ObjectiveTo unveil the mechanism of Jianpi Huogu prescription （JPHGP） in ameliorating the dyslipidemia of steroid-induced osteonecrosis of the femur head （SONFH） by oxylipidomics combined with transcriptomics. MethodsSixty SD rats were assigned into normal， model， low-， medium-， and high-dose （2.5， 5， 10 g·kg^-1， respectively） JPHGP， and Jiangushengwan （1.53 g·kg^-1） groups. Lipopolysaccharide was injected into the tail vein at a dose of 20 μg·kg^-1 on days 1 and 2， and methylprednisolone sodium succinate was injected at a dose of 40 mg·kg^-1 into the buttock muscle on days 3 to 5. The normal group received an equal volume of normal saline. Drug administration by gavage began 4 weeks after the last injection， and samples were taken after administration for 8 weeks. Hematoxylin-eosin staining was conducted to reveal the histopathological changes of the femoral head， and the number of adipocytes， the rate of empty bone lacunae， and the trabecular area were calculated. Micro-computed tomography was used for revealing the histological and histomorphometrical changes of the femoral head. Enzyme-linked immunosorbent assay was employed to measure the serum levels of triglyceride （TG）， total cholesterol （TC）， low-density lipoprotein （LDL）， high-density lipoprotein （HDL）， apolipoprotein A1 （ApoA1）， and apolipoprotein B （ApoB）. At the same time， the femoral head was collected for oxylipidomic and transcriptomic detection. The differential metabolites and differential genes were enriched and analyzed， and the target genes regulating lipid metabolism were predicted. The predicted target proteins were further verified by molecular docking， immunohistochemistry， and Western blot. ResultsCompared with the normal group， the model group showcased thinning of the femoral head， trabecular fracture， karyopyknosis， subchondral cystic degeneration， increases in the number of adipocytes and the rate of empty bone lacunae （P<0.01）， a reduction in the trabecular area （P<0.01）， decreases in BMD， Tb.Th， Tb.N， and BV/TV， and increases in Tb.Sp and BS/BV （P<0.01）. Compared with the model group， the JPHGP groups showed no obvious thinning of the femoral head or subchondroidal cystic degeneration. The high- and medium-dose JPHGP groups presented declines in the number of adipocytes and the rate of empty bone lacunae， an increase in the trabecular area （P<0.05， P<0.01）， rises in BMD， Tb.Th， Tb.N， and BV/TV， and decreases in Tb.Sp and BS/BV （P<0.05， P<0.01）. Compared with the normal group， the model group showcased raised serum levels of TG， TC， LDL， and ApoB and lowered serum levels of HDL and ApoA1 （P<0.01）. Compared with the model group， the JPHGP groups had lowered serum levels of TG， TC， LDL， and ApoB （P<0.05， P<0.01） and a risen serum level of ApoA1 （P<0.05， P<0.01）. Moreover， the serum level of HDL in the high-dose JPHGP group increased （P<0.01）. A total of 19 different metabolites of disease set and drug set were screened out by oxylipidomics of the femoral head， and 119 core genes with restored expression were detected by transcriptomics. The enriched pathways were mainly concentrated in inflammation， lipids， apoptosis， and osteoclast differentiation. Molecular docking， immunohistochemistry， and Western blot results showed that compared with the normal group， the model group displayed increased content of 5-lipoxygenase （5-LO） and peroxisome proliferator-activated receptor γ （PPARγ） in the femoral head （P<0.01）. Compared with the model group， medium- and high-dose JPHGP reduced the content of 5-LO and PPARγ （P<0.05， P<0.01）. ConclusionJPHGP can restore the levels of oxidized lipid metabolites by regulating the 5-LO-PPARγ axis to treat SONFH in rats. Relevant studies provide experimental evidence for the efficacy mechanism of JPHGP in the treatment of SONFH.

ObjectiveTo unveil the mechanism of Jianpi Huogu prescription （JPHGP） in ameliorating the dyslipidemia of steroid-induced osteonecrosis of the femur head （SONFH） by oxylipidomics combined with transcriptomics. MethodsSixty SD rats were assigned into normal， model， low-， medium-， and high-dose （2.5， 5， 10 g·kg^-1， respectively） JPHGP， and Jiangushengwan （1.53 g·kg^-1） groups. Lipopolysaccharide was injected into the tail vein at a dose of 20 μg·kg^-1 on days 1 and 2， and methylprednisolone sodium succinate was injected at a dose of 40 mg·kg^-1 into the buttock muscle on days 3 to 5. The normal group received an equal volume of normal saline. Drug administration by gavage began 4 weeks after the last injection， and samples were taken after administration for 8 weeks. Hematoxylin-eosin staining was conducted to reveal the histopathological changes of the femoral head， and the number of adipocytes， the rate of empty bone lacunae， and the trabecular area were calculated. Micro-computed tomography was used for revealing the histological and histomorphometrical changes of the femoral head. Enzyme-linked immunosorbent assay was employed to measure the serum levels of triglyceride （TG）， total cholesterol （TC）， low-density lipoprotein （LDL）， high-density lipoprotein （HDL）， apolipoprotein A1 （ApoA1）， and apolipoprotein B （ApoB）. At the same time， the femoral head was collected for oxylipidomic and transcriptomic detection. The differential metabolites and differential genes were enriched and analyzed， and the target genes regulating lipid metabolism were predicted. The predicted target proteins were further verified by molecular docking， immunohistochemistry， and Western blot. ResultsCompared with the normal group， the model group showcased thinning of the femoral head， trabecular fracture， karyopyknosis， subchondral cystic degeneration， increases in the number of adipocytes and the rate of empty bone lacunae （P<0.01）， a reduction in the trabecular area （P<0.01）， decreases in BMD， Tb.Th， Tb.N， and BV/TV， and increases in Tb.Sp and BS/BV （P<0.01）. Compared with the model group， the JPHGP groups showed no obvious thinning of the femoral head or subchondroidal cystic degeneration. The high- and medium-dose JPHGP groups presented declines in the number of adipocytes and the rate of empty bone lacunae， an increase in the trabecular area （P<0.05， P<0.01）， rises in BMD， Tb.Th， Tb.N， and BV/TV， and decreases in Tb.Sp and BS/BV （P<0.05， P<0.01）. Compared with the normal group， the model group showcased raised serum levels of TG， TC， LDL， and ApoB and lowered serum levels of HDL and ApoA1 （P<0.01）. Compared with the model group， the JPHGP groups had lowered serum levels of TG， TC， LDL， and ApoB （P<0.05， P<0.01） and a risen serum level of ApoA1 （P<0.05， P<0.01）. Moreover， the serum level of HDL in the high-dose JPHGP group increased （P<0.01）. A total of 19 different metabolites of disease set and drug set were screened out by oxylipidomics of the femoral head， and 119 core genes with restored expression were detected by transcriptomics. The enriched pathways were mainly concentrated in inflammation， lipids， apoptosis， and osteoclast differentiation. Molecular docking， immunohistochemistry， and Western blot results showed that compared with the normal group， the model group displayed increased content of 5-lipoxygenase （5-LO） and peroxisome proliferator-activated receptor γ （PPARγ） in the femoral head （P<0.01）. Compared with the model group， medium- and high-dose JPHGP reduced the content of 5-LO and PPARγ （P<0.05， P<0.01）. ConclusionJPHGP can restore the levels of oxidized lipid metabolites by regulating the 5-LO-PPARγ axis to treat SONFH in rats. Relevant studies provide experimental evidence for the efficacy mechanism of JPHGP in the treatment of SONFH.

[Objective] : To improve the accuracy and professionalism of question-answering (QA) model in traditional Chinese medicine (TCM) lung cancer by integrating large language models with structured knowledge graphs using the knowledge graph (KG) to text-enhanced retrieval-augmented generation (KG2TRAG) method. [Methods] : The TCM lung cancer model (TCMLCM) was constructed by fine-tuning ChatGLM2-6B on the specialized datasets Tianchi TCM, HuangDi, and ShenNong-TCM-Dataset, as well as a TCM lung cancer KG. The KG2TRAG method was applied to enhance the knowledge retrieval, which can convert KG triples into natural language text via ChatGPT-aided linearization, leveraging large language models (LLMs) for context-aware reasoning. For a comprehensive comparison, MedicalGPT, HuatuoGPT, and BenTsao were selected as the baseline models. Performance was evaluated using bilingual evaluation understudy (BLEU), recall-oriented understudy for gisting evaluation (ROUGE), accuracy, and the domain-specific TCM-LCEval metrics, with validation from TCM oncology experts assessing answer accuracy, professionalism, and usability. [Results] : The TCMLCM model achieved the optimal performance across all metrics, including a BLEU score of 32.15%, ROUGE-L of 59.08%, and an accuracy rate of 79.68%. Notably, in the TCM-LCEval assessment specific to the field of TCM, its performance was 3% − 12% higher than that of the baseline model. Expert evaluations highlighted superior performance in accuracy and professionalism. [Conclusion] TCMLCM can provide an innovative solution for TCM lung cancer QA, demonstrating the feasibility of integrating structured KGs with LLMs. This work advances intelligent TCM healthcare tools and lays a foundation for future AI-driven applications in traditional medicine.

Objective:To develop deep learning models for ulcerative colitis (UC) classification based on Mayo endoscopic score.Methods:A total of 2400 endoscopic images from the Gastrointestinal Endoscopy Centre of the First Affiliated Hospital of Soochow University and the HyperKvasir database were extracted for training classification models, and 200 endoscopic images from Affiliated Jintan Hospital of Jiangsu University were extracted for evaluating the models, both scored by endoscopists according to Mayo endoscopic score (score 0-3). Four deep convolutional neural networks (MobileNetV2, ResNetV2, Xception, EfficientNetV2S), which were pre-trained in the ImageNet database, were used to develop the UC classification models by transfer learning. Models were evaluated in the test set based on the confusion matrix using accuracy, Matthews correlation coefficient (MCC) and Cohen′s kappa, and compared with the performance of senior and junior physicians. Meanwhile, the model was visualized by gradient-weighted class activation mapping.Results:Four deep learning Mayo score models based on UC endoscopic image classification models were successfully developed. In the test set, the accuracy of MobileNetV2, ResNetV2, Xception and EfficientNetV2S was 0.785, 0.800, 0.815, 0.830, respectively (average accuracy 0.808). Amoug them, EfficientNetV2S model was the best, higher than junior physician′s accuracy (accuracy 0.785), and slightly lower than senior physician′s (accuracy 0.870) .Conclusions:The UC endoscopic severity classification models developed by deep learning show good performance, which can be further improved by larger sample size and optimizing the framework.

Diabetes, hypertension, and dyslipidemia, collectively referred to the " Three Highs, " represent increasingly prevalent metabolic risk factors in China. Many individuals experience all three conditions concurrently, significantly heightening the risk of cardiovascular disease and mortality. Although the National Metabolic Management Center(MMC) has been established for over eight years and has its unique features, the awareness, treatment, and control rates of these diseases in China remain low, and the efficiency of community management is insufficient. According to the previous two editions of management guidelines and the most recent domestic and international diagnostic and treatment guidelines, this paper conducts an in-depth analysis of the operational experience and management strategies of the MMC. Its aim is to improve the efficiency of grassroots MMC mode management for " Three Highs" patients and ensure that patients receive more standardized management.

Objective:To develop deep learning models for ulcerative colitis (UC) classification based on Mayo endoscopic score.Methods:A total of 2400 endoscopic images from the Gastrointestinal Endoscopy Centre of the First Affiliated Hospital of Soochow University and the HyperKvasir database were extracted for training classification models, and 200 endoscopic images from Affiliated Jintan Hospital of Jiangsu University were extracted for evaluating the models, both scored by endoscopists according to Mayo endoscopic score (score 0-3). Four deep convolutional neural networks (MobileNetV2, ResNetV2, Xception, EfficientNetV2S), which were pre-trained in the ImageNet database, were used to develop the UC classification models by transfer learning. Models were evaluated in the test set based on the confusion matrix using accuracy, Matthews correlation coefficient (MCC) and Cohen′s kappa, and compared with the performance of senior and junior physicians. Meanwhile, the model was visualized by gradient-weighted class activation mapping.Results:Four deep learning Mayo score models based on UC endoscopic image classification models were successfully developed. In the test set, the accuracy of MobileNetV2, ResNetV2, Xception and EfficientNetV2S was 0.785, 0.800, 0.815, 0.830, respectively (average accuracy 0.808). Amoug them, EfficientNetV2S model was the best, higher than junior physician′s accuracy (accuracy 0.785), and slightly lower than senior physician′s (accuracy 0.870) .Conclusions:The UC endoscopic severity classification models developed by deep learning show good performance, which can be further improved by larger sample size and optimizing the framework.

Diabetes, hypertension, and dyslipidemia, collectively referred to the " Three Highs, " represent increasingly prevalent metabolic risk factors in China. Many individuals experience all three conditions concurrently, significantly heightening the risk of cardiovascular disease and mortality. Although the National Metabolic Management Center(MMC) has been established for over eight years and has its unique features, the awareness, treatment, and control rates of these diseases in China remain low, and the efficiency of community management is insufficient. According to the previous two editions of management guidelines and the most recent domestic and international diagnostic and treatment guidelines, this paper conducts an in-depth analysis of the operational experience and management strategies of the MMC. Its aim is to improve the efficiency of grassroots MMC mode management for " Three Highs" patients and ensure that patients receive more standardized management.

ObjectiveTo investigate the risk factors of fertility behaviors with preterm birth and low birth weight， and to develop a nomogram model to predict the occurrence of low birth weight. MethodsBirth registration information in Shanghai from 2010 to 2020 was collected， and ANOVA and Chi-square tests were used to compare the differences in reproductive behavior factors and newborn health status across time. The odds ratio （OR） value and 95%CI were calculated by a multi-classification logistic regression model to determine the association between reproductive behavior factors and preterm birth or low birth weight infants. A nomogram model was established based on logistic model and the area under the ROC curve was used to assess the effect of the model. ResultsThis analysis included 2 089 384 live newborns. The incidence of full-term low birth weight， preterm normal weight and preterm low birth weight in Shanghai was 0.94%， 2.48% and 2.01%， respectively. From 2010 to 2020， 40.00% women had a history of abortion， the proportion of women who gave birth at age ≥40 years old increased from 1.05% to 2.24%， the proportion of fathers aged ≥40 years increased from 4.79% to 7.48%， and the proportion of women with postgraduate or above increased from 4.81% to 11.74%. The incidence of preterm low birth weight in Shanghai showed an increasing trend over time. Logistic regression analysis showed that the risk of preterm low birth weight was lower in female than in male infants （OR=0.97， 95%CI： 0.95‒0.98）， and the risk of full-term low birth weight was higher than in male infants （OR=1.85， 95%CI： 1.80‒1.90）. The risk of preterm birth and low birth weight was lower for couples of childbearing age with higher education. The risk of preterm low birth weight in newborns tended to increase with maternal age at childbirth >30 years， paternal age ≥40 years， and the number of abortions >2 times. Mother <25 or >35 years， father aged 30‒34 years， and the number of abortions >3 times were the risk factors of full-term low birth weight infants. ConclusionCouples of childbearing age who choose to have children at too high or too low age may increase the risk of preterm birth or low birth weight， so it is necessary to strengthen population awareness and promote age-appropriate childbirth. Multiple abortions are also associated with preterm birth and low birth weight， and it is advisable to popularize the scientific knowledge of contraception and birth control to reduce unnecessary abortions. The nomogram in the study can visualize the risk of full-term and low birth weight infant at different levels of factors， which can assist couples preparing for pregnancy in making decisions about the timing of childbirth and understanding the level of risk.

OBJECTIVE: To establish a machine learning model based on extreme gradient boosting (XGBoost) algorithm for early prediction of severe acute pancreatitis (SAP), and explore its predictive efficiency. METHODS: A retrospective cohort study was conducted. The patients with acute pancreatitis (AP) who admitted to the First Affiliated Hospital of Soochow University, the Second Affiliated Hospital of Soochow University and Changshu Hospital Affiliated to Soochow University from January 1, 2020 to December 31, 2021 were enrolled. Demography information, etiology, past history, and clinical indicators and imaging data within 48 hours of admission were collected according to the medical record system and image system, and the modified CT severity index (MCTSI), Ranson score, bedside index for severity in acute pancreatitis (BISAP) and acute pancreatitis risk score (SABP) were calculated. The data sets of the First Affiliated Hospital of Soochow University and Changshu Hospital Affiliated to Soochow University were randomly divided into training set and validation set according to 8 : 2. Based on XGBoost algorithm, the SAP prediction model was constructed on the basis of hyperparameter adjustment by 5-fold cross validation and loss function. The data set of the Second Affiliated Hospital of Soochow University was served as independent test set. The predictive efficacy of the XGBoost model was evaluated by drawing the receiver operator characteristic curve (ROC curve), and compared it with the traditional AP related severity score; variable importance ranking diagram and Shapley additive explanation (SHAP) diagram were drawn to visually explain the model. RESULTS: A total of 1 183 AP patients were enrolled finally, of which 129 (10.9%) developed SAP. Among the patients from the First Affiliated Hospital of Soochow University and Changshu Hospital Affiliated to Soochow University, there were 786 patients in the training set and 197 in the validation set; 200 patients from the Second Affiliated Hospital of Soochow University were used as the test set. Analysis of all three datasets showed that patients who advanced to SAP exhibited pathological manifestation such as abnormal respiratory function, coagulation function, liver and kidney function, and lipid metabolism. Based on the XGBoost algorithm, an SAP prediction model was constructed, and ROC curve analysis showed that the accuracy for prediction of SAP reached 0.830, the area under the ROC curve (AUC) was 0.927, which was significantly improved compared with the traditional scoring systems including MCTSI, Ranson, BISAP and SABP, the accuracy was 0.610, 0.690, 0.763, 0.625, and the AUC was 0.689, 0.631, 0.875, and 0.770, respectively. The feature importance analysis based on the XGBoost model showed that the top ten items ranked by the importance of model features were admission pleural effusion (0.119), albumin (Alb, 0.049), triglycerides (TG, 0.036), Ca²⁺ (0.034), prothrombin time (PT, 0.031), systemic inflammatory response syndrome (SIRS, 0.031), C-reactive protein (CRP, 0.031), platelet count (PLT, 0.030), lactate dehydrogenase (LDH, 0.029), and alkaline phosphatase (ALP, 0.028). The above indicators were of great significance for the XGBoost model to predict SAP. The SHAP contribution analysis based on the XGBoost model showed that the risk of SAP increased significantly when patients had pleural effusion and decreased Alb. CONCLUSIONS A SAP prediction scoring system was established based on the machine automatic learning XGBoost algorithm, which can predict the SAP risk of patients within 48 hours of admission with good accuracy.
Humans ; Pancreatitis ; Acute Disease ; Retrospective Studies ; Hospitalization ; Algorithms

Acute pancreatitis （AP） is a gastrointestinal disease that requires early intervention， and when it progresses to moderate-severe AP （MSAP） or severe AP （SAP）， there will be a significant increase in the mortality rate of patients. Machine learning （ML） has achieved great success in the early prediction of AP using clinical data with the help of its powerful computational and learning capabilities. This article reviews the research advances in ML in predicting the severity， complications， and death of AP， so as to provide a theoretical basis and new insights for clinical diagnosis and treatment of AP through artificial intelligence.