Objective To develop a risk assessment scale for immune checkpoint inhibitor (ICI)-associated myocarditis based on multidisciplinary collaboration, and to evaluate its diagnostic performance. Methods Based on multidisciplinary cooperation, integrating clinical experience from oncology and cardiology, literature data, and patient conditions, a risk assessment scale for ICI-associated myocarditis was developed. A total of 101 patients with malignancies who received immunotherapy at Zhongshan Hospital, Fudan University, from October 2020 to October 2024 were included as the validation cohort. Patients were stratified into low-risk (0-1 point), medium-risk (2-4 points), and high-risk (≥5 points) groups based on their scale scores. The association between pretictive risk stratifications and actual assessment results was assessed using the Cox proportional hazards regression model. The predictive value of the scale for ICI-associated myocarditis was evaluated using receiver operating characteristic (ROC) curve. Agreement between the scale scores and actual assessment results was assessed using Cohen’s Kappa coefficient. Results Based on the scale pretictive results, 28(27.7%), 8(7.9%), 65(64.4%) patients were at low risk, medium risk, and high risk for ICI-related myocarditis, respectively; however, 46(45.5%), 8(7.9%), 47(46.5%) were at low risk, medium risk, and high risk actually. Kaplan-Meier survival analysis showed that the cumulative incidence of ICI-related myocarditis in the high-risk group was significantly higher than that in the medium- and low-risk groups (P＜0.05). In the multivariable-adjusted Cox proportional hazards model, the ICI-related myocarditis risk in high-risk group was about 4 times that in the low-risk group. ROC curve analysis demonstrated that the average area under the curve (AUC) for predicting ICI-related myocarditis was 0.81, with an accuracy of 0.74. The Cohen’s Kappa coefficient was 0.55, indicating moderate agreement. In the actual high-risk group, no patient was predicted to be at low risk; in the actual low-risk group, 16 patients were predicted to be at high risk. Conclusions This risk assessment scale for ICI-associated myocarditis shows high predictive performance. It provides oncologists with a simple yet effective multidisciplinary diagnostic reference tool, potentially enhancing early identification of ICI-associated myocarditis.

Objective:To explore the establishment of a Cox proportional hazard prediction model for the re-cent(within a week)delivery of pregnant women with threatened early preterm labor(28-33+6 weeks).Meth-ods:The clinical data of 293 patients with threatened early preterm labor at 28-33+6 weeks admitted to the De-partment of Obstetrics,Affiliated Hospital of North Sichuan Medical College from January 2021 to December 2022 were retrospectively collected.According to whether the patient gave birth recently(within a week)during hospi-talization,they were divided into delivery group(n=88)and non-delivery group(n=205).Demographic data,specialty conditions and clinical biochemical indicators were collected,and the differences of each observation in-dex between the two groups were compared.Cox multivariate analysis was used to screen variables,and then the Cox proportional hazard prediction model was constructed to calculate the hazard ratio(HR).The C-index was used to test the predictive ability of the model,and the nomogram was drawn to visualize the Cox proportional hazard prediction model.The calibration curve was used to test the consistency between the model prediction re-sults and the actual situation.Results:The average age of the delivery group was 30.2±5.0 years old,and the average prolonged pregnancy time was 61.3±47.5 hours.Compared to the group that gave birth and did not give birth,the proportion of patients with body mass index(BMI)at admission,parity＞1,gestational diabetes mellitus or diabetic pregnancy combined(GDM/PGDM),abnormal fetal orientation,regular contractions at admis-sion,vaginal cleanliness≥ Ⅲ,white blood cell count ≥10 ×109/L at admission and cervical length＜20 mm,the difference was statistically significant(P＜0.05).Multivariate Cox regression analysis:BMI＞30 kg/m2(HR 2.178,95％CI 1.208-3.928,P=0.010),parity＞1(HR 3.095,95％CI 1.759-5.447,P＜0.001),regular uterine contraction(HR3.447,95％CI 2.216-5.362,P＜0.001),cervical length＜20 mm(HR3.594,95％CI 2.289-5.646,P＜0.001)and white blood cell count≥10 ×109/L(HR2.124,95％CI 1.352-3.335,P=0.001)on admission were independent risk factors for recent(within a week)delivery of pregnant women with threatened early preterm birth.When the above 5 indicators were included in the prediction model,the C-index was 0.797(95％CI 0.750-0.844),indicating that the model had good prediction ability.The calibration chart shows that the prediction results of the model are in good agreement with the actual results.Conclusions:Preg-nant women with BMI＞30 kg/m2 at admission,parity＞1,regular uterine contraction at admission,cervical length＜20 mm at admission and white blood cell count≥10 ×109/L at admission are prone to premature delivery in the near future.Cox proportional risk prediction model based on the above risk factors has certain accuracy,which can be used to guide clinical workers to take corresponding intervention measures in advance to avoid or reduce the possibility of premature delivery in the near future and improve maternal and infant outcomes.

Objective:To assess the effectiveness of a 6-month Ba Duan Jin exercise program in improving the balance of community-dwelling older adults.Methods:A two arms, parallel-group, cluster randomized controlled trial was conducted in 1 028 community residents aged 60-80 years in 40 communities in 5 provinces of China. Participants in the intervention group (20 communities, 523 people) received Ba Duan Jin exercise 5 days/week, 1 hour/day for 6 months, and three times of falls prevention health education, and the control group (20 communities, 505 people) received falls prevention health education same as the intervention group. The Berg balance scale (BBS) score was the leading outcome indicator, and the secondary outcome indicators included the length of time of standing on one foot (with eyes open and closed), standing in a tandem stance (with eyes open and closed), the closed circle test, and the timed up to test.Results:A total of 1 028 participants were included in the final analysis, including 731 women (71.11%) and 297 men (28.89%), and the age was (69.87±5.67) years. After the 3-month intervention, compared with the baseline data, the BBS score of the intervention group was significantly higher than the control group by 3.05 (95% CI: 2.23-3.88) points ( P<0.001). After the 6-month intervention, compared with the baseline data, the BBS score of the intervention group was significantly higher than the control group by 4.70 (95% CI: 4.03-5.37) points ( P<0.001). Ba Duan Jin showed significant improvement ( P<0.05) in all secondary outcomes after 6 months of exercise in the intervention group compared with the control group. Conclusions:This study showed that Ba Duan Jin exercise can improve balance in community-dwelling older adults aged 60-80. The longer the exercise time, the better the improvement.

【Objective】 To study and compare the effects of different storage temperature and time on coagulation factor after cryoprecipitated antihemophilic factor(CAF) melting, and to provide reference for the establishment of industry standards. 【Methods】 From June 2021 to May 2023, a total of 96 bags of CAF were sampled in 4 bags per month, and timely detected in the same month. After the CAF was melted in a 37℃ water bath, the mild to moderate lipemic blood was labeled. Each bag of CAF and two 50 mL transfer bags were divided into two bags and two groups of 20 mL each using a sterile adapter. One group was placed in a 4℃ refrigerator and the other in a 22℃ water bath for 0 h, 4 h, 8 h, 12 h, 24 h and 48 h. Then 2 mL of aseptic sample was taken separately and put into the test tube, and 1mL of sample and 3 mL of buffer were added into the other test tube with the sampling gun and mixed on the machine for testing. The experimental data of 60 bags without mild to moderate lipemic blood cryoprecipitation and coagulation factor were randomly selected and statistically analyzed by SPSS21.0. 【Results】 After melting, CAF was stored for 0 h, 4 h, 8 h, 12 h, 24 h and 48 h to detect the average content and growth rate of coagulation factor in the two groups: 1) Storage at 4℃, factor Ⅷ content was 118.62, 111.57(-5.95%), 105.51(-11.05%), 103.30(-12.92%), 94.35(-20.46%) and 83.25(-29.82%) IU/ bag, respectively; Storage at 22℃, the factor Ⅷ content was 118.62, 112.69(-5.00%), 111.41(-6.08%), 109.01(-8.10%), 101.55(-14.39%) and 92.75(-21.81%) IU/ bag, and the storage results of the two groups were compared. At 24 h at 4℃ and 48 h at 22℃, the content of factor Ⅷ had significant statistical significance(P<0.01), and when stored at 22℃, the decay rate of factor Ⅷ was slower; 2) When stored at 4℃, the content of factor V was 41.19, 41.31(0.29%), 40.52(-1.64%), 40.27(-2.23%), 39.05(-5.19%) and 36.99(-10.21%) IU/ bag, respectively; Stored at 22℃, the factor V content was 41.19, 41.71(1.25%), 42.54(3.28%), 41.94(1.80%), 39.21(-4.80%) and 35.64(-13.48%) IU/ bag, respectively. Comparison of storage results between the two groups showed that the content of factor V was statistically significant(P<0.05) and significantly significant(P<0.01) at 4℃48 h and 22℃48 h, respectively, and the decay rate of factor V was faster when stored at 22℃; 3) When stored at 4℃, the Fbg content was 268.86, 268.17(-0.26%), 262.46(-2.38%), 270.50(0.61%), 267.52(-0.50%) and 261.92(-2.58%) mg/ bag, respectively; Stored at 22℃, the Fbg content was 268.86, 265.86(-1.12%), 264.12(-1.77%), 265.89(-1.11%), 266.04(-1.05%) and 261.04(-2.91%) mg/ bag, respectively. There was no statistical significance between the 2 groups and the original 0 h content in each time period(P>0.05). 【Conclusion】 After CAF melting, coagulation factor decreased with the extension of storage time, especially the decrease of factor Ⅷ, followed by factor V, while Fbg basically unchanged. Comparison between the two groups showed that, factor Ⅷ decay rate is slower, factor V decay rate is faster of storage at 22℃. CAF should be transfused as soon as possible after melting. If the delay is unavoidable, for the delay time less than 12 h, storage at 4℃ is recommended, fot the delay time more than 12 h and less than 24 h, storage at 22℃ is recommended.

OBJECTIVE To explore the mechanism of limonin treating in hepatic fibrosis through network pharmacology, and validate its mechanism by molecular docking and animal experiments. METHODS Firstly, the targets of limonin and hepatic fibrosis were screened from the SwissTargetPrediction, GeneCards and DisGeNet database, etc. Meanwhile, the common targets of limonin and hepatic fibrosis were obtained from the bioinformatics website. The protein protein interaction network of common target was constructed by using STRING database and Cytoscape software, and the CytoNCA plug-in was used to screen core targets. And then the enrichment analysis of GO and KEGG on the common target was performed by Metascape database. Thereby, the possible mechanism of limonin against hepatic fibrosis were predicted. Finally, the AutoDock Vina was used for molecular docking verification, and the prediction results of network pharmacology were verified by animal experiments. RESULTS The prediction results indicated that limonin might acted on 86 targets including AKT1, VEGFA and HIF1A, and participated in biological processes including hormone response, protein phosphorylation, angiogenesis, and PI3K-Akt pathway, HIF-1 pathway, VEGF pathway and other signaling pathways related to hepatic fibrosis. The results of protein protein interaction network topology analysis showed that the 11 core targets including AKT1, VEGFA, HIF1A and PIK3CA, etc. Molecular docking results showed that limonin had strong affinity and relatively stable binding conformation with the core targets. In the animal experiments, compared with the model group, hyaluronidase(HA) and laminin(LN) in rat serume in high-dose group of limonin(LH) and low-dose group of limonin(LL)(except for LN in LL group) were declined(P<0.01 or P<0.05), and the degree of inflammation and hepatic fibrosis were relieved to different degrees in liver tissue of the LH group and LL group; Western blotting and qPCR detection showed that protein and mRNA expression levels of AKT, HIF-1α and VEGF(except for VEGF in LL group) was down-regulated in the LH group and LL group(P<0.01 or P<0.05). CONCLUSION Limonin may acts on AKT1, VEGFA, HIF1A and other core targets to treat hepatic fibrosis angiogenesis, which may be related to the inhibition of AKT/HIF-1α/VEGF signaling pathway.

Objective:To explore the establishment of a Cox proportional hazard prediction model for the re-cent(within a week)delivery of pregnant women with threatened early preterm labor(28-33+6 weeks).Meth-ods:The clinical data of 293 patients with threatened early preterm labor at 28-33+6 weeks admitted to the De-partment of Obstetrics,Affiliated Hospital of North Sichuan Medical College from January 2021 to December 2022 were retrospectively collected.According to whether the patient gave birth recently(within a week)during hospi-talization,they were divided into delivery group(n=88)and non-delivery group(n=205).Demographic data,specialty conditions and clinical biochemical indicators were collected,and the differences of each observation in-dex between the two groups were compared.Cox multivariate analysis was used to screen variables,and then the Cox proportional hazard prediction model was constructed to calculate the hazard ratio(HR).The C-index was used to test the predictive ability of the model,and the nomogram was drawn to visualize the Cox proportional hazard prediction model.The calibration curve was used to test the consistency between the model prediction re-sults and the actual situation.Results:The average age of the delivery group was 30.2±5.0 years old,and the average prolonged pregnancy time was 61.3±47.5 hours.Compared to the group that gave birth and did not give birth,the proportion of patients with body mass index(BMI)at admission,parity＞1,gestational diabetes mellitus or diabetic pregnancy combined(GDM/PGDM),abnormal fetal orientation,regular contractions at admis-sion,vaginal cleanliness≥ Ⅲ,white blood cell count ≥10 ×109/L at admission and cervical length＜20 mm,the difference was statistically significant(P＜0.05).Multivariate Cox regression analysis:BMI＞30 kg/m2(HR 2.178,95％CI 1.208-3.928,P=0.010),parity＞1(HR 3.095,95％CI 1.759-5.447,P＜0.001),regular uterine contraction(HR3.447,95％CI 2.216-5.362,P＜0.001),cervical length＜20 mm(HR3.594,95％CI 2.289-5.646,P＜0.001)and white blood cell count≥10 ×109/L(HR2.124,95％CI 1.352-3.335,P=0.001)on admission were independent risk factors for recent(within a week)delivery of pregnant women with threatened early preterm birth.When the above 5 indicators were included in the prediction model,the C-index was 0.797(95％CI 0.750-0.844),indicating that the model had good prediction ability.The calibration chart shows that the prediction results of the model are in good agreement with the actual results.Conclusions:Preg-nant women with BMI＞30 kg/m2 at admission,parity＞1,regular uterine contraction at admission,cervical length＜20 mm at admission and white blood cell count≥10 ×109/L at admission are prone to premature delivery in the near future.Cox proportional risk prediction model based on the above risk factors has certain accuracy,which can be used to guide clinical workers to take corresponding intervention measures in advance to avoid or reduce the possibility of premature delivery in the near future and improve maternal and infant outcomes.

Objective:To explore the establishment of a Cox proportional hazard prediction model for the re-cent(within a week)delivery of pregnant women with threatened early preterm labor(28-33+6 weeks).Meth-ods:The clinical data of 293 patients with threatened early preterm labor at 28-33+6 weeks admitted to the De-partment of Obstetrics,Affiliated Hospital of North Sichuan Medical College from January 2021 to December 2022 were retrospectively collected.According to whether the patient gave birth recently(within a week)during hospi-talization,they were divided into delivery group(n=88)and non-delivery group(n=205).Demographic data,specialty conditions and clinical biochemical indicators were collected,and the differences of each observation in-dex between the two groups were compared.Cox multivariate analysis was used to screen variables,and then the Cox proportional hazard prediction model was constructed to calculate the hazard ratio(HR).The C-index was used to test the predictive ability of the model,and the nomogram was drawn to visualize the Cox proportional hazard prediction model.The calibration curve was used to test the consistency between the model prediction re-sults and the actual situation.Results:The average age of the delivery group was 30.2±5.0 years old,and the average prolonged pregnancy time was 61.3±47.5 hours.Compared to the group that gave birth and did not give birth,the proportion of patients with body mass index(BMI)at admission,parity＞1,gestational diabetes mellitus or diabetic pregnancy combined(GDM/PGDM),abnormal fetal orientation,regular contractions at admis-sion,vaginal cleanliness≥ Ⅲ,white blood cell count ≥10 ×109/L at admission and cervical length＜20 mm,the difference was statistically significant(P＜0.05).Multivariate Cox regression analysis:BMI＞30 kg/m2(HR 2.178,95％CI 1.208-3.928,P=0.010),parity＞1(HR 3.095,95％CI 1.759-5.447,P＜0.001),regular uterine contraction(HR3.447,95％CI 2.216-5.362,P＜0.001),cervical length＜20 mm(HR3.594,95％CI 2.289-5.646,P＜0.001)and white blood cell count≥10 ×109/L(HR2.124,95％CI 1.352-3.335,P=0.001)on admission were independent risk factors for recent(within a week)delivery of pregnant women with threatened early preterm birth.When the above 5 indicators were included in the prediction model,the C-index was 0.797(95％CI 0.750-0.844),indicating that the model had good prediction ability.The calibration chart shows that the prediction results of the model are in good agreement with the actual results.Conclusions:Preg-nant women with BMI＞30 kg/m2 at admission,parity＞1,regular uterine contraction at admission,cervical length＜20 mm at admission and white blood cell count≥10 ×109/L at admission are prone to premature delivery in the near future.Cox proportional risk prediction model based on the above risk factors has certain accuracy,which can be used to guide clinical workers to take corresponding intervention measures in advance to avoid or reduce the possibility of premature delivery in the near future and improve maternal and infant outcomes.

Objective:To explore the establishment of a Cox proportional hazard prediction model for the re-cent(within a week)delivery of pregnant women with threatened early preterm labor(28-33+6 weeks).Meth-ods:The clinical data of 293 patients with threatened early preterm labor at 28-33+6 weeks admitted to the De-partment of Obstetrics,Affiliated Hospital of North Sichuan Medical College from January 2021 to December 2022 were retrospectively collected.According to whether the patient gave birth recently(within a week)during hospi-talization,they were divided into delivery group(n=88)and non-delivery group(n=205).Demographic data,specialty conditions and clinical biochemical indicators were collected,and the differences of each observation in-dex between the two groups were compared.Cox multivariate analysis was used to screen variables,and then the Cox proportional hazard prediction model was constructed to calculate the hazard ratio(HR).The C-index was used to test the predictive ability of the model,and the nomogram was drawn to visualize the Cox proportional hazard prediction model.The calibration curve was used to test the consistency between the model prediction re-sults and the actual situation.Results:The average age of the delivery group was 30.2±5.0 years old,and the average prolonged pregnancy time was 61.3±47.5 hours.Compared to the group that gave birth and did not give birth,the proportion of patients with body mass index(BMI)at admission,parity＞1,gestational diabetes mellitus or diabetic pregnancy combined(GDM/PGDM),abnormal fetal orientation,regular contractions at admis-sion,vaginal cleanliness≥ Ⅲ,white blood cell count ≥10 ×109/L at admission and cervical length＜20 mm,the difference was statistically significant(P＜0.05).Multivariate Cox regression analysis:BMI＞30 kg/m2(HR 2.178,95％CI 1.208-3.928,P=0.010),parity＞1(HR 3.095,95％CI 1.759-5.447,P＜0.001),regular uterine contraction(HR3.447,95％CI 2.216-5.362,P＜0.001),cervical length＜20 mm(HR3.594,95％CI 2.289-5.646,P＜0.001)and white blood cell count≥10 ×109/L(HR2.124,95％CI 1.352-3.335,P=0.001)on admission were independent risk factors for recent(within a week)delivery of pregnant women with threatened early preterm birth.When the above 5 indicators were included in the prediction model,the C-index was 0.797(95％CI 0.750-0.844),indicating that the model had good prediction ability.The calibration chart shows that the prediction results of the model are in good agreement with the actual results.Conclusions:Preg-nant women with BMI＞30 kg/m2 at admission,parity＞1,regular uterine contraction at admission,cervical length＜20 mm at admission and white blood cell count≥10 ×109/L at admission are prone to premature delivery in the near future.Cox proportional risk prediction model based on the above risk factors has certain accuracy,which can be used to guide clinical workers to take corresponding intervention measures in advance to avoid or reduce the possibility of premature delivery in the near future and improve maternal and infant outcomes.

Objective:To explore the establishment of a Cox proportional hazard prediction model for the re-cent(within a week)delivery of pregnant women with threatened early preterm labor(28-33+6 weeks).Meth-ods:The clinical data of 293 patients with threatened early preterm labor at 28-33+6 weeks admitted to the De-partment of Obstetrics,Affiliated Hospital of North Sichuan Medical College from January 2021 to December 2022 were retrospectively collected.According to whether the patient gave birth recently(within a week)during hospi-talization,they were divided into delivery group(n=88)and non-delivery group(n=205).Demographic data,specialty conditions and clinical biochemical indicators were collected,and the differences of each observation in-dex between the two groups were compared.Cox multivariate analysis was used to screen variables,and then the Cox proportional hazard prediction model was constructed to calculate the hazard ratio(HR).The C-index was used to test the predictive ability of the model,and the nomogram was drawn to visualize the Cox proportional hazard prediction model.The calibration curve was used to test the consistency between the model prediction re-sults and the actual situation.Results:The average age of the delivery group was 30.2±5.0 years old,and the average prolonged pregnancy time was 61.3±47.5 hours.Compared to the group that gave birth and did not give birth,the proportion of patients with body mass index(BMI)at admission,parity＞1,gestational diabetes mellitus or diabetic pregnancy combined(GDM/PGDM),abnormal fetal orientation,regular contractions at admis-sion,vaginal cleanliness≥ Ⅲ,white blood cell count ≥10 ×109/L at admission and cervical length＜20 mm,the difference was statistically significant(P＜0.05).Multivariate Cox regression analysis:BMI＞30 kg/m2(HR 2.178,95％CI 1.208-3.928,P=0.010),parity＞1(HR 3.095,95％CI 1.759-5.447,P＜0.001),regular uterine contraction(HR3.447,95％CI 2.216-5.362,P＜0.001),cervical length＜20 mm(HR3.594,95％CI 2.289-5.646,P＜0.001)and white blood cell count≥10 ×109/L(HR2.124,95％CI 1.352-3.335,P=0.001)on admission were independent risk factors for recent(within a week)delivery of pregnant women with threatened early preterm birth.When the above 5 indicators were included in the prediction model,the C-index was 0.797(95％CI 0.750-0.844),indicating that the model had good prediction ability.The calibration chart shows that the prediction results of the model are in good agreement with the actual results.Conclusions:Preg-nant women with BMI＞30 kg/m2 at admission,parity＞1,regular uterine contraction at admission,cervical length＜20 mm at admission and white blood cell count≥10 ×109/L at admission are prone to premature delivery in the near future.Cox proportional risk prediction model based on the above risk factors has certain accuracy,which can be used to guide clinical workers to take corresponding intervention measures in advance to avoid or reduce the possibility of premature delivery in the near future and improve maternal and infant outcomes.

Objective:To explore the establishment of a Cox proportional hazard prediction model for the re-cent(within a week)delivery of pregnant women with threatened early preterm labor(28-33+6 weeks).Meth-ods:The clinical data of 293 patients with threatened early preterm labor at 28-33+6 weeks admitted to the De-partment of Obstetrics,Affiliated Hospital of North Sichuan Medical College from January 2021 to December 2022 were retrospectively collected.According to whether the patient gave birth recently(within a week)during hospi-talization,they were divided into delivery group(n=88)and non-delivery group(n=205).Demographic data,specialty conditions and clinical biochemical indicators were collected,and the differences of each observation in-dex between the two groups were compared.Cox multivariate analysis was used to screen variables,and then the Cox proportional hazard prediction model was constructed to calculate the hazard ratio(HR).The C-index was used to test the predictive ability of the model,and the nomogram was drawn to visualize the Cox proportional hazard prediction model.The calibration curve was used to test the consistency between the model prediction re-sults and the actual situation.Results:The average age of the delivery group was 30.2±5.0 years old,and the average prolonged pregnancy time was 61.3±47.5 hours.Compared to the group that gave birth and did not give birth,the proportion of patients with body mass index(BMI)at admission,parity＞1,gestational diabetes mellitus or diabetic pregnancy combined(GDM/PGDM),abnormal fetal orientation,regular contractions at admis-sion,vaginal cleanliness≥ Ⅲ,white blood cell count ≥10 ×109/L at admission and cervical length＜20 mm,the difference was statistically significant(P＜0.05).Multivariate Cox regression analysis:BMI＞30 kg/m2(HR 2.178,95％CI 1.208-3.928,P=0.010),parity＞1(HR 3.095,95％CI 1.759-5.447,P＜0.001),regular uterine contraction(HR3.447,95％CI 2.216-5.362,P＜0.001),cervical length＜20 mm(HR3.594,95％CI 2.289-5.646,P＜0.001)and white blood cell count≥10 ×109/L(HR2.124,95％CI 1.352-3.335,P=0.001)on admission were independent risk factors for recent(within a week)delivery of pregnant women with threatened early preterm birth.When the above 5 indicators were included in the prediction model,the C-index was 0.797(95％CI 0.750-0.844),indicating that the model had good prediction ability.The calibration chart shows that the prediction results of the model are in good agreement with the actual results.Conclusions:Preg-nant women with BMI＞30 kg/m2 at admission,parity＞1,regular uterine contraction at admission,cervical length＜20 mm at admission and white blood cell count≥10 ×109/L at admission are prone to premature delivery in the near future.Cox proportional risk prediction model based on the above risk factors has certain accuracy,which can be used to guide clinical workers to take corresponding intervention measures in advance to avoid or reduce the possibility of premature delivery in the near future and improve maternal and infant outcomes.