Objective To develop a neural network-based deep learning model for predicting postoperative recurrence in thyroid tumor patients and validate the model with external datasets for providing clinicians with a reliable decision support tool.Methods An artificial neural network structure was adopted in the study,with thyroid tumor data from the SEER database serving as the training set.External validation was conducted with open-source data from the University of California,Irvine(UCIrvine),and the data from 100 patients at a general tertiary hospital in Hunan province.The model's accuracy and reliability in predicting recurrence were evaluated through multiple performance metrics.Results Experimental results showed that the model outperformed Logistic model in recurrence prediction,with accuracy,recall rate,precision and F1 score reaching 0.915 3,0.981 8,0.921 1 and 0.947 4 in internal validation.Moreover,the model achieved accuracies,recall rates,precisions,F1 scores and ROC_AUC values of 0.832 9,0.945 5,0.841 4,0.890 4 and 0.78 on the UCIrvine validation set,while 0.870 0,0.880 0,0.862 7,0.871 3 and 0.80 on the local validation set.Conclusion This neural network-based predictive model exhibits excellent performance in thyroid tumor recurrence prediction,providing clinicians with a valuable decision support tool that can help optimize postoperative treatment plans and improve patient prognosis management.

Objective:To report the blood group antigen and antibody specificity identification methods for a patient with high-frequency antibodies, and the process of finding and providing compatible blood for the patient.Methods:A patient sent from the Blood Transfusion Department of Shanxi Provincial People′s Hospital to Taiyuan Blood Center in November 2022 was selected for the study. Classical serological methods were used to determine the patient′s blood type, screen for unexpected antibodies, identify antibodies, and perform crossmatching. High-frequency antibody identification was carried out using red blood cells treated with various enzymes. Blood group genotyping was conducted using Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (MALDI-TOF) and Sanger sequencing. Multiple strategies were employed to address the patient′s blood source problem. The study was approved by the Medical Ethics Committee of Taiyuan Blood Center [Ethics No. 2024 Ethics Review No.(2)].Results:①The patient′s blood type was B, RhD positive. Initial screening of the patient′s serum with multiple screening cells and antibody identification cells in saline medium was negative, but positive in antiglobulin medium. The patient′s serum showed varying reaction intensities with red blood cells treated with different enzymes. ②MALDI-TOF mass spectrometry and Sanger sequencing revealed a homozygous nonsense variant c. 376C>T (p.Gln126Ter) in the ABCG2 gene, resulting in the Jr(a-) phenotype. During family donor selection, the patient′s son was found to have a heterozygous variant c. 376C>T (p.Gln126Ter), and another heterozygous variant c. 421C>A (p.Gln141Lys), which predicted a Jr(a+ w) phenotype. ③Crossmatch tests confirmed the compatibility of blood from the patient′s son, which was used to address the urgent blood requirement. Later, rare blood from a Jr(a-) donor from the Guangzhou Blood Center was used for the patient′s ongoing treatment, saving the patient′s life. Conclusion:Combining classic serological testing with blood group gene typing techniques successfully identified the rare Jr(a-) blood type and high-frequency anti-Jra antibodies. Enzyme-treated red blood cell identification methods confirmed the presence of anti-Jra antibodies. By searching within the family and seeking help from other blood centers, compatible blood was found. This approach may provide insights for resolving similar complex blood matching problems in the future.

OBJECTIVE: To report the blood group antigen and antibody specificity identification methods for a patient with high-frequency antibodies, and the process of finding and providing compatible blood for the patient. METHODS: A patient sent from the Blood Transfusion Department of Shanxi Provincial People's Hospital to Blood Transfusion Technology Research Laboratory of Taiyuan Blood Center in November 2022 was selected for the study. Classical serological methods were used to determine the patient's blood type, screen for unexpected antibodies, identify antibodies, and perform crossmatching. High-frequency antibody identification was carried out using red blood cells treated with various enzymes. Blood group genotyping was conducted using Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (MALDI-TOF) and Sanger sequencing. Multiple strategies were employed to address the patient's blood source problem. The study was approved by the Medical Ethics Committee of Taiyuan Blood Center [Ethics No. 2024 Ethics Review No.(2)]. RESULTS: The patient's blood type was B, RhD positive. Initial screening of the patient's serum with multiple screening cells and antibody identification cells in saline medium was negative, but positive in antiglobulin medium. The patient's serum showed varying reaction intensities with red blood cells treated with different enzymes. MALDI-TOF mass spectrometry and Sanger sequencing revealed a homozygous nonsense variant c.376C>T (p.Gln126Ter) in the ABCG2 gene, resulting in the Jr(a-) phenotype. During family donor selection, the patient's son was found to have a heterozygous variant c.376C>T (p.Gln126Ter), and another heterozygous variant c.421C>A (p.Gln141Lys), which predicted a Jr(a+w) phenotype. Crossmatch tests confirmed the compatibility of blood from the patient's son, which was used to address the urgent blood requirement. Later, rare blood from a Jr(a-) donor from the Guangzhou Blood Center was used for the patient's ongoing treatment, saving the patient's life. CONCLUSION Combining classic serological testing with blood group gene typing techniques successfully identified the rare Jr(a-) blood type and high-frequency anti-Jra antibodies. Enzyme-treated red blood cell identification methods confirmed the presence of anti-Jra antibodies. By searching within the family and seeking help from other blood centers, compatible blood was found. This approach may provide insights for resolving similar complex blood matching problems in the future.
Humans ; Blood Grouping and Crossmatching/methods* ; Blood Group Antigens/immunology* ; Blood Transfusion ; Male ; Isoantibodies/blood* ; Female ; Genotype

Objective To investigate the prevalence of metabolic associated fatty liver disease(MAFLD)and its correlation with metabolic components among personnel on tropical islands.Methods The data of personnel who received health examination on islands in 2024 were analyzed,and they were grouped with the age limit of 30 years old to compare the detection rates of MAFLD and metabolic components in different age groups.In people aged≥ 30 years old,the age,gender,body mass index(BMI),waist circumference(WC),fasting blood glucose,blood lipids,liver function,kidney function and other indexes were compared between MAFLD and non-MAFLD groups.Univariate and multivariate logistic regression models were conducted to analyze the factors affecting the occurrence of MAFLD.The effects of various metabolic components on the risk of MAFLD in different age groups were analyzed by subgroup analyses.Results Among 1213 personnel,175(14.4％)cases had MAFLD,of which 141(80.6％)cases were mild,32(18.3％)were moderate,and 2(1.1％)were severe.The detection rates of MAFLD(25.6％[74/289]vs 10.9％[101/924])and overweight/obesity(55.7％[161/289]vs 37.7％[348/924])in age ≥ 30 years old were significantly higher than those in age＜30 years old(both P＜0.001).In people aged≥ 30 years old,compared with the non-MAFLD group,the BMI,WC,systolic blood pressure,diastolic blood pressure,triglyceride(TG),low density lipoprotein-cholesterol,alanine transaminase,aspartate transaminase,gamma glutamyltransferase and uric acid(UA)in the MAFLD group were significantly higher(all P＜0.05),and the high density lipoprotein-cholesterol(HDL-C)was significantly lower(P＜0.05).There were no significant differences in age,gender,fast blood glucose,total cholesterol,alkaline phosphatase,total bilirubin,serum creatinine,or blood urea nitrogen(all P＞0.05).Logistic regression analysis showed that WC was an independent risk factor for MAFLD(odds ratio[OR]=1.101,95％confidence interval[95％CI]1.030-1.176,P=0.004);HDL-C was an independent protective factor for MAFLD(OR=0.071,95％CI0.016-0.323,P=0.001);and BMI ≥24.0 kg/m2 and WC≥90 cm were positively correlated with MAFLD(both P＜0.01).In people aged≥30 years old,the risk of MAFLD was increased in those with overweight/obesity,arterial blood pressure≥ 130/85 mmHg(1 mmHg=0.133 kPa),TG≥1.7 mmol/L,HDL-C≤1.0 mmol/L and UA＞420 μmol/L(all P＜0.05),and the risk of MAFLD was most significantly increased in overweight/obesity people(hazard ratio[HR]=5.088,95％CI 2.724-9.504,P＜0.001).Among people aged＜30 years old,the risk of MAFLD was increased in those with overweight/obesity and UA＞420 μmol/L(both P＜0.01),and the risk of MAFLD was most significantly increased in overweight/obesity individuals(HR=6.305,95％CI3.973-10.006,P＜0.001).Conclusion The detection rates of MAFLD and various metabolic components are higher in the personnel on tropical islands,and the risk of MAFLD is higher in those with overweight/obesity,TG≥1.7 mmol/L and hyperuricemia.

Objective:To report the blood group antigen and antibody specificity identification methods for a patient with high-frequency antibodies, and the process of finding and providing compatible blood for the patient.Methods:A patient sent from the Blood Transfusion Department of Shanxi Provincial People′s Hospital to Taiyuan Blood Center in November 2022 was selected for the study. Classical serological methods were used to determine the patient′s blood type, screen for unexpected antibodies, identify antibodies, and perform crossmatching. High-frequency antibody identification was carried out using red blood cells treated with various enzymes. Blood group genotyping was conducted using Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (MALDI-TOF) and Sanger sequencing. Multiple strategies were employed to address the patient′s blood source problem. The study was approved by the Medical Ethics Committee of Taiyuan Blood Center [Ethics No. 2024 Ethics Review No.(2)].Results:①The patient′s blood type was B, RhD positive. Initial screening of the patient′s serum with multiple screening cells and antibody identification cells in saline medium was negative, but positive in antiglobulin medium. The patient′s serum showed varying reaction intensities with red blood cells treated with different enzymes. ②MALDI-TOF mass spectrometry and Sanger sequencing revealed a homozygous nonsense variant c. 376C>T (p.Gln126Ter) in the ABCG2 gene, resulting in the Jr(a-) phenotype. During family donor selection, the patient′s son was found to have a heterozygous variant c. 376C>T (p.Gln126Ter), and another heterozygous variant c. 421C>A (p.Gln141Lys), which predicted a Jr(a+ w) phenotype. ③Crossmatch tests confirmed the compatibility of blood from the patient′s son, which was used to address the urgent blood requirement. Later, rare blood from a Jr(a-) donor from the Guangzhou Blood Center was used for the patient′s ongoing treatment, saving the patient′s life. Conclusion:Combining classic serological testing with blood group gene typing techniques successfully identified the rare Jr(a-) blood type and high-frequency anti-Jra antibodies. Enzyme-treated red blood cell identification methods confirmed the presence of anti-Jra antibodies. By searching within the family and seeking help from other blood centers, compatible blood was found. This approach may provide insights for resolving similar complex blood matching problems in the future.

Objective To develop a neural network-based deep learning model for predicting postoperative recurrence in thyroid tumor patients and validate the model with external datasets for providing clinicians with a reliable decision support tool.Methods An artificial neural network structure was adopted in the study,with thyroid tumor data from the SEER database serving as the training set.External validation was conducted with open-source data from the University of California,Irvine(UCIrvine),and the data from 100 patients at a general tertiary hospital in Hunan province.The model's accuracy and reliability in predicting recurrence were evaluated through multiple performance metrics.Results Experimental results showed that the model outperformed Logistic model in recurrence prediction,with accuracy,recall rate,precision and F1 score reaching 0.915 3,0.981 8,0.921 1 and 0.947 4 in internal validation.Moreover,the model achieved accuracies,recall rates,precisions,F1 scores and ROC_AUC values of 0.832 9,0.945 5,0.841 4,0.890 4 and 0.78 on the UCIrvine validation set,while 0.870 0,0.880 0,0.862 7,0.871 3 and 0.80 on the local validation set.Conclusion This neural network-based predictive model exhibits excellent performance in thyroid tumor recurrence prediction,providing clinicians with a valuable decision support tool that can help optimize postoperative treatment plans and improve patient prognosis management.

Objective To explore the application value of chromosomal microarray analysis(CMA)technolo-gy in children with abnormal development at the endocrine clinic,and to summarize the data of diagnosis and treatment. Methods A retrospective analysis of 15 children with abnormal development was performed at the endocrinology clinic of Guangzhou Women and Childrenˊs Medical Center from January 2015 to December 2017. The whole genome CMA was applied according to the standard operation procedure of CytoScan 750 arrays of Affymetrix,USA. The results were analyzed by chromosome analysis suite( CHAS)software and related bioinformatics methods. Results The report on CMA showed that the genomes of 15 children had the pathogenic copy number variation(CNVs)or variants of uncer-tain significance. The chromosomal abnormalities were consistent with the clinical manifestations of all children. There were deletions in 14 cases and duplications in 3 cases. Among the 15 cases,loss of heterozygosity was found in 2 cases, uniparental disomy in 1 case,trisomy in 2 cases,Turner syndrome in 2 cases,Smith-Magenis syndrome in 1 case,and wolf Hirschhorn syndrome in 1 case. Only 2 of 15 children were diagnosed as chromosomal abnormalities by routine kar-yotype analysis. Conclusions The whole genome high resolution CMA can significantly improve the rate of diagnosis in children with abnormal development at endocrinology clinic,and is worthy of recommendation.

Objectives To explore the clinical and gene mutation characteristics of Gitelman syndrome in children. Method The clinical data of 3 children with Gitelman syndrome were retrospectively analyzed. Results All three cases were male and their age were 3, 8 and 10 years . The clinical manifestations were hypokalemia, hypomagnesemia, alkalosis, hyperreninemia,and hyperaldosteronemia.Gene detection revealed a complex heterozygous mutation in the SLC12A3 gene.A total of 5 mutation sites were found in the SLC12A3 gene,c.179C>T(Thr60Met),c.248 G>A(Arg83Gln),c.2129 C>A(Ser710X), c.2660+1G>A, c.1456G>A (Asp486Asn). After the diagnosis was confirmed, they were treated with potassium supplement, magnesium supplement, and spironolactone and the conditions were improved in all cases. Conclusions In children with hypokalemia, be aware of Gitelman syndrome, and gene detection is helpful for the diagnosis.

Objective To investigate the correlation of previous hepatitis B virus (HBV) infection with the incidence risk of chronic pancreatitis (CP).Methods This was a case control study.Five hundred and seventy-one patients with CP admitted in the Department of Gastroenterology, Changhai Hospital, Second Military Medical University between January 2015 and October 2016 were enrolled, and 1216 sex and age matched health individuals were also enrolled as the control group.The 5 serum HBV markers(HBsAg,HBsAb, HBeAg, HBeAb and HBcAb) were detected and their correlation with CP incidence was analyzed.Results The positive rate of HBsAg in the CP group and the control group were 3.0% and 3.8%, respectively, and the difference was statistical significant.(OR=0.039, 95% CI 0.02~0.80, P<0.00), but in all the HBsAg positive models (HBVM) the difference of CP and control groups was not statistical significant.HBsAb positive rate in CP group and the control group were 51.8% and 75.0%, respectively, and the difference was statistical significant(P<0.000).HBeAg positive rate in CP group and the control group were 1.1% and 0.1%, the difference was statistical significant (P<0.05), but in all the HBeAg positive models, the CP group and the control group had no statistical difference (P>0.05).The positive rate of HBeAb in the CP group and the control group were 24.3% and 10.8%, respectively, and the difference was statistical significant(P<0.00).The positive rate of HBcAb in the CP group and the control group were 50.1% and 16.5%, respectively,and the difference was statistical significant(P<0.000).In the(HBsAb+, HBeAb+, HBcAb+), (HBsAb+, HBcAb+), (HBeAb+, HBcAb+), (HBcAb+) models, the positive rate in CP group was significantly higher than that of the control group(P<0.000).Multivariate regression analysis showed that the positivity of HBsAb and HBeAb were the protection factors for the occurrence of CP(P<0.05),and HBcAb positivity was the independent risk factor for CP (OR=6.931,P<0.000).Conclusions HBsAb and HBeAb poitivity were the protectors for CP, while HBcAb positivity could be considered as an independent risk factor for CP.

Objective To investigate the effect of pulmonary surfactant on neonatal with acute respiratory distress syndrome.Methods98 cases with neonatal respiratory distress syndrome inthe fourth hospital of Ningbo were randomly divided into experimental group and control group, 49 cases in each group.All patients were given warm, anti infection, and maintain the internal environment stability, prevention of bleeding and other conventional treatment.The control group were treated with mechanical CPAP, the experimental group were given pulmonary surfactant 70mg/kg, concentration is 35mg/mL.Pulmonary surfactant was injected slowly by tracheal intubation at supine, lateral(left, right) and semi recumbent position.The drug was distributed evenly in the lung of the patients who were given the drug 1-2 times.Respiratory frequency (RR), pH, oxygen partial pressure (PaO2), partial pressure of carbon dioxide (PaCO2) levels and the incidence of complications, clinical effective rate of the tthe two groups were observed and compared.ResultsCompared with pre-treatment, RR and PaCO2 levels were decreased, pH and PaO2 levels were increased after treatment in the two groups, the differences were statistically significant (P<0.05);compared with the control group, RR, PaCO2 level were lower, pH and PaO2 levels were higher in the experimental group, the differences has statistical significance (P<0.05);compared with the control group, the experimental groupwith a low incidence of complications, clinical effective rate is higher, the difference was statistically significant (P<0.05).ConclusionPulmonary surfactant can reduce the respiratory frequency in neonatal with acute respiratory distress syndrome, improve arterial blood gas levels, which get better clinical curative effect.