[Objective] To perform genetic analysis on samples with weak agglutination and mixed agglutination of ABO blood group antigens in neonates, and to investigate the molecular biological characteristics of ABO subtypes in neonates. [Methods] Serological identification of ABO blood group was performed by tube method and microcolumn gel method. The ABO exons 2-7 were amplified by PCR, and the amplified products were sequenced by Sanger sequencing method to determine the genotype. [Results] Among the ABO blood group serological results of 14 neonates, 8 cases showed weakened A antigen, and 6 cases showed weakened B antigen. Seven samples were identified with ABO subtype alleles, with genotypes as A102/B101+c.538C>T, Aw26/B102, A205/O02, A205/B101(2 cases), Aw26/O02, B(A)06/O01, B101/O01(3 cases), A102/O01(2 cases), A102/B101 and B101/O02. Additionally, three other family members were also found to carry B(A)06 allele in a pedigree investigation. [Conclusion] For samples showing weakened antigens in ABO blood type identification of neonates, it is necessary to consider the possibility of ABO subtype in addition to age factors, and genetic testing can be used to prevent missed detection of ABO subtypes in neonates.

ObjectiveTo investigate the effect of laminin subunit α3 （LAMA3） on the epithelial-mesenchymal transition （EMT）， invasion， and metastasis abilities of pancreatic cancer （PC）. MethodsA comprehensive analysis was performed for tumor- and EMT-related databases to identify the EMT genes associated with PC， especially LAMA3. The methods of qRT-PCR and Western blot were used to measure the expression level of LAMA3 in PC tissue and cell lines； immunofluorescence assay was used to determine the localization of LAMA3 in PANC-1 cells； Transwell assay was used to investigate the effect of LAMA3 on the invasion and migration abilities of PC cells. The t-test was used for comparison of continuous data between groups. ResultsThe analysis of the TCGA database identified 3 EMT-related oncogenes for PC， i.e.， LAMA3， AREG， and SDC1. The LASSO-Cox regression model showed that LAMA3 had the most significant impact on the prognosis of PC （risk score=0.256 1×LAMA3+0.043 1×SDC1+0.071 4×AREG）. The Cox model and nomogram showed that the high expression of LAMA3 was an independent risk factor for the poor prognosis of PC （hazard ratio=1.32， 95% confidence interval： 1.07‍ ‍—‍ 1.62， P<0.01）. Experimental results showed that there was a significant increase in the expression of LAMA3 in pancreatic cancer tissue compared with the normal pancreatic tissue. Compared with the HPDE cell line， there were varying degrees of increase in the expression of LAMA3 in pancreatic cancer AsPC-1， BxPC-3， PANC-1， MIA PaCa-2， and SW1990 cell lines， with the highest expression level in PANC-1 cells. The enrichment analysis showed that LAMA3 was associated with the biological processes and signaling pathways such as EMT， collagen metabolism， extracellular matrix degradation， the TGF-β pathway， and the PI3K pathway. After the knockdown of LAMA3， there were significant reductions in the expression levels of N-Cadherin， Vimentin， and Snail， while there was a significant increase in the expression level of E-Cadherin. Transwell assay showed that there were significant reductions in the invasion and migration abilities of PANC-1 cells after the knockdown of LAMA3. ConclusionLAMA3 is highly expressed in PC and can promote the EMT， invasion， and migration of PC cells， and therefore， LAMA3 may be used as a novel diagnostic marker and a new therapeutic target for PC.

Objective To introduce an innovative technique, the "balance-shaped sternal elevation device" and its application in the subxiphoid uniportal video-assisted thoracoscopic surgery (VATS) for anterior mediastinal masses resection. Methods Patients who underwent single-port thoracoscopic assisted anterior mediastinal tumor resection through the xiphoid process at the Department of Thoracic Surgery, West China Hospital, Sichuan University from May to June 2024 were included, and their clinical data were analyzed. Results A total of 7 patients were included, with 3 males and 4 females, aged 28-72 years. The diameter of the tumor was 1.9-17.0 cm. The operation time was 62-308 min, intraoperative blood loss was 5-100 mL, postoperative chest drainage tube retention time was 0-9 days, pain score on the 7th day after surgery was 0-2 points, and postoperative hospital stay was 3-12 days. All patients underwent successful and complete resection of the masses and thymus, with favorable postoperative recovery. Conclusion The "balance-shaped sternal elevation device" effectively expands the retrosternal space, providing surgeons with satisfactory surgical views and operating space. This technique significantly enhances the efficacy and safety of minimally invasive surgery for anterior mediastinal masses, reduces trauma and postoperative pain, and accelerates patient recovery, demonstrating important clinical significance and application value.

ObjectiveTo construct a large language model （LLM）-based intelligent pre-consultation system for traditional Chinese medicine （TCM） to improve efficacy of clinical practice. MethodsA TCM large language model was fine-tuned using DeepSpeed ZeRO-3 distributed training strategy based on YAYI 2-30B. A weighted undirected graph network was designed and an agent-based syndrome differentiation model was established based on relationship data extracted from TCM literature and clinical records. An agent collaboration framework was developed to integrate the TCM LLM with the syndrome differentiation model. Model performance was comprehensively evaluated by Loss function， BLEU-4， and ROUGE-L metrics， through which training convergence， text generation quality， and language understanding capability were assessed. Professional knowledge test sets were developed to evaluate system proficiency in TCM physician licensure content， TCM pharmacist licensure content， TCM symptom terminology recognition， and meridian identification. Clinical tests were conducted to compare the system with attending physicians in terms of diagnostic accuracy， consultation rounds， and consultation duration. ResultsAfter 100 000 iterations， the training loss value was gradually stabilized at about 0.7±0.08， indicating that the TCM-LLM has been trained and has good generalization ability. The TCM-LLM scored 0.38 in BLEU-4 and 0.62 in ROUGE-L， suggesting that its natural language processing ability meets the standard. We obtained 2715 symptom terms， 505 relationships between diseases and syndromes， 1011 relationships between diseases and main symptoms， and 1 303 600 relationships among different symptoms， and constructed the Agent of syndrome differentiation model. The accuracy rates in the simulated tests for TCM practitioners， licensed pharmacists of Chinese materia medica， recognition of TCM symptom terminology， and meridian recognition were 94.09%， 78.00%， 87.50%， and 68.80%， respectively. In clinical tests， the syndrome differentiation accuracy of the system reached 88.33%， with fewer consultation rounds and shorter consultation time compared to the attending physicians （P＜0.01）， suggesting that the system has a certain pre- consultation ability. ConclusionThe LLM-based intelligent TCM pre-diagnosis system could simulate diagnostic thinking of TCM physicians to a certain extent. After understanding the patients' natural language， it collects all the patient's symptom through guided questioning， thereby enhancing the diagnostic and treatment efficiency of physicians as well as the consultation experience of the patients.

BACKGROUND:Human periodontal ligament stem cells are potential functional cells for periodontal tissue engineering.However,long-term in vitro culture may lead to reduced stemness and replicative senescence of periodontal ligament stem cells,which may impair the therapeutic effect of human periodontal ligament stem cells. OBJECTIVE:To investigate the effect of oxymatrine on the stemness maintenance and osteogenic differentiation of periodontal ligament stem cells in vitro,and to explore the potential mechanism. METHODS:Periodontal ligament stem cells were isolated from human periodontal ligament tissues by tissue explant enzyme digestion and cultured.The surface markers of mesenchymal cells were identified by flow cytometry.Periodontal ligament stem cells were incubated with 0,2.5,5,and 10 μg/mL oxymatrine.The effect of oxymatrine on the proliferation activity of periodontal ligament stem cells was detected by CCK8 assay.The appropriate drug concentration for subsequent experiments was screened.Western blot assay was used to detect the expression of stem cell non-specific proteins SOX2 and OCT4 in periodontal ligament stem cells.qRT-PCR and western blot assay were used to detect the expression levels of related osteogenic genes and proteins in periodontal ligament stem cells. RESULTS AND CONCLUSION:(1)The results of CCK8 assay showed that 2.5 μg/mL oxymatrine significantly enhanced the proliferative activity of periodontal stem cells,and the subsequent experiment selected 2.5 μg/mL oxymatrine to intervene.(2)Compared with the blank control group,the protein expression level of SOX2,a stem marker of periodontal ligament stem cells in the oxymatrine group did not change significantly(P>0.05),and the expression of OCT4 was significantly up-regulated(P<0.05).(3)Compared with the osteogenic induction group,the osteogenic genes ALP,RUNX2 mRNA expression and their osteogenic associated protein ALP protein expression of periodontal ligament stem cells were significantly down-regulated in the oxymatrine+osteogenic induction group(P<0.05).(4)The oxymatrine up-regulated the expression of stemness markers of periodontal ligament stem cells and inhibited the bone differentiation of periodontal ligament stem cells,and the results of high-throughput sequencing showed that it may be associated with WNT2,WNT16,COMP,and BMP6.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

BACKGROUND: Congenital heart disease (CHD) is a leading cause of birth defect-related mortality. However, more recent CHD mortality data for China are lacking. Additionally, limited studies have evaluated sex, rural-urban, and region-specific disparities of CHD mortality in China. METHODS: We designed a population-based study using data from the Dataset of National Mortality Surveillance in China between 2008 and 2021. We calculated age-adjusted CHD mortality using the sixth census data of China in 2010 as the standard population. We assessed the temporal trends in CHD mortality by age, sex, area, and region from 2008 to 2021 using the joinpoint regression model. RESULTS: From 2008 to 2021, 33,534 deaths were attributed to CHD. The period witnessed a two-fold decrease in the age-adjusted CHD mortality from 1.61 to 0.76 per 100,000 persons (average annual percent change [AAPC] = -5.90%). Females tended to have lower age-adjusted CHD mortality than males, but with a similar decline rate from 2008 to 2021 (females: AAPC = -6.15%; males: AAPC = -5.84%). Similar AAPC values were observed among people living in urban (AAPC = -6.64%) and rural (AAPC = -6.12%) areas. Eastern regions experienced a more pronounced decrease in the age-adjusted CHD mortality (AAPC = -7.86%) than central (AAPC = -5.83%) and western regions (AAPC = -3.71%) between 2008 and 2021. Approximately half of the deaths (46.19%) due to CHD occurred during infancy. The CHD mortality rates in 2021 were lower than those in 2008 for people aged 0-39 years, with the largest decrease observed among children aged 1-4 years (AAPC = -8.26%), followed by infants (AAPC = -7.01%). CONCLUSIONS CHD mortality in China has dramatically decreased from 2008 to 2021. The slower decrease in CHD mortality in the central and western regions than in the eastern regions suggested that public health policymakers should pay more attention to health resources and health education for central and western regions.
Humans ; Heart Defects, Congenital/mortality* ; Male ; Female ; China/epidemiology* ; Infant ; Child, Preschool ; Adult ; Child ; Adolescent ; Infant, Newborn ; Middle Aged ; Young Adult ; Aged ; Rural Population

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.