Blood transfusion is an important life-support measure in modern medicine. As clinical demand for transfusion continues to rise, blood supply remains under persistent strain, making the efficient utilization of this precious medical resource a critical component of blood management. Accurate prediction of transfusion needs is of great significance for optimizing blood resource allocation and conserving blood products. Traditional transfusion decision-making relies on clinical experience and simplified scoring systems, which are insufficient to meet the demands of precision medicine. Machine learning (ML) technology can integrate multidimensional data—including demographic characteristics, laboratory indicators, surgical information, and dynamic physiological waveforms—to construct high-performance predictive models, offering a new approach to transfusion prediction. This article systematically reviews the application progress of ML in transfusion prediction across clinical scenarios such as trauma, the perioperative period, and obstetrics. In trauma, ML has been applied to early massive transfusion prediction, prehospital transfusion decision-making, and precise prediction in pediatric trauma. In the perioperative field, applications span specialties including traumatic brain injury, orthopedic surgery, cardiovascular and major vascular surgery, and hepatic surgery. In obstetrics, models can effectively predict postpartum hemorrhage and transfusion requirements associated with cesarean section. Studies have shown that algorithms such as random forest, gradient boosting machines, and deep neural networks achieve area under the receiver operating characteristic curve (AUC) values of 0.83-0.98 in massive transfusion prediction, 0.75-0.97 in perioperative scenarios, and 0.80-0.89 in obstetric transfusion prediction—significantly outperforming traditional scoring tools. However, current studies are generally limited by single-center retrospective designs, insufficient external validation, and inadequate reporting standards; limited model interpretability and challenges in clinical integration further constrain practical translation. Future research should focus on constructing multicenter data cohorts, integrating multimodal data, conducting prospective implementation studies, and deeply embedding models within electronic health record systems, thereby advancing the application of ML to optimize clinical transfusion decision-making.

Objective:To investigate the risk factors associated with post-prematurity respiratory disease (PPRD) in very preterm infants.Methods:A prospective cohort study was conducted, enrolling 369 very preterm infants who were admitted to the neonatal intensive care unit of Xinhua Hospital, Shanghai Jiao Tong University School of Medicine, within one week of birth from January 2019 to June 2023. Data on maternal and infant clinical characteristics, neonatal morbidities, and treatments during hospitalization were collected. The very preterm infants were divided into 2 groups based on whether they developed PPRD. Continuous variables were compared using Mann-Whitney U test, while categorical variables were compared using χ2 tests or continuity correction χ2 test. Multivariate Logistic regression analysis was used to identify the independent risk factors for PPRD in very preterm infants. Results:Among the 369 very preterm infants, 217 cases(58.8%) were male, with a gestational age of 30 (28, 31) weeks at birth and a birth weight of 1 320 (1 085, 1 590) g. Of these, 116 cases (31.4%) developed PPRD, while 253 cases (68.6%) did not. The very preterm infants in the PPRD group had a lower gestational age and lower birth weight (both, P<0.001). The PPRD group also had a higher proportion of males, lower Apgar scores at the 1 th minute after birth and the 5 th minutes after birth, a higher rate of born via cesarean delivery, and a higher incidence of bronchopulmonary dysplasia, more pulmonary surfactant treatment, longer durations of mechanical ventilation, longer total oxygen therapy, and lower Z-score for weight at discharge (all P<0.05). Multivariate Logistic regression analysis showed that gestational age ( OR=0.85, 95% CI 0.73-0.99, P=0.037), born via cesarean delivery ( OR=2.23, 95% CI 1.21-4.10, P=0.010), a duration of mechanical ventilation ≥7 days ( OR=2.51, 95% CI 1.43-4.39, P=0.001), and a Z-score for weight at discharge ( OR=0.82, 95% CI 0.67-0.99, P=0.040) were all independent risk factors for PPRD in very preterm infants. Conclusion:Very preterm infants with a small gestational age, born via cesarean section, mechanical ventilation ≥7 days, and a low Z-score for weight at discharge should be closely monitored for PPRD, and provided with standardized respiratory management after discharge.

Objective:To investigate the clinical efficacy of cervical cerclage in singleton pregnancy with different degrees of short cervix.Methods:The clinical data of singleton pregnant women who underwent transvaginal ultrasound examination at 18-24 +6 weeks of gestation and found cervical dilation with cervical length (CL) ≤20 mm, and without history of spontaneous preterm delivery or late abortion in Women′s Hospital, Zhejiang University School of Medicine from January 2021 to September 2023 were collected and retrospectively analyzed. According to the case control matching, 78 pregnant women in the cerclage group and 78 women in the conservative treatment group were finally included. The pregnancy outcomes and neonatal prognosis of the two groups were compared. Meanwhile, the two groups of pregnant women were divided into three subgroups for stratification (CL≤10, 11-15, 16-20 mm). Multivariate logistic regression analysis and Kaplan-Meier curve were used to evaluate the effect of cervical cerclage on pregnancy outcomes in different subgroups. Results:(1) Compared with the conservative treatment group, the gestational age at delivery (median: 36 vs 37 weeks) and the duration of pregnancy extension (median: 90 vs 97 days) in the cerclage group were not significantly prolonged (all P>0.05). There was no significant difference in the cumulative non-delivery rate between the cerclage group and the conservative treatment group ( P=0.143). The rate of neonatal intensive care unit (NICU) admission in the cerclage group was significantly higher than that in the conservative treatment group (46.1% vs 31.2%, P<0.001), but there were no significant differences in other neonatal outcomes between the two groups (all P>0.05). (2) CL 16-20 mm subgroup: compared with the conservative treatment group (28 cases), the cerclage group (18 cases) had a significantly higher rate of preterm birth before 37 weeks of gestation (3.6% vs 6/18, P<0.001) and a significantly lower neonatal birth weight (median: 3 370 vs 2 925 g, P<0.001). There was no significant difference in the cumulative non-delivery rate between the cerclage group and the conservative treatment group ( P=0.168). (3) CL 11-15 mm subgroup: compared with the conservative treatment group (26 cases), the gestational age of delivery in the cerclage group (32 cases) was later (median: 36 and 37 weeks, respectively), and the difference was statistically significant ( P<0.05). The cumulative non-delivery rate in the cerclage group was significantly higher than that in the conservative treatment group ( P=0.001). (4) CL≤10 mm subgroup: compared with the conservative treatment group (24 cases), the pregnant women in the cerclage group (28 cases) had a later gestational age at delivery (median: 34 vs 37 weeks), a longer duration of pregnancy (median: 74 vs 97 days), and a larger newborn birth weight (median: 2 300 vs 3 165 g). The rates of preterm birth before 34 weeks of gestation (45.8% vs 14.3%) and before 37 weeks of gestation (83.3% vs 39.3%) were lower, and the differences were all statistically significant (all P<0.05). The cumulative non-delivery rate of pregnant women in the cerclage group was significantly higher than that in the conservative treatment group ( P<0.001). Conclusion:Cervical cerclage could significantly prolong the gestational weeks and improve the perinatal outcomes for singleton pregnant women with cervical internal orifice dilation and CL≤15 mm without a history of spontaneous preterm delivery or late abortion.

Objective:To examine the clinical effects of the SCERTS model-based peer-mediated intervention (PMI) on the social skills of children with autism spectrum disorder (ASD) and provide new approaches to the rehabilitation treatment of ASD children.Methods:A randomized controlled study.A total of 120 children with mild-to-moderate ASD diagnosed Xi′an Traditional Chinese Medicine Hospital of Encephalopathy between April 2023 and April 2024 were selected.They were assigned to either the experimental or control group using the random number table method, with 60 cases in each group.The experimental group was treated with the SCERTS model-based PMI, whereas the control group underwent conventional rehabilitation training, comprising cognitive, language and behavioural interventions, acupuncture and other techniques.Changes in Social Responsiveness Scale (SRS), Autism Social Skills Scale (ASSS) and Autism Treatment Evaluation Checklist (ATEC) scores were observed in the 2 groups of children before and after treatment.The statistical analyses were performed using a paired-samples t-test for intra-group comparisons and an independent-samples t-test for inter-group comparisons. Results:Following 12 weeks of intervention, the SRS and ATEC scores decreased while the ASSS scores increased after treatment in both groups.The differences were statistically significant (all P<0.05).The SRS [(83.25±14.56) points] and ATEC [(79.41±15.36) points] of the experimental group were lower than those of the control group [SRS(89.80±12.69) points, ATEC(85.95±16.13) points].The ASSS score of the experimental group was higher than that of the control group [(112.77±22.42) points vs.(103.80±24.13) points] ( t=2.627, 2.274, -2.109; all P<0.05). Conclusions:The SCERTS model-based PMI is an efficacious approach for the improvement of social skills in ASD children, and thus merits further investigation and application.

Objective:To investigate the clinical efficacy of cervical cerclage in singleton pregnancy with different degrees of short cervix.Methods:The clinical data of singleton pregnant women who underwent transvaginal ultrasound examination at 18-24 +6 weeks of gestation and found cervical dilation with cervical length (CL) ≤20 mm, and without history of spontaneous preterm delivery or late abortion in Women′s Hospital, Zhejiang University School of Medicine from January 2021 to September 2023 were collected and retrospectively analyzed. According to the case control matching, 78 pregnant women in the cerclage group and 78 women in the conservative treatment group were finally included. The pregnancy outcomes and neonatal prognosis of the two groups were compared. Meanwhile, the two groups of pregnant women were divided into three subgroups for stratification (CL≤10, 11-15, 16-20 mm). Multivariate logistic regression analysis and Kaplan-Meier curve were used to evaluate the effect of cervical cerclage on pregnancy outcomes in different subgroups. Results:(1) Compared with the conservative treatment group, the gestational age at delivery (median: 36 vs 37 weeks) and the duration of pregnancy extension (median: 90 vs 97 days) in the cerclage group were not significantly prolonged (all P>0.05). There was no significant difference in the cumulative non-delivery rate between the cerclage group and the conservative treatment group ( P=0.143). The rate of neonatal intensive care unit (NICU) admission in the cerclage group was significantly higher than that in the conservative treatment group (46.1% vs 31.2%, P<0.001), but there were no significant differences in other neonatal outcomes between the two groups (all P>0.05). (2) CL 16-20 mm subgroup: compared with the conservative treatment group (28 cases), the cerclage group (18 cases) had a significantly higher rate of preterm birth before 37 weeks of gestation (3.6% vs 6/18, P<0.001) and a significantly lower neonatal birth weight (median: 3 370 vs 2 925 g, P<0.001). There was no significant difference in the cumulative non-delivery rate between the cerclage group and the conservative treatment group ( P=0.168). (3) CL 11-15 mm subgroup: compared with the conservative treatment group (26 cases), the gestational age of delivery in the cerclage group (32 cases) was later (median: 36 and 37 weeks, respectively), and the difference was statistically significant ( P<0.05). The cumulative non-delivery rate in the cerclage group was significantly higher than that in the conservative treatment group ( P=0.001). (4) CL≤10 mm subgroup: compared with the conservative treatment group (24 cases), the pregnant women in the cerclage group (28 cases) had a later gestational age at delivery (median: 34 vs 37 weeks), a longer duration of pregnancy (median: 74 vs 97 days), and a larger newborn birth weight (median: 2 300 vs 3 165 g). The rates of preterm birth before 34 weeks of gestation (45.8% vs 14.3%) and before 37 weeks of gestation (83.3% vs 39.3%) were lower, and the differences were all statistically significant (all P<0.05). The cumulative non-delivery rate of pregnant women in the cerclage group was significantly higher than that in the conservative treatment group ( P<0.001). Conclusion:Cervical cerclage could significantly prolong the gestational weeks and improve the perinatal outcomes for singleton pregnant women with cervical internal orifice dilation and CL≤15 mm without a history of spontaneous preterm delivery or late abortion.

Objective:To examine the clinical effects of the SCERTS model-based peer-mediated intervention (PMI) on the social skills of children with autism spectrum disorder (ASD) and provide new approaches to the rehabilitation treatment of ASD children.Methods:A randomized controlled study.A total of 120 children with mild-to-moderate ASD diagnosed Xi′an Traditional Chinese Medicine Hospital of Encephalopathy between April 2023 and April 2024 were selected.They were assigned to either the experimental or control group using the random number table method, with 60 cases in each group.The experimental group was treated with the SCERTS model-based PMI, whereas the control group underwent conventional rehabilitation training, comprising cognitive, language and behavioural interventions, acupuncture and other techniques.Changes in Social Responsiveness Scale (SRS), Autism Social Skills Scale (ASSS) and Autism Treatment Evaluation Checklist (ATEC) scores were observed in the 2 groups of children before and after treatment.The statistical analyses were performed using a paired-samples t-test for intra-group comparisons and an independent-samples t-test for inter-group comparisons. Results:Following 12 weeks of intervention, the SRS and ATEC scores decreased while the ASSS scores increased after treatment in both groups.The differences were statistically significant (all P<0.05).The SRS [(83.25±14.56) points] and ATEC [(79.41±15.36) points] of the experimental group were lower than those of the control group [SRS(89.80±12.69) points, ATEC(85.95±16.13) points].The ASSS score of the experimental group was higher than that of the control group [(112.77±22.42) points vs.(103.80±24.13) points] ( t=2.627, 2.274, -2.109; all P<0.05). Conclusions:The SCERTS model-based PMI is an efficacious approach for the improvement of social skills in ASD children, and thus merits further investigation and application.

Objective: To analyze the frequency and investigate the causes of unexpected antibodies in D-negative blood donors. Methods: From January 2022 to December 2024, 3 768 D-negative blood donors sent to our laboratory were selected as research subjects. D-negative confirmation test and RhCE phenotype detection were applied by saline tube method and microcolumn gel indirect antiglobulin test (IAT), respectively. Antibody screening and identification were performed using the polybrene method and IAT column agglutination methods. Anti-D, anti-C and anti-G specificities were identified by a two-step adsorption-elution method, and the genotypes of D-negative samples were determined by RHD gene amplification, Sanger sequencing, and PacBio Single Molecule Real-Time (SMRT) sequencing. Results: Among D-negative donors, ccee and Ccee phenotypes accounted for the highest proportion, 55.68% (2 098/3 768) and 29.56% (1 114/3 768), respectively, while CcEE and CCEe phenotypes were the least, with one case detected in each, accounting for 0.03% (1/3 768). A total of 165 cases with D variant phenotype were detected, and the proportion of D variant was 4.38% (165/3 768) in the donors detected by D-negative confirmation test. Antibody screening positive blood donors were identified in 93 cases with a proportion of 2.47% (93/3 768). Antibody specificity was determined in 84 blood donors, and 9 samples showed no clear specificity. Anti-D was detected most frequently (n=68), in which 6 of them were detected having multiple antibodies, anti-D + anti-C (n=2), anti-D + anti-G(n=1), and anti-D + anti-E(n=3). The other antibodies detected were anti-E (n=1), anti-M(n=9), anti-P1 (n=3), anti-Le (n=1), and anti-HI(n=2). Fourteen cases were detected with anti-D in serological D-negative donors with C+ or E+ phenotype, in which three of them were DVI type 3 individuals and 11 cases were D negative individuals. Conclusion: The incidence of unexpected antibodies was higher in D-negative blood donors than in the total donors, with anti-D being the most common. The data provide insights for prevention and monitoring hemolytic disease of the fetus and newborn (HDFN) caused by anti-D. To ensure the safety of blood transfusion, routine unexpected antibody screening for RhD-negative blood donors is recommended to prevent the use of unexpected antibodies positive plasma in the clinic.

Objective To analyze the application of serological test results in the diagnosis and treatment of anti-M-in-duced hemolytic disease of the fetus and newborn(HDFN),and to explore HDFN prevention strategies.Methods The se-rological test results of 12 cases of HDFN caused by anti-M diagnosed in our laboratory from January 2017 to December 2023 were retrospectively analyzed,including blood group identification of mothers and children,serum total bilirubin/hemoglo-bin/antibody titer test,and three hemolysis tests in newborns.Clinical data of the children and mothers were collected,in-cluding pregnancy history,blood transfusion history,prenatal antibody testing,history of intrauterine blood transfusion and gestational week of delivery,and the prognosis of the children was followed up.Results All 12 cases of fetal neonatal he-molytic disease due to anti-M were RhD+MN phenotype newborn born to RhD+NN mother,with maternal-fetal incompati-blility in MN blood groups.In the ABO blood group system,ABO incompatibility between mother and child accounted for 41.7%(5/12).None of the mothers had a history of blood transfusion,and the median titer of the test at 4℃was 32,and the median titer at 37℃was 4.The mothers of 3 cases had a history of multiple intrauterine blood transfusions,with an inci-dence of 25%(3/12).One case had an abnormal first pregnancy,with an incidence of 8.3%(1/12),and seven cases had an abnormal pregnancy with a miscarriage,with an incidence of abnormal pregnancy and birth history of 58.3%(7/12).There were 6 cases of premature labor,with an incidence of 50%(6/12).The mothers in three cases underwent regular ob-stetric examination and the specificity of the antibodies was determined,accounting for 25%(3/12).Twelve children had free antibodies with a median titer of 6 at 4℃and 2 at 37℃.Two children had anti-M antibodies that were not reactive at 37℃,with a negative rate of 16.7%(2/12).The positive rate of DAT and elution test was respectively 8.3%(1/12)and 16.7%(2/12)in the children.The median minimum hemoglobin value was 75 g/L,and all 12 children received blood transfusions.The median peak total bilirubin value was 157.5 μmol/L,and none of them reached the threshold for blood ex-change.The rate of delayed anemia was16.7%(2/12),the postnatal mortality rate was8.3%(1/12),and 11 children was free of growth and neurodevelopmental delay in prognosis.Conclusion Anti-M can cause severe HDFN,which can also oc-cur in primigravida.The intensity of antibody titer does not correlate with the severity of the disease,and it is prone to cause delayed anemia,which should be monitored regularly according to the serological characteristics of anti-M and clinical symp-toms,and should be treated timely.

Objective To investigate the serological characteristics and gene mutation mechanism of RhD variant blood donors in Zhengzhou.Methods From January 2023 to December 2023,1 619 RhD-negative blood donors sent to our labora-tory were selected for the study,and RhD negative confirmation test and RhCE phenotype detection were applied by tube method and microcolumn gel indirect antiglobulin test method.RHD gene amplification and Sanger sequencing were used to detect RhD variant sample genotypes.Results A total of 69 cases of RhD variants were detected in the RhD negative con-firmation test,with a proportion of 4.26%(69/1 619).The RhCE phenotypes were ccEe,Ccee,CcEe and CCee.There were 17 genotypes and 15 phenotypes of the D variant.The RHD?weak partial 15 allele was the most frequent(33 cases),with a frequency of 47.83%(33/69),and the main phenotype was the ccEe.This was followed by the RHD?DVI.3 allele in 20 ca-ses with a frequency of 28.99%(20/69)and the predominant phenotype was Ccee.The RHD?weak partial 15/RHD?01EL.01 heterozygote was found in 3 cases with a frequency of 4.35%(3/69),all with the CcEe phenotype.Other rare genotypes were present in 13 cases with a frequency of 18.84%(13/69).Antibody screening was positive in 3 cases with a frequency of 4.35%(3/69).Two cases of female blood donors,both with history of pregnancy and childbirth,were identified as anti-D;one case of male donor was anti-M.Conclusion The RHD?weak partial 15 genotype was the most common among the RhD variants in blood donors in Zhengzhou,followed by the RHD?DVI.3 genotype.It plays an important role in guarantee-ing the safety of blood supply and guiding precision transfusion.

ObjectiveTo investigate the prevalence of polycystic ovary syndrome （PCOS） among female college students at a university in Fuzhou City， Jiangxi Province， and to facilitate early detection and intervention of PCOS. MethodsUsing a stratified sampling method， a total of 450 female freshmen were randomly selected for PCOS screening. A self-designed questionnaire was used for data collection， covering menstrual status， high androgen signs， lifestyle， dietary habits， and awareness of PCOS. Sample t test and χ² test were used to compare the basic information and dietary habits between PCOS and non-PCOS cases. The correlation between various indicators and the prevalence of PCOS was analyzed by a logistic regression model. ResultsA total of 12 PCOS cases were identified， with a prevalence rate of 2.99%. PCOS cases exhibited statistically significant differences compared to non-PCOS cases in terms of waist-to-hip ratio， waist circumference， abdominal obesity， the proportion of overweight or obese individuals， and a preference for sweet food （all P<0.05）. Multivariate logistic regression analysis revealed a significant correlation between preference for sweet food and the occurrence of PCOS （OR=4.858， 95%CI=1.199‒19.675，P=0.027）， as well as a significant correlation with PCOS accompanied by abdominal obesity （OR=7.083， 95%CI=0.773‒64.937， P=0.048）. Among the female college students surveyed， 37.90% had never heard of PCOS， 51.62% were only familiar with the name of the disease， and 10.47% had attempted to search for PCOS-related information. ConclusionThe prevalence of PCOS among female college students should not be overlooked and unhealthy dietary habits may be a crucial factor contributing to the occurrence of PCOS during this period. Early screening for PCOS during puberty is crucial.