignant risk of micro lesions in breast areola region.Methods:The case data of Beijing Tiantan Hospital affiliated to Capital Medical University from May 2020 to July 2024 were retrospectively analyzed. A total of 50 patients with benign intraductal papilloma（bIDP group）and 54 patients with malignant risk breast tumor（mrBT group）were found to have micro lesions in breast areola region and confirmed by puncture or surgical pathology. Clinical data，conventional ultrasound and contrast-enhanced ultrasound features were compared between the two groups. Multivariate Logistic regression analysis and Lasso regression analysis were performed on statistically significant factors to screen out influencing factors. ROC curves were plotted to evaluate diagnostic efficacy，nomogram model and clinical decision curves were constructed to evaluate clinical benefits.Results:The differences of age，nipple discharge presentation，conventional ultrasound features（including boundary，morphology，aspect ratio，internal echo，internal microcalcification，far-field echo，peripheral irregular hyperechoic ring，dilate of peripheral ducts），and contrast-enhanced ultrasound features（including wash-in time，enhancement intensity，enhancement mode，enhancement scope，blood perfusion defect，crab foot sign，penetrating vessels）were statistically significant between the bIDP group and mrBTgroup（all P<0.05）. Regression analysis showed that age，uniformity of internal echo within the lesion，dilation of surrounding ducts，and enhanced crab foot sign were the affect factors for the diagnosis of mrBT（all P<0.05）. Based on these factors，a nomogram model was constructed with an area under ROC curve（AUC）of 0.907（95% CI=0.851-0.963），a sensitivity of 0.907，and a specificity of 0.780. The decision curve analysis showed that the collective model had good predictive performance. Conclusions:The nomogram model based on multimodal ultrasound features has good value in predicting malignant risk micro breast tumor of areola region.

Objective:To develop a deep learning model based on super-resolution endorectal ultrasound（ERUS）images for the preoperative prediction of perineural invasion（PNI）in patients with rectal cancer，thereby providing a reference for risk stratification and individualized treatment planning.Methods:A retrospective analysis was conducted on 382 patients with rectal cancer who underwent total mesorectal excision at Fujian Medical University Union Hospital between June 2019 and February 2024. Patients were randomly divided into a training set（ n=305）and a test set（ n=77）at a ratio of 8∶2，and further grouped into PNI-negative group and PNI-positive group subgroups based on pathological results. Super-resolution ultrasound images were generated from original ERUS images using a generative adversarial network（GAN）. Deep convolutional neural networks were developed based on features from intratumoral and peritumoral regions to identify the optimal region of interest（ROI）. The dSR5_ResNet18 and dSR5_ResNet50 models were constructed using the super-resolution images with a 5-pixel peritumoral extension. Representative clinical features were selected for subgroup analysis based on sample size and intergroup statistical differences between PNI-positive and PNI-negative patients. Forest plots were used to evaluate model applicability and robustness across subgroups. Results:The dSR5_ResNet18 model，built using super-resolution images of the tumor combined with a 5-pixel peritumoral region，achieved the best predictive performance，with an AUC of 0.867（95% CI=0.782 - 0.952）in the test set. Decision curve analysis demonstrated that the dSR5_ResNet18 model provided the greatest net clinical benefit. Forest plot analysis indicated strong generalizability of the models across subgroups such as pathological N stage，maximum lesion length，and lymph node enlargement，though relatively weaker performance was observed in the carcinoembryonic antigen（CEA）subgroup. Among all models，dSR5_ResNet18 exhibited the most consistent performance across subgroups，with the narrowest confidence intervals and highest robustness. Conclusions:The deep learning model incorporating ERUS-based super-resolution reconstruction demonstrated excellent performance in the preoperative prediction of PNI in rectal cancer. It offers significant advantages in image quality and generalizability，and may serve as a valuable tool to assist clinicians in formulating personalized treatment strategies.

Objective:To establish a model based on the characteristics of breast cancer ultrasound images through deep learning methods to predict the risk of axillary lymph node metastasis（ALNM）in patients with breast cancer（maximum diameter ≤2.5 cm）before surgery.Methods:A total of 419 patients（3 433 breast tumor ultrasound images）with breast cancer（maximum diameter ≤2.5 cm）who underwent axillary lymph node dissection at Beijing Tiantan Hospital，Capital Medical University from January 2019 to December 2024 were retrospectively included. According to the pathological results of axillary lymph nodes，they were divided into 220 cases in the ALNM occurrence group（positive group）and 199 cases in the non-ALNM occurrence group（negative group）. The breast cancer ultrasound images of the two groups of cases were randomly classified into the training set（2 404 images），the validation set（687 images）and the test set（342 images）according to a ratio of 7∶2∶1. YOLOv8 was used as the basic model of You Only Look Once（YOLO）and optimized. The optimized model was applied to locate and capture the potential ultrasound features of breast cancer cases in the training set. A prediction model was constructed based on the captured ultrasound features. The model was adjusted and optimized through the validation set，and then matched with the case images in the test set. The confusion classification matrix graph and the curve graph for measuring the model performance were used to evaluate the model prediction performance and interpret the model，and the efficacy of this model in identifying breast cancer patients at risk of ALNM was analyzed.Results:There were statistically significant differences between the positive and negative groups in terms of the pathological maximum diameter of breast tumors，pathological T staging，the differentiation degree，the presence of distant metastasis，the maximum diameter measured by ultrasound，the quadrant of breast tumor occurrence，the Breast Imaging - Reporting and Data System（BI-RADS）classification of breast tumors，and the presence of abnormal ultrasound features of lymph node（all P<0.05）. The established deep learning model could automatically perform bounding box localization for the breast cancer of patients.The breast tumors in the positive group had potential ultrasound features that could be captured by the model compared with those in the negative group. The mean average precision（mAP）50 was 0.883，mAP 50-95 was 0.636，PR-AUC was 0.884 5，strict PR-AUC was 0.636 4，the sensitivity was 90.5%，and the specificity was 91.2%，and it had a good predictive efficacy. Conclusions:This prediction model based on the ultrasound characteristics of breast cancer through deep learning can effectively predict breast cancer（maximum diameter ≤ 2.5 cm）with the risk of ALNM，providing an effective basis for the clinical management of axillary lymph nodes in breast cancer patients.

Aim To investigate the effect of long-chain non-coding RNA(lncRNA)GS1-124K5.4 targeting regulation of PRDX6 on proliferation,migration and in-vasion of lung squamous carcinoma(LUSC)cells and the underlying mechanism.Methods The expression level of lncRNA GS1-124K5.4 in lung cancer tissues and adjacent tissues of 60 patients with LUSC were de-termined by fluorescence in situ hybridization.The ex-pression level of lncRNA GS1-124K5.4 in human nor-mal lung cells and LUSC cells were determined by qRT-PCR.Two kinds of LUSC cells(NCI-H 1703,SK-MES-1)with highest expression level of lncRNA GS1-124K5.4 were selected for subsequent experi-ments.The distribution of lncRNA GS1-124K5.4 in cells was studied by fluorescence in situ hybridization and prokaryotic separation.The effect of knockdown of lncRNA GS1-124K5.4 on proliferation of NCI-H1703 and SK-MES-1 cells was studied by CCK-8 experiment and cell clone formation experiment;the effect of knockdown of lncRNA GS1-124K5.4 on migration of NCI-H1703 and SK-MES-1 cells was studied by cell scratch experiment and Transwell cell migration experi-ment;and the effect of knockdown of lncRNA GS1-124K5.4 on invasion of NCI-H1703 and SK-MES-1 cells was studied by Transwell invasion experiment.The protein to be bound by lncRNA GS1-124K5.4 was detected by RNA pull-down combined with mass spec-trometry and immune-precipitation.The effect of knockdown of lncRNA GS1-124K5.4 targeting PRDX6 on proliferation,migration and invasion of NCI-H1703 and SK-MES-1 cells was studied.Results(1)The fluorescence intensity of lncRNA GS1-124K5.4 in lung squamous cell carcinoma increased compared with that in adjacent tissues(P＜0.05),and the expression of lncRNA GS1-124K5.4 was related with lymph node metastasis and clinical stage(P＜0.05).(2)The ex-pression level of lncRNA GS1-124K5.4 in NCI-H1703,NCI-H520 and SK-MES-1 cells significantly increased(P＜0.05).(3)The result of fluorescence in situ hybridization experiment and nucleoplasm sepa-ration experiment showed that lncRNA GS1-124K5.4 was mainly distributed in cell nucleus.(4)The prolif-eration,migration and invasion ability of NCI-H1703 and SK-MES-1 cells with knockdown of lncRNA GS1-124K5.4 significantly decreased(P＜0.05).(5)PRDX6 protein to be bound to LncRNA GS1-124K5.4 was determined by RNA pull-down combined with mass spectrometry and immunoprecipitation.(6)The prolif-eration,migration and invasion ability of NCI-H1703 and SK-MES-1 cells with overexpression of lncRNA GS1-124K5.4 significantly increased(P＜0.05);the proliferation,migration and invasion ability of NCI-H1703 and SK-MES-1 cells with knockdown of PRDX6 significantly decreased(P＜0.05);the proliferation,migration and invasion ability of NCI-H1703 and SK-MES-1 cells with overexpression of lncRNAGS1-124K5.4 and knockdown of PRDX6 showed no signifi-cant change(P＞0.05).Conclusions LncRNA GS1-124K5.4 is highly expressed in lung squamous cell carcinoma,and it may promote the proliferation,migration and invasion of lung squamous carcinoma cells by targeting the expression of PRDX6 protein.