OBJECTIVE:Based on different algorithms of machine learning,the prediction model of lumbar disc herniation has become a trend and hot spot in the development of precision medicine.However,there is limited evidence on the reporting quality and methodological quality of prediction models of lumbar disc herniation outcomes using machine learning.This article is aimed to explore the performance of machine learning algorithms in predicting the prognosis of lumbar disc herniation by comprehensively analyzing the report quality and risk of bias of previous studies that developed and validated prognosis prediction models based on machine learning through a comprehensive literature search,in order to explore the performance of machine learning algorithms in predicting the prognosis of lumbar disc herniation.METHODS:The databases of CNKI,WanFang,VIP,SinOMED,PubMed,Web of Science,Embase,and The Cochrane Library were searched by computer.Studies on the use of machine learning to develop(and/or validate)prognostic prediction models for lumbar disc herniation were collected from the inception of the database to December 31,2023.Two researchers independently screened the literature,extracted data,and assessed the risk of bias of the included studies.The reporting quality and risk of bias of the included studies were assessed by the Multivariable Transparent Reporting of Predictive Models(TRIPOD)statement and the Predictive Model Risk of Bias Assessment Tool(PROBAST).The results of the evaluation were analyzed using descriptive statistics and visual charts.RESULTS:(1)A total of 23 articles were included,and the TRIPOD compliance of each study ranged from 11％to 87％,with a median compliance of 54％.The quality of reporting of titles,detailed descriptions of treatment measures,blinding of predictors,handling of missing data,details of risk stratification,specific procedures for enrollment,model interpretation,and model performance was mostly poor,with TRIPOD adherence rates ranging from 4％to 35％.(2)Of all included studies,61％had a high risk of bias and 39％had an unclear overall risk of bias.The area under the curve,accuracy,sensitivity and specificity were used to evaluate the performance of the model.The areas under the curve of 20 models were reported,ranging from 0.561 to 0.999.Three models reported the accuracy of the model,ranging from 82.07％to 89.65％.(3)Among all included studies,the statistical analysis domain was most often assessed as having a high risk of bias,mainly due to the small number of valid samples,the selection of predictors based on univariate analysis and the lack of calibration and discrimination assessment of the model in the study.CONCLUSION:These results indicate that machine learning can achieve good predictive ability in the development and validation of prognostic models for lumbar disc herniation.The commonly used algorithms include regression algorithm,support vector machine,decision tree,random forest,artificial neural network,naive Bayes and other algorithms.Reasonable algorithms combined with clinical practice can improve the accuracy of prognosis prediction of lumbar disc herniation.However,the reporting and methodological quality of prognosis prediction models based on machine learning are poor,the prediction performance of different models varies greatly,and the generalization and extrapolation of research models are unclear.There is an urgent need to improve the design,implementation and reporting of such studies.To promote the application of machine learning in the clinical practice of lumbar disc herniation prediction models,it is necessary to comprehensively consider various predictors related to the prognosis of the disease before modeling,and strictly follow the relevant standards of PROBAST tool during modeling.

Objective: To investigate the impact of RhE antigen-matched transfusion during liver transplantation on early postoperative recovery and complications. Methods: In this retrospective cohort study, ninety-five patients undergoing liver transplantation at Kunming First People's Hospital between January 2022 and July 2025 were enrolled. Patients were divided into two groups: Group 1 (RhE-mismatched transfusion, n=57) and Group 2 (RhE-matched transfusion, n=38). The baseline data, complete blood counts, hepatic and renal function, coagulation parameters, and complication rates between the two groups were compared at postoperative days 1, 3, 5, 7, and 10. Survival analysis was performed using the Kaplan-Meier method. Results: The baseline characteristics were well-balanced and comparable between the two groups (all P>0.05). The early postoperative mortality rate in the mismatched group (31.58%, 18/57) was significantly higher than that in the matched group (10.53%, 4/38) (P=0.017). The incidence of postoperative hepatic encephalopathy was significantly higher in the mismatched group (50.88%, 29/57) than in the matched group (10.53%, 4/38) (P<0.001). The incidence of postoperative haemorrhage in the mismatched group (24.56%, 14/57) was higher than that in the matched group (5.26%, 2/38), with a statistically significant difference (P=0.014). The incidence of perioperative infection in the mismatched group (28.07%, 16/57) was higher than that in the matched group (10.53%, 4/38), with a statistically significant difference (P=0.04). Corresponding odds ratios (OR) and 95% confidence intervals indicated a lower risk of these adverse events in the matched group. On postoperative day 1, the change in activated partial thromboplastin time (-1.6, 20.5) in the mismatched group was greater than in the matched group (-0.2, 5.5). The change in international normalised ratio (-0.56, 1.22) in the mismatched group was greater than in the matched group (-0.18, 0.32), while the change in albumin (-4.0, 4.8) was smaller in the mismatched group than in the matched group (-2.5, 8.8). On postoperative day 5, the change in albumin (-0.41±7.83) in the mismatched group was smaller than in the matched group (2.68±4.53). At postoperative day 7, the change in albumin in the mismatched group (-0.61±7.38) was smaller than that in the matched group (2.51±5.85), while the change in D-dimer in the mismatched group (0.73, 7.4) was greater than that in the matched group (-1.6, 4.3). On postoperative day 10, the mismatched group exhibited significantly higher fibrinogen levels (-1.21, 1.78) than the matched group (-0.49, 0.97), and significantly longer prothrombin times (-11.3, -2.7) than the matched group (-6.2, -0.8) (all P<0.05). The matched group exhibited a mean overall survival (OS) of 32.803 months (95% CI:29.171-36.436 months), significantly exceeding the mismatched group's 28.996 months (95% CI:24.202-33.790 months). The log-rank test yielded statistically significant results (χ =4.307, P=0.038). Conclusion: Implementing RhE blood group-matched transfusion during liver transplantation may help reduce early postoperative mortality and the incidence of major complication rates, promote faster recovery of coagulation and liver function, and thereby improve short-term patient outcomes.

Bone metastasis is one of the common complications of lung cancer, which seriously affects the quality of life and survival of patients. At present, the clinical diagnosis of bone metastasis of lung cancer mainly depends on imaging methods, but due to its lack of sensitivity and potential radiation risk, about half of patients have already had bone-related events when they are diagnosed clearly. The treatment of bone metastasis of lung cancer mainly depends on surgery, radiotherapy and chemotherapy, targeted therapy, immunotherapy, etc. Although the treatment of bone metastasis of lung cancer has made some progress in recent years, there are still some problems such as high risk of other distant metastasis. This article mainly reviews the pathogenesis, diagnostic biomarkers and treatment progress of bone metastasis of lung cancer, in order to provide reference for the diagnosis and treatment of bone metastasis of lung cancer.

ObjectiveTo investigate the imaging features of calcium salt deposition and serological markers in patients with alveolar echinococcosis through a retrospective analysis， as well as independent risk factors for the degree of calcium salt deposition in lesions， and to provide a basis for assessing disease process. MethodsA retrospective analysis was performed for the imaging and clinical data of 107 patients with alveolar echinococcosis who were admitted to The First Affiliated Hospital of Shihezi University from December 2023 to June 2025， and according to the volume of calcium salt deposition， they were divided into non-deposition group with 16 patients， mild deposition group with 52 patients， moderate deposition group with 16 patients， and severe deposition group with 23 patients. A one-way analysis of variance or the Kruskal-Wallis H test was used for comparison of continuous data between groups， and the χ² test or Fisher’s exact test was used for comparison of categorical data between groups. The four groups were further combined into the low deposition group （no/mild deposition） and the high deposition group （moderate/severe deposition）. A binary logistic regression analysis was used to investigate the independent influencing factors for calcium salt deposition， and a predictive model was established. The receiver operating characteristic （ROC） curve was used to assess the predictive performance of the model， and the Bootstrap method was used for internal validation. ResultsThere were significant differences between the four groups in sex distribution， involvement of other sites， white blood cell count， lymphocyte percentage， fibrinogen， uric acid， sodium ion， chloride ion， and calcium ion （all P<0.05）. The univariate analysis showed that there were significant differences between the four groups in sex， involvement of other sites， white blood cell count， lymphocyte percentage， fibrinogen， alanine aminotransferase， albumin， creatinine， uric acid， sodium ion， chloride ion， and calcium ion （all P<0.1）. The multi-collinearity diagnosis showed that the VIF values for all continuous variables ranged from 1.104 to 1.760， suggesting that collinearity did not affect modeling. An ordinal logistic regression model was established based on sex， involvement of other sites， calcium ion， lymphocyte percentage， and uric acid. The multivariate analysis showed that lymphocyte percentage （odds ratio ［OR］=1.106， 95% confidence interval ［CI］： 1.041 — 1.174， P=0.001） and blood calcium level （OR=0.005， 95%CI： 0.000 —0.230， P=0.007） were independent influencing factors for the degree of calcium salt deposition. The regression equation was established as Logit（P）=8.231 + 0.100 × lymphocyte percentage -5.344 × calcium ion. The ROC curve analysis showed that the model had an area under the ROC curve of 0.716， with a Youden index of 0.353， a sensitivity of 1.000， and a specificity of 0.353. The Hosmer-Lemeshow test showed that the model had poor calibration （χ²=20.688， P=0.008）. The Bootstrap method with 1000 repeated samples showed that the estimated values of lymphocyte percentage （OR=1.106， 95%CI： 1.049 — 1.186， P=0.002） and calcium ion （OR=0.005， 95%CI： 0.000 — 0.214， P=0.010） were consistent with the original model， and the confidence intervals did not include 1， which further supported the reliability of the model. ConclusionBoth lymphocyte percentage and blood calcium level are independent influencing factors for calcium salt deposition in alveolar echinococcosis， and the degree of calcium salt deposition in alveolar echinococcosis lesions increases with the reduction in blood calcium level and the increase in lymphocyte percentage.

OBJECTIVE: To explore the prenatal and postnatal phenotypes of 22q11.2 microdeletion syndrome (22q11.2DS) and enhance clinical understanding of this condition. METHODS: Data were collected from 86 fetuses diagnosed with 22q11.2DS at four prenatal diagnostic centers across China between January 2014 and August 2025. Prenatal imaging findings, pregnancy outcomes, and postnatal conditions were analyzed. RESULTS: Among the 86 fetuses, complete ultrasound data were available for 65 cases. Cardiovascular abnormalities were observed in 42 cases, thymic hypoplasia or aplasia in 7 cases, urinary system anomalies in 6 cases, nuchal translucency (NT) thickening in 7 cases, butterfly vertebrae, clubfoot, omphalocele and diaphragmatic hernia in 1 case each, cleft lip and palate in 2 cases, and ultrasound soft markers in 13 cases. The parents of 9 fetuses opted to continue with the pregnancy. Among these, 6 showed no significant ultrasound abnormalities and no related phenotypes postnatally, while the remaining 3 exhibited ultrasound anomalies with postnatal manifestations including developmental delay, immunodeficiency, and cardiac defects. CONCLUSION Fetuses with 22q11.2DS may exhibit various ultrasound abnormalities in multiple systems before and after birth. In addition to cardiovascular anomalies, they may also present with thymic hypoplasia or aplasia, thickened NT, and urinary abnormalities. Fetuses with thickened NT or thymic anomalies should be closely monitored, and thymic assessment should be included in routine prenatal imaging evaluations. For fetuses with 22q11.2DS who show no ultrasound abnormalities, the risk of developing severe phenotypes after birth is relatively low, but occult palate clefts and psychiatric disorders cannot be ruled out. Due to limitations in sample size and follow-up duration, above conclusions require further validation through large-scale prospective studies.
Humans ; Female ; Pregnancy ; Ultrasonography, Prenatal ; DiGeorge Syndrome/genetics* ; Adult ; Male ; Follow-Up Studies ; Fetus/diagnostic imaging* ; Phenotype ; Infant, Newborn

Prostate cancer is one of the most common malignant tumors of male genitourinary system. Prostate cancer has the following characteristics： insidious onset， early asymptomatic or not obvious symptoms， complex etiology and pathogenesis， long incubation period and so on. Therefore， the realization of its early diagnosis and treatment is of great significance to the prognosis of patients. Prostate-specific membrane antigen （PSMA） is a type 2 transmembrane glycoprotein that is highly expressed on the membrane of almost all primary and metastatic prostate cancer cells， and is an ideal target for prostate cancer imaging and treatment. In recent years， with the approval of urea-containing small molecule PET （positron emission computed tomography） radiopharmaceutical based on PSMA （⁶⁸Ga-PSMA-11， ¹⁸F-PSMA-1007）， PET-CT （positron emission computed tomography/computed tomography） has shown new potential for early diagnosis and accurate staging of prostate cancer patients. This review mainly summarizes the research progress of urea-containing PSMA PET imaging agents and finds that they have defects such as uptake in non-target tissues like the kidneys， lacrimal glands， and salivary glands. Thus， further optimizing their structure to reduce the uptake in non-target tissues， providing provide convenience for the labeling of therapeutic radiopharmaceuticals， thereby achieving the goal of integrated diagnosis and treatment， is an important development direction in this field.

OBJECTIVE:Based on different algorithms of machine learning,the prediction model of lumbar disc herniation has become a trend and hot spot in the development of precision medicine.However,there is limited evidence on the reporting quality and methodological quality of prediction models of lumbar disc herniation outcomes using machine learning.This article is aimed to explore the performance of machine learning algorithms in predicting the prognosis of lumbar disc herniation by comprehensively analyzing the report quality and risk of bias of previous studies that developed and validated prognosis prediction models based on machine learning through a comprehensive literature search,in order to explore the performance of machine learning algorithms in predicting the prognosis of lumbar disc herniation.METHODS:The databases of CNKI,WanFang,VIP,SinOMED,PubMed,Web of Science,Embase,and The Cochrane Library were searched by computer.Studies on the use of machine learning to develop(and/or validate)prognostic prediction models for lumbar disc herniation were collected from the inception of the database to December 31,2023.Two researchers independently screened the literature,extracted data,and assessed the risk of bias of the included studies.The reporting quality and risk of bias of the included studies were assessed by the Multivariable Transparent Reporting of Predictive Models(TRIPOD)statement and the Predictive Model Risk of Bias Assessment Tool(PROBAST).The results of the evaluation were analyzed using descriptive statistics and visual charts.RESULTS:(1)A total of 23 articles were included,and the TRIPOD compliance of each study ranged from 11％to 87％,with a median compliance of 54％.The quality of reporting of titles,detailed descriptions of treatment measures,blinding of predictors,handling of missing data,details of risk stratification,specific procedures for enrollment,model interpretation,and model performance was mostly poor,with TRIPOD adherence rates ranging from 4％to 35％.(2)Of all included studies,61％had a high risk of bias and 39％had an unclear overall risk of bias.The area under the curve,accuracy,sensitivity and specificity were used to evaluate the performance of the model.The areas under the curve of 20 models were reported,ranging from 0.561 to 0.999.Three models reported the accuracy of the model,ranging from 82.07％to 89.65％.(3)Among all included studies,the statistical analysis domain was most often assessed as having a high risk of bias,mainly due to the small number of valid samples,the selection of predictors based on univariate analysis and the lack of calibration and discrimination assessment of the model in the study.CONCLUSION:These results indicate that machine learning can achieve good predictive ability in the development and validation of prognostic models for lumbar disc herniation.The commonly used algorithms include regression algorithm,support vector machine,decision tree,random forest,artificial neural network,naive Bayes and other algorithms.Reasonable algorithms combined with clinical practice can improve the accuracy of prognosis prediction of lumbar disc herniation.However,the reporting and methodological quality of prognosis prediction models based on machine learning are poor,the prediction performance of different models varies greatly,and the generalization and extrapolation of research models are unclear.There is an urgent need to improve the design,implementation and reporting of such studies.To promote the application of machine learning in the clinical practice of lumbar disc herniation prediction models,it is necessary to comprehensively consider various predictors related to the prognosis of the disease before modeling,and strictly follow the relevant standards of PROBAST tool during modeling.

As the center of energy metabolism and a key hub of intracellular signal transduction in platelets, platelet mitochondria play a core role in maintaining platelet activation, apoptosis, and functional homeostasis. Mitochondria not only provide approximately 90% of platelet ATP through oxidative phosphorylation (OXPHOS), but also deeply participate in platelet activation, secretion, aggregation, and apoptosis-like programmed death by regulating reactive oxygen species (ROS) generation, mitochondrial membrane potential (ΔΨm), and the opening of the mitochondrial permeability transition pore (mPTP). During platelet storage, mitochondrial dysfunction is a core mechanism underlying platelet storage lesions. Therefore, targeting mitochondrial function protection—such as through supplementation of metabolic substrates, application of mitochondria-targeted antioxidants, or mPTP inhibitors—holds promise as a novel intervention strategy to ameliorate platelet storage lesions, prolong platelet survival in vivo, and reduce adverse transfusion reactions.

Background/Aims: Transmembrane 4 L six family member 1 (TM4SF1) is highly expressed and contributes to the progression of various malignancies. However, how it modulates hepatocellular carcinoma (HCC) progression and senescence remains to be elucidated. Methods: TM4SF1 expression in HCC samples was evaluated using immunohistochemistry and flow cytometry. Cellular senescence was assessed through SA-β-gal activity assays and Western blot analysis. TM4SF1-related protein interactions were investigated using immunoprecipitation-mass spectrometry, co-immunoprecipitation, bimolecular fluorescence complementation, and immunofluorescence. Tumor-infiltrating immune cells were analyzed by flow cytometry. The HCC mouse model was established via hydrodynamic tail vein injection. Results: TM4SF1 was highly expressed in human HCC samples and murine models. Knockdown of TM4SF1 suppressed HCC proliferation both in vitro and in vivo, inducing non-secretory senescence through upregulation of p16 and p21. TM4SF1 enhanced the interaction between AKT1 and PDPK1, thereby promoting AKT phosphorylation, which subsequently downregulated p16 and p21. Meanwhile, TM4SF1-mediated AKT phosphorylation enhanced PD-L1 expression while reducing major histocompatibility complex class I level on tumor cells, leading to impaired cytotoxic function of CD8+ T cells and an increased proportion of exhausted CD8+ T cells. In clinical HCC samples, elevated TM4SF1 expression was associated with resistance to anti-PD-1 immunotherapy. Targeting TM4SF1 via adeno-associated virus induced tumor senescence, reduced tumor burden and synergistically enhanced the efficacy of anti-PD-1 therapy. Conclusions Our results revealed that TM4SF1 regulated tumor cell senescence and immune evasion through the AKT pathway, highlighting its potential as a therapeutic target in HCC, particularly in combination with first-line immunotherapy.

Objective To establish a rat model of venous thrombosis in a plateau hypobaric hypoxic environment and to investigate the effect of this environment on venous thrombosis.Methods A total of 144 healthy male SD rats were assigned randomly to four groups(n=36 rats per group):a plains sham operation(A)group,plains operation(B)group,plateau altitude 6000 m+sham operation(C)group,and plateau altitude 6000 m+surgery(D)group.Rats in A and B groups were maintained in a plains normoxic environment,while rats in C and D groups C and D were subjected to a plateau environment.Rats in the surgical groups underwent quantitative constriction to incompletely obstruct the inferior vena cava blood flow.Each group was further divided into subgroups based on time:1,3,5,7,14,and 21 d(n=6 rats per group).Regular vascular ultrasound monitoring was conducted,and blood samples were taken for whole blood viscosity testing and the assessment of inflammatory indicators,including endothelin-1(ET-1),interleukin-6(IL-6)and tissue factor(TF).Coagulation function was evaluated through the activated partial thromboplastin time(APTT),prothrombin time(PT),thrombin time(TT),fibrinogen(FIB)and D-dimer.After the observation period,the experimental animals were sacrificed and the limbs were removed.Thrombus samples were stained with hematoxylin/eosin(HE),and the thrombus wet mass was measured.Results The thrombosis incidence was significantly higher in the plateau D group than in B group,accompanied by a marked increase in blood viscosity and hematocrit(P＜0.01).Additionally,levels of ET-1,IL-6,and TF were significantly elevated(P＜0.05),indicating a coagulation disorder.Conclusions A plateau hypoxic environment model can be successfully simulated by quantitative coarctation of the inferior vena cava,combined with a specialized environmental chamber.The findings of this study suggest that a plateau hypoxic environment promotes venous thrombosis.