With the rapid advancement of hemodynamic indices and monitoring technologies, their classification methods and application processes have become increasingly complex. Currently, no unified standard hasbeen established, making it difficult to fully meet the clinical requirements for hemodynamic management. To assist in hemodynamic monitoring assessment and therapeutic decision-making in critically ill patients, the Critical Hemodynamic Therapy Collaborative Group, in conjunction with the Critical Ultrasound Study Group, has jointly developed the Standard for the Application of Hemodynamic Monitoring Techniques in Critical Care. The first part of this standard systematically categorizes hemodynamic indicators into flow indicators, pressure and its derivative indicators, and tissue perfusion indicators, while elaborating on the clinical application of each. The second part establishes a standardized clinical implementation pathway for hemodynamic monitoring. It proposes a tiered monitoring strategy-comprising basic, advanced, indication-specific, and special scenario monitoring-tailored to different clinical settings. It emphasizes the central role of critical care ultrasound across all levels of monitoring and establishes hemodynamic assessment standards for organs such as the brain, kidneys, and gastrointestinal tract. This standard aims to provide a unified framework for clinical practice, teaching, training, and research in critical care medicine, thereby promoting standardized development within the discipline.

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.

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.

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.

The safety of traditional Chinese medicine(TCM) has always been taken very seriously, and rich and valuable theories and experiences have been developed to ensure the safe and precise use of TCM in clinical practices. In recent years, the cognitive theory of toxicity of TCM, has undergone a profound change. TCM is characterized by the existence of intrinsic toxicity, idiosyncratic toxicity, and indirect toxicity related to organic factors. Therefore, the traditional theories and experiences of TCM, which focus on the prevention and control of intrinsic toxicity, fail to be used for the development of risk prevention and control countermeasures for newly discovered TCM with idiosyncratic toxicity and indirect toxicity. Accordingly, based on the toxicity classification and mechanism characteristics of TCM, this paper proposed a new strategy and method in TCM compatibility for detoxification based on componenttarget-effect interaction. The strategy based on component-target-effect interaction is to carry out TCM compatibility for detoxification by blocking the occurrence of drug-mediated damage and promoting damage repair through component interactions, target interactions,and/or effect interactions. Based on this theory, the paper established a strategy for TCM compatibility that aligned with the cognitive theory of toxicity of TCM, so as to achieve safe and precise use of TCM in clinical practices. The strategy based on component-targeteffect interaction has been exemplarily applied to the development of countermeasures to reduce the toxicity of TCM, including Polygonum Multiflorum, Epimedii Folium, and Psoraleae Fructus, and a new mechanism of Glycyrrhizae Radix et Rhizoma to " harmonize various medicines and detoxify myriad poisons" was illustrated, providing a scientific basis for the safe and precise use of TCM in clinical practice. This paper explained the scientific connotation, application forms, and application examples of componenttarget-effect interaction, aiming to provide a theoretical and methodological basis for guaranteeing the precise use of TCM in clinical practice and innovate the theories and methods of TCM compatibility for detoxification.
Drugs, Chinese Herbal/chemistry* ; Humans ; Medicine, Chinese Traditional/methods* ; Animals ; Drug-Related Side Effects and Adverse Reactions/prevention & control*

OBJECTIVE: To explore the biomechanical characteristics and clinical application effects of three-dimensional (3D) printed osteotomy guide plate combined with Ilizarov technique in the treatment of rigid clubfoot. METHODS: A retrospective analysis was performed on the clinical data of 11 patients with rigid clubfoot who met the inclusion criteria and were admitted between January 2019 and December 2024. There were 6 males and 5 females, aged 21-60 years with an average of 43.2 years. Among them, 5 cases were untreated congenital rigid clubfoot, 4 cases were recurrent rigid clubfoot after previous treatment, and 2 cases were rigid clubfoot due to disease sequelae. All 11 patients first received slow distraction using Ilizarov technique combined with circular external fixator until the force lines of the foot and ankle joint were basically normal. Then, 1 male patient aged 24 years was selected, and CT scanning was used to obtain imaging data of the ankle joint and foot. A 3D finite element model was established and validated using the plantar stress distribution nephogram of the patient. After validation, the biomechanical changes of the tibiotalar joint under the same load were simulated after triple arthrodesis and fixation. The optimal correction angle of the hindfoot was determined to fabricate 3D-printed osteotomy guide plates, and all 11 patients underwent triple arthrodesis using these guide plates. The functional recovery was evaluated by comparing the American Orthopaedic Foot and Ankle Society (AOFAS) score, International Clubfoot Study Group (ICFSG) score, and 36-Item Short Form Survey (SF-36) score before and after operation. RESULTS: Finite element analysis showed that the maximum peak von Mises stress of the tibiotalar joint was at hindfoot varus 3° and the minimum at valgus 6°; the maximum peak von Mises stress of the 3 naviculocuneiform joints under various conditions appeared at lateral naviculocuneiform joint before operation, and the minimum appeared at lateral naviculocuneiform joint at neutral position 0°; the maximum peak von Mises stress of the 5 tarsometatarsal joints under various conditions appeared at the 2nd tarsometatarsal joint at hindfoot neutral position 0°, and the minimum appeared at the 1st tarsometatarsal joint at valgus 6°. Clinical application results showed that the characteristics of clubfoot deformity observed during operation were consistent with the preoperative 3D reconstruction model. All 11 patients were followed up 8-24 months with an average of 13.1 months. One patient had postoperative incision exudation, which healed after dressing change; the remaining patients had good incision healing. All patients achieved good healing of the osteotomy segments, with a healing time of 3-6 months and an average of 4.1 months. At last follow-up, the AOFAS score, SF-36 score, and ICFSG score significantly improved when compared with those before operation ( P<0.05). CONCLUSION The 3D-printed osteotomy guide plate combined with Ilizarov technique has favorable biomechanical advantages in the treatment of rigid clubfoot, with significant clinical application effects. It can effectively improve the foot function of patients and achieve precise and personalized treatment.
Humans ; Clubfoot/diagnostic imaging* ; Printing, Three-Dimensional ; Male ; Osteotomy/instrumentation* ; Female ; Ilizarov Technique/instrumentation* ; Adult ; Retrospective Studies ; Biomechanical Phenomena ; Middle Aged ; Bone Plates ; Young Adult ; Finite Element Analysis ; Treatment Outcome ; Ankle Joint/physiopathology* ; Tomography, X-Ray Computed ; External Fixators