ObjectiveTo observe the changes in the levels of different subtypes of epidermal growth factor receptor （ErbB）， namely ErbB1， ErbB2， ErbB3， and ErbB4， in the spinal dorsal horn of inflammatory pain model rats， and to explore their mechanism of mediating hyperalgesia as well as the intervention mechanism of electroacupuncture at "Zusanli （ST 36）" and "Kunlun （BL 60）". MethodsThe study was divided into five parts. In experiment 1， 14 Sprague Dawley （SD） rats were randomly divided into control and inflammatory pain group （7 rats each group） to observe the pain behavior and the protein expression of different ErbB receptor subtypes in the spinal dorsal horn. In experiment 2， 30 rats were randomly divided into control group 1， inflammatory pain group 1， and low-， medium-， and high-concentration TX1-85-1 groups， with 6 rats in each group， to observe the effect of inhibiting spinal ErbB3 on inflammatory pain. In experiment 3， 12 rats were randomly divided into control virus group and ErbB3 knockdown virus group， with 6 rats in each group， to observe the effect of knocking down ErbB3 in the spinal dorsal horn on inflammatory pain. In experiment 4， 44 rats were randomly divided into control group 2， inflammatory pain group 2， electroacupuncture group， and sham electroacupuncture group， with 11 rats in each group， to observe the effect of electroacupuncture. In experiment 5， 40 rats were randomly divided into control group 3， inflammatory pain group 3， electroacupuncture group 1， and electroacupuncture + NRG1 group， with 10 rats in each group， to observe the effect of activating ErbB3 on electroacupuncture. A rat model of inflammatory pain was established by subcutaneous injection of 100 μl of complete Freund's adjuvant into the sole of the unilateral hind foot of SD rats. Rats in the low-， medium-， and high-concentration TX1-85-1 groups were intrathecally injected with ErbB3 inhibitor TX1-85-1 on day 5 to day 7 after modeling. Rats in the ErbB3 knockdown virus group were injected with ErbB3 knockdown virus packaged with adenovirus vector-based short hairpin RNA （shRNA） into the spinal dorsal horn in situ 3 weeks before modeling. Rats in each electroacupuncture group received electroacupuncture at bilateral "Zusanli （ST 36）" and "Kunlun （BL 60）" from day 1 to day 7 after modeling， with dense-sparse waves at a frequency of 2 Hz/100 Hz and a current of 0.5-1.5 mA for 30 minutes once a day. Rats in the electroacupuncture + NRG1 group were intrathecally injected with ErbB3 ligand recombinant human neuregulin-1 （NRG1） after electroacupuncture intervention from day 5 to day 7 after modeling. The mechanical withdrawal threshold and thermal withdrawal latency of rats were measured on day 1， 3， 5， and 7 after modeling to evaluate behavior， and Western Blot was used to detect the protein and phosphorylation levels of each ErbB subtype in the spinal dorsal horn. ResultsCompared with the control group， rats in the inflammatory pain group showed decreased mechanical withdrawal threshold and thermal withdrawal latency of rats， and increased expression of phosphorylated ErbB3 （p-ErbB3） protein in the spinal dorsal horn on days 1， 3， 5， and 7 after modeling （P＜0.01）. On day 5 and day 7 after modeling， compared with the inflammatory pain group 1， the mecha-nical withdrawal threshold and thermal withdrawal latency of rats in the medium- and high-concentration TX1-85-1 groups increased， and the expression of p-ErbB3 protein decreased （P＜0.05）. On day 1， 3， 5， and 7 after modeling， compared with the control virus group， the mechanical withdrawal threshold and thermal withdrawal latency of rats in the ErbB3 knockdown virus group increased （P＜0.05）. On day 5 and day 7 after modeling， compared with the inflammatory pain group 2 and the sham electroacupuncture group， the mechanical withdrawal threshold and thermal withdrawal latency of rats in the electroacupuncture group increased， and the expression of p-ErbB3 protein decreased （P＜0.05）. On day 5 and day 7 after modeling， compared with the electroacupuncture + NRG1 group， the mechanical withdrawal threshold and thermal withdrawal latency of rats in the electroacupuncture group 1 increased （P＜0.05）. ConclusionThe p-ErbB3 in the spinal dorsal horn involved in hyperalgesia in rats with inflammatory pain， and electroacupuncture at "Zusanli （ST 36）" and "Kunlun （BL 60）" can alleviate inflammatory pain by inhibiting the expression of p-ErbB3 protein in the spinal dorsal horn of rats.

ObjectiveTo evaluate the clinical efficacy of Huayu Tongluo Formula （化瘀通络方， HTF） in patients with mid-stage diabetic kidney disease of blood stasis syndrome and explore its potential mechanisms. MethodsA multi-center， randomized， double-blind， placebo-controlled clinical trial was conducted. Ninety patients of mid-stage diabetic kidney disease of blood stasis syndrome were divided into a control group of 46 cases and a treatment group of 44 cases. Both groups received conventional western medicine treatment， the treatment group additionally taking HTF， while the control group taking a placebo of the formula. The treatment was administered once daily for 24 weeks. The primary outcomes included 24-hour urine total protein （24 h-UTP）， serum albumin （Alb）， glycated hemoglobin （HbA1c）， and serum creatinine （Scr）.The secondary outcomes included changes in levels of endothelin-1 （ET-1）， nitric oxide （NO）， vascular endothelial growth factor （VEGF）， and traditional Chinese medicine （TCM） syndrome scores before and after treatment. Clinical efficacy was evaluated based on TCM syndrome scores and overall disease outcomes. Adverse reactions and endpoint events were recorded. ResultsIn the treatment group after treatment， 24 h-UTP， ET-1， and VEGF levels significantly decreased （P＜0.05）， Alb and NO levels significantly increased （P＜0.05）； while the TCM syndrome scores for edema， lumbar pain， numbness of limbs， dark purple lips， dark purple tongue or purpura， and thin， rough pulse all significantly decreased （P＜0.05）. In the control group， no significant changes were observed in any of the indicators after treatment （P＞0.05）.Compared with the control group， the treatment group showed significant reductions in 24 h-UTP， ET-1， and VEGF levels， and increases in Alb and NO levels （P＜0.05）. The TCM syndrome scores for edema， lumbar pain， dark purple tongue or purpura， and thin， rough pulse were all lower in the treatment group than in the control group （P＜0.05）. The total effective rate of TCM syndrome in the treatment group was 59.09% （26/44）， and the overall clinical effective rate was 45.45% （20/44）. In the control group， these rates were 15.22% （7/46） and 8.7% （4/46）， respectively， with the treatment group showing significantly better outcomes （P＜0.05）. A total of 7 adverse events occurred across both groups， with no significant difference （P＞0.05）. No endpoint events occurred during the study. ConclusionOn the basis of conventional treatment of Western medicine， HTF can further reduce urinary protein levels and improve clinical symptoms in patients with mid-stage diabetic kidney disease of blood stasis syndrome. The mechanism may be related to its effects on endothelial function.

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.

Objective To assess the situation and influencing factors of occupational burnout among the staff at the Center for Disease Control and Prevention (CDC) in Sichuan Province. Methods A total of 1 038 CDC staff members in Sichuan Province were selected as the study subjects using the stratified random sampling method. Occupational burnout of the staff was assessed using the Maslach Burnout Inventory General Survey via an online questionnaire. Results The detection rate of occupational burnout was 42.3% (439/1 038). Binary logistic regression analysis result showed that, after controlling for confounding factors such as education level and alcohol consumption, CDC staffs aged at 20-<31, 31-<41, and 41-<51 years were at higher risk of occupational burnout compared with those ≥51 years (all P<0.05). CDC staffs with 5-<10 or ≥10 years of service had higher occupational burnout risk compared with those with <5 years (both P<0.05). CDC staffs with poor or fair health status, irregular diet, and poor sleep quality had higher risk of occupational burnout compared with those healthy, have regular diet, and good sleep quality (all P<0.05). The risk of occupational burnout increased with higher overtime frequency (all P<0.05). Conclusion Occupational burnout among CDC staffs in Sichuan Province is relatively high. Age, years of service, health status, diet, sleep quality, and overtime frequency are key influencing factors.

Consensus method is not only a common technical approach to the formulation of the acupuncture-moxibustion guidelines, but also an important way to form the recommended treatment protocols of acupuncture-moxibustion guidelines. Based on the theory of implementation science, the paper explores the influencing factors of consensus-reaching to acupuncture-moxibustion guidelines, and puts forward methodological suggestions on the consensus method performed in the formulation of acupuncture-moxibustion guidelines, so as to advance the rational application of consensus method and enhance the scientificity and transparency of acupuncture-moxibustion guidelines.
Moxibustion/standards* ; Humans ; Acupuncture Therapy/standards* ; Consensus ; Practice Guidelines as Topic

Recent advances in large models demonstrate significant prospects for transforming the field of medical imaging. These models, including large language models, large visual models, and multimodal large models, offer unprecedented capabilities in processing and interpreting complex medical data across various imaging modalities. By leveraging self-supervised pretraining on vast unlabeled datasets, cross-modal representation learning, and domain-specific medical knowledge adaptation through fine-tuning, large models can achieve higher diagnostic accuracy and more efficient workflows for key clinical tasks. This review summarizes the concepts, methods, and progress of large models in medical imaging, highlighting their potential in precision medicine. The article first outlines the integration of multimodal data under large model technologies, approaches for training large models with medical datasets, and the need for robust evaluation metrics. It then explores how large models can revolutionize applications in critical tasks such as image segmentation, disease diagnosis, personalized treatment strategies, and real-time interactive systems, thus pushing the boundaries of traditional imaging analysis. Despite their potential, the practical implementation of large models in medical imaging faces notable challenges, including the scarcity of high-quality medical data, the need for optimized perception of imaging phenotypes, safety considerations, and seamless integration with existing clinical workflows and equipment. As research progresses, the development of more efficient, interpretable, and generalizable models will be critical to ensuring their reliable deployment across diverse clinical environments. This review aims to provide insights into the current state of the field and provide directions for future research to facilitate the broader adoption of large models in clinical practice.
Humans ; Diagnostic Imaging/methods* ; Precision Medicine/methods* ; Image Processing, Computer-Assisted/methods*