Objective:To preliminarily establish a sample clot warning model for coagulation screening tests using 5 machine learning methods.Methods:This cross-sectional study collected 7 401 routine screening test samples from Peking Union Medical College Hospital, Chinese Academy of Medical Sciences, from January 1st, 2015, to August 18th, 2024, including 4 786 clotted (positive) and 2 615 qualified (negative) samples for model development. The dataset was divided into Dataset 1 and Dataset 2 based on a reagent change for APTT in December 2018, with separate models developed for each. An additional 2 493 samples (October 31st to November 8th, 2024) were used to evaluate consistency between the model and manual assessment, while 23 200 samples (October 17th to December 31st, 2024) were used for assessing real-world predictive performance. Five machine learning algorithms were employed to develop the clot prediction model: logistic regression (LR), random forest (RF), extreme gradient boosting (XGBoost), naive bayes (NB), and artificial neural network (ANN), with the ANN model constructed using two different hidden layer and neuron parameter settings. Model selection was based on AUC, accuracy, sensitivity, specificity, F1-score, PPV, and NPV, with the optimal model integrated into the LIS for validation.Results:Among the six models using 5 machine learning algorithms, XGBoost demonstrated the highest performance (AUC=0.961, sensitivity=0.945, F1-score=0.934) and robustness to reagent changes ( Z=-1.333, P=0.113). When deployed, the differences between the model's predictions and manual pre-judgment were statistically significant ( Z=-5.289 to 8.933, all P<0.01). The predictive efficacy indices AUC (95% CI), sensitivity, specificity, and accuracy of the XGBoost model deployed in real-world operation of the LIS were 0.939 (0.918—0.960), 0.958, 0.921, and 0.921 respectively. Conclusion:In this study, a clot warning model for coagulation screening samples was established based on the XGBoost algorithm, and its prediction efficacy is good, providing a foundation for intelligent pre-analytical quality control for coagulation screening tests.

Objective:To preliminarily establish a sample clot warning model for coagulation screening tests using 5 machine learning methods.Methods:This cross-sectional study collected 7 401 routine screening test samples from Peking Union Medical College Hospital, Chinese Academy of Medical Sciences, from January 1st, 2015, to August 18th, 2024, including 4 786 clotted (positive) and 2 615 qualified (negative) samples for model development. The dataset was divided into Dataset 1 and Dataset 2 based on a reagent change for APTT in December 2018, with separate models developed for each. An additional 2 493 samples (October 31st to November 8th, 2024) were used to evaluate consistency between the model and manual assessment, while 23 200 samples (October 17th to December 31st, 2024) were used for assessing real-world predictive performance. Five machine learning algorithms were employed to develop the clot prediction model: logistic regression (LR), random forest (RF), extreme gradient boosting (XGBoost), naive bayes (NB), and artificial neural network (ANN), with the ANN model constructed using two different hidden layer and neuron parameter settings. Model selection was based on AUC, accuracy, sensitivity, specificity, F1-score, PPV, and NPV, with the optimal model integrated into the LIS for validation.Results:Among the six models using 5 machine learning algorithms, XGBoost demonstrated the highest performance (AUC=0.961, sensitivity=0.945, F1-score=0.934) and robustness to reagent changes ( Z=-1.333, P=0.113). When deployed, the differences between the model's predictions and manual pre-judgment were statistically significant ( Z=-5.289 to 8.933, all P<0.01). The predictive efficacy indices AUC (95% CI), sensitivity, specificity, and accuracy of the XGBoost model deployed in real-world operation of the LIS were 0.939 (0.918—0.960), 0.958, 0.921, and 0.921 respectively. Conclusion:In this study, a clot warning model for coagulation screening samples was established based on the XGBoost algorithm, and its prediction efficacy is good, providing a foundation for intelligent pre-analytical quality control for coagulation screening tests.

As a threat to human health， steroid-induced osteonecrosis of femur head is a common refractory orthopedic disease mainly caused by glucocorticoids， with poor prognosis and unclear pathogenesis. Osteogenesis-associated signaling pathways play an important role in bone formation. Glucocorticoid-induced abnormal activation and transport of these signaling pathways lead to abnormal differentiation of bone marrow mesenchymal stem cells， dysfunction of bone metabolism， and osteogenesis disorders， which may be the main reasons for the occurrence and development of steroid-induced osteonecrosis of femur head. Bone formation and remodeling need the participation of bone marrow mesenchymal stem cells， which are stem cells characterized by continuous self-renewal and differentiation. The key to strengthening bone remodeling is to improve the osteogenic differentiation capacity， which is the key point to inhibit bone resorption and prevent bone marrow mesenchymal stem cells from differentiating into osteoclasts. Traditional Chinese medicine （TCM） has been used in the treatment of osteonecrosis in ancient times. It is recorded in the Treasury of Words on Materia Medica （《本草汇编》） that "The deficiency in the lower energizer cannot be tonified without Eucommiae Cortexz.The soreness in lower legs cannot be alleviated without Eucommiae Cortex...The pain in the waist and knee cannot be relieved without Eucommiae Cortex...Tonifying liver and invigorating kidney， Eucommiae Cortex is an essential medicine." This indicates that ancient physicians have already begun to use the liver-tonifying， kidney-invigorating， and sinew-bone-strengthening effects of Eucommiae Cortex for the treatment of osteonecrosis. As the national support for the development of TCM strengthens， increasing studies have been conducted on the TCM prevention and treatment of steroid-induced osteonecrosis of femur head. Studies have suggested that Chinese medicinal herbs can exert a positive effect on the differentiation of bone marrow mesenchymal stem cells by affecting targeted signaling molecules， and promote osteogenesis and bone defect repair， thus combating the occurrence and development of steroid-induced osteonecrosis of femur head. The regulation of osteogenic signaling pathway by Chinese medicines to prevent steroid-induced osteonecrosis of femoral head has become a hot research topic. This article reviews the studies about the prevention and treatment of steroid-induced osteonecrosis of femur head with the active components in Chinese medicinal herbs by regulating osteogenic signaling pathways. We then explore the mechanism of the active components in promoting the differentiation of bone marrow mesenchymal stem cells into osteoblasts and inhibiting their differentiation into osteoclasts to facilitate bone formation， aiming to provide a reference for the further study of treating steroid-induced osteonecrosis of femoral head with Chinese medicinal herbs.

Objective To explore the value of the 7th and 8th edition AJCC TNM staging systems for hepatocellular cancer about disease free survival (DFS) after surgery.Methods Clinical data of hepatocellular cancinoma patients were analyzed retrospectively.The difference of the two staging systems in predicting DFS were compared by Kaplan-Meier analytical method and ROC test.Results Based on AJCC 7th edition,there were 114 phase Ⅰ patients,64 phase Ⅱ patients,18 phase Ⅲ patients,4 phase Ⅳ patients,while based on 8th edition,there were 33 phase ⅠA patients,85 Ⅰ B patients,60 phase Ⅱ patients,18 phase Ⅲ patients and 4 phase Ⅳ patients.There was a significant difference in the survival curve between the two stages (x2 =31.177,40.073,P ＜ 0.01).At the same time,the area under the ROC curve in the 8th edition was better than that in the 7th edition.In addition,in the 8th edition the DFS curve of phase ⅠA was superior to that of phase Ⅰ in 7th edition,and to that of phase ⅠB in the 8th edition (x2 =5.701,P =0.017;x2 =7.865,P =0.005).There was no significant difference between that of phase Ⅰ in the 7th edition and that of phase ⅠB in the 8th edition (~ =0.753,P =0.385).Conclusion The value of the 8th AJCC TNM staging in evaluating postoperative DFS is better than the 7th stage,especially for stage I patients.

Purpose To construct a recombined antitumor peptide and to analyze its bioactivity. Methods Constructing a recombined gene and inserting the pGEX-4T-3 vector. The recombined protein was expressed in E. coli BL21 and purified with Amylose Resin. Then, citrostatin was subjected to the following tests separately: inhibition of endothelial cell proliferation, MTT test of cytotoxicity and inhibition of endothelial cell tube formation on ECMatrix. Results Citrostatin significantly inhibited the proliferation of human endothelial cell ECV304(IC_(50) = 2.28 μmol/L) .It also significantly inhibited the proliferation of human tumor cell 1990 and NCI-H64O(IC_(50) = 9.24,2.74 μmol/L) ,and the inhibitory effect became more marked with the increase of citrostatin concentration. The inhibitory effects of citrostatin on endothelial cell tube formation was also confirmed . Conclusion An antitumor peptide, citrostatin, has been successfully constructed and purified, which showed anti-angiogenesis effect and direct cytotoxic effect on tumor cells.