BACKGROUND:Rapid and accurate segmentation of brain tissue in medical images is of great significance for three-dimensional biomechanical modeling and diagnosis of craniocerebral injuries.Currently,artificial intelligence(AI)-based baseline models exhibit excellent generalization capabilities on large-scale datasets.However,due to the specificity and complexity of craniocerebral tissues,these models have certain limitations in their application to craniocerebral tissue segmentation.Additionally,the scarcity of craniocerebral tissue samples makes it difficult for baseline models to achieve precise segmentation results through fine-tuning.OBJECTIVE:To construct a craniocerebral tissue segmentation model based on texture feature retrieval enhancement to improve segmentation accuracy under a small number of samples.METHODS:Segment Anything in Medical Images(MedSAM)model was selected as the basic framework,and texture features were combined with deep learning to build a brain tissue segmentation model based on texture feature retrieval enhancement(DP-MedSAM).Dice Coefficient and mean intersection over union(MIoU)were selected to evaluate the efficiency of image segmentation results.In comparison with the original MedSAM model,the ablation experiment systematically evaluated the influence of key components on the model performance.The sensitivities of MedSAM,the Segment Anything Model(SAM)for medical image segmentation(SAM-Med2D)and DP-MedSAM in the mandible,left optic nerve,and left parotid gland were compared.RESULTS AND CONCLUSION:(1)By verifying the impact of the number of point prompts on segmentation results on the HaN-Seg dataset,the experimental results indicated that the optimal Dice score was achieved with the addition of three points.(2)DP-MedSAM demonstrated performance improvements compared with MedSAM and SAM-Med2D on two datasets(HaN and Public Domain Database for Computational Anatomy).Especially on the Public Domain Database for Computational Anatomy dataset,in terms of the MIoU metric,DP-MedSAM outperformed MedSAM by 6.59％and SAM-Med2D by 37.35％;in terms of the Dice metric,DP-MedSAM outperformed MedSAM and SAM-Med2D by 4.34％and 25.32％,respectively.(3)The ablation experiment results showed that removing the texture feature extraction module in the DP-MedSAM model,relying solely on original image features,led to a significant decrease in results on the test set.Furthermore,removing the vector cache database and its retrieval enhancement function from the model,which deprived the ability of the model to perform similarity retrieval using an external knowledge base,further reduced model performance.(4)Under conditions of limited data resources,the DP-MedSAM model outperformed the other two models in all evaluation metrics.The DP-MedSAM model performed excellently when processing simple and moderately difficult samples,demonstrating a clear advantage over the other two models and indicating good generalization ability.Processing the fine structures of difficult samples placed higher demands on the model's segmentation capabilities.Although the performance of the DP-MedSAM model declined slightly,it still outperformed the other two models.(5)This study proposes an innovative craniocerebral tissue segmentation model,DP-MedSAM,which improves the baseline model's performance in capturing local details and global structural information in medical images by introducing target region texture feature extraction.Through vector similarity retrieval technology,DP-MedSAM can retrieve the feature vector most similar to the current target region from a pre-constructed vector database,providing more precise guiding information for the segmentation process.

BACKGROUND:Inter-limb asymmetry is a common phenomenon observed during human growth and development.Prolonged specialized training can lead to specific adaptations in inter-limb asymmetry among athletes.OBJECTIVE:To review the formation causes,manifestations,and impacts of inter-limb asymmetry on sports performance,and provide an overview of the relevant assessment methods and intervention strategies.METHODS:A literature search was conducted in the CNKI,WanFang,PubMed,and Web of Science databases from their inception to September 2024.The search terms included"asymmetry,asymmetries,asymmetric,asymmetrical,imbalance,strength,power,force,jump,sprint,athletic performance,anthropometry,injury"in English and Chinese.After excluding duplicate publications,irrelevant content,and conference papers,a total of 131 articles were finally included for analysis.RESULTS AND CONCLUSION:(1)Inter-limb asymmetry can be influenced by various factors including genetics,task demands,training regimens,injuries,fatigue,and limb preference.These factors lead to being primarily manifested in anatomical structure,strength performance,and task-specific asymmetry.(2)An increase in inter-limb asymmetry can result in impaired performance in bilateral in-phase symmetric movements.However,the relationship between increased inter-limb asymmetry and bilateral out-of-phase symmetric movements remains unclear and requires further investigation.(3)Training interventions have been shown to effectively mitigate inter-limb asymmetry,with unilateral training demonstrating superior outcomes compared with bilateral training.The choice of training methods and content should be tailored to meet the specific demands of the sport.(4)To further clarify the relationship between inter-limb asymmetry and athletic performance,it is recommended that future research adopt the concept of"task specificity"in inter-limb asymmetry.This includes standardizing study designs,selecting sensitive testing methods and indicators,unifying calculation methods to provide more high-quality evidence,and establishing categorized warning threshold standards for inter-limb asymmetry in different sports.

Objective: To analyze the demographic characteristics, return donation patterns, and risk of adverse reactions among donors temporarily deferred due to blood pressure abnormalities, so as to provide an evidence-based foundation for optimizing pre-donation blood pressure screening strategies, enhancing donor retention, and ensuring blood supply safety. Methods: Data from 2.939 million donor instances were collected through the Information Management System at the Beijing Red Cross Blood Center between January 2015 and August 2025. The analysis specifically focused on the 11 600 instances of donors temporarily deferred due to abnormal blood pressure, examining demographic characteristics (age, and gender) and donation-related features (number of donations, donation site, and type of donation). Further analysis was conducted on the return donation patterns, including the return rate, time interval to return, and the incidence, type, and severity of adverse reactions among returned donors. Results: Distribution of abnormal blood pressure: Among the 11 600 instances of abnormal blood pressure, the prevalence was significantly higher in males (0.48%, 10 111/2 086 909) than in females (0.17%, 1 465/852 090). The 46-55 age group had the highest prevalence (0.88%, 2 925/329 235), and the differences across age groups were statistically significant. The prevalence was, higher among repeat donors (0.41%, 5 242/1 276 452) than first-time donors (0.38%, 6 334/1 662 547). The prevalence at mobile donation sites outside the blood center (0.06%, 350/596 104) was higher than fixed donor centers (0.50%, 10 225/2 052 290) and group donation drives (0.34%, 1 001/290 608). Return donations: A total of 19.49% (2 256 out of 11 576) deferred donors returner and successfully donated. Among these donors, 36.17% (816 out of 2 256) returned within 7 days, while the highest proportion of returns was observed within 31-182 days (25.44%, 574/2 256). A higher return rate was observed among male donors (20.17%, 2 039/1 0111) compared to female donors (14.81%, 217/1 465). The return rate for repeat donors (43.02%, 2 255/5 242) was significantly higher than that of first-time donors (0.02%, 1/6 334). Individual donors showed a higher return rate (20.95%, 1 986/9 479) than group donors (12.88%, 270/2 097), with all differences being statistically significant (P<0.05). The differences in return rates across age groups were not statistically significant (P>0.05). Adverse reactions: The incidence of adverse reactions after return was 0.09% (2/2 256), significantly lower than the overall adverse reaction incidence during the same period (0.20%, 5 981/2 938 999). Both adverse reactions were local reaction (category A1, pain or bruising at the puncture site), with no reported cases of systemic vasovagal reactions (VR) or severe adverse events. Conclusion: The current blood pressure screening criteria may lead to the unnecessary deferral of eligible donors. The risk of adverse reactions is extremely low among returned donors who were deferred for abnormal blood pressure. A relaxation of the blood pressure screening criteria is therefore suggested, coupled with the optimization of donation site environment and blood pressure measurement procedure with reference to expert consensus to enhance donor retention and blood supply safety.

Objective: To explore the distribution characteristics and influencing factors of adverse reactions in whole blood donation in Jinan, Shandong, so as to provide evidence for the prevention and control of such adverse reactions in this region. Methods: A retrospective analysis was conducted on whole blood donors and adverse reaction cases in Jinan during 2023. To explore influencing factors of adverse reactions, univariate and multivariate logistic regression analyses were used to examine the relationships between adverse reactions and factors such as gender, age, donation organization mode, donation frequency, donation volume, time slot, and health examination results. Results: A total of 122 961 whole blood donations were recorded in Jinan in 2023. Donation-related adverse reactions occurred in 2 054 cases, with an incidence rate of 1.67%. Univariate analysis revealed significant differences in the incidence of adverse reactions across donor characteristics: the rate was higher in females (2.35%, 921/39 192) than in males (1.35%, 1 133/83 769), donors aged 18-25 years had the highest incidence (3.48%, 1 799/51 733), the incidence in group donations (3.13%, 1,737/55 534) was significantly higher than in individual donations (0.47%, 317/67 427), and insufficient blood collection was closely associated with adverse reactions (all P<0.001). Multivariate logistic regression analysis identified group donation, female gender, and a pulse rate of 81-99 beats per minute as risk factors for adverse reactions (all P<0.001), while systolic blood pressure of 116-139 mmHg and diastolic blood pressure of 76-89 mmHg were protective factors (all P<0.05). Compared to younger and lower-weight donor groups, older and higher-weight donors had a significantly lower risk of adverse reactions (all P<0.05). Donors giving 400 mL had a higher risk than those giving 200 mL (P<0.001). In addition, compared with the donation time slot of 7:00-8:59, the risk of adverse reactions was significantly higher during 9:00-16:59, with the time slot of 13:00-14:59 showing the most prominent risk (all P<0.05). However, no statistically significant difference was observed between the time slot of 17:00-20:59 and that of 7:00-8:59 (P>0.05). The primary clinical manifestation of adverse reactions was donation-related vasovagal reaction, with mental tension being the leading precipitating factor, accounting for 69.08% (1 419/2 054) of cases. Conclusion: The occurrence of adverse reactions in whole blood donation in the Jinan is influenced by multiple factors, including donor demographic characteristics, donation organization mode, physiological indicators, and time of donation. It is recommended to enhance the identification and intervention for high-risk groups, and optimize donation processes and service models to reduce the incidence of adverse reactions, thereby ensuring donor safety and blood quality.

AIM:To investigate the current status and influencing factors of knowledge-attitude-practice in myopia prevention and control among children and adolescents in Ningbo City, thereby providing a scientific basis for formulating targeted prevention strategies.METHODS: Children and adolescents aged 6-12 years old were selected from the medical-school collaborative myopia prevention network in Ningbo City between August 2024 and May 2025 using stratified cluster sampling. Information on myopia prevention knowledge(15 items)and practice(9 items)was collected through questionnaire surveys. Logistic regression models were used to analyze factors influencing myopia occurrence in children and adolescents.RESULTS: A total of 664 children and adolescents aged 6-12 years were enrolled in this study. Participants were divided by age into three groups: 6-7 years old(n=221), 8-9 years old(n=221), and 10-12 years old(n=222). Of the 664 questionnaires distributed, 637 valid questionnaires were returned(201 from the 6-7 age group, 235 from the 8-9 age group, and 201 from the 10-12 age group), yielding an effective response rate of 95.9%. Based on myopia screening results, the non-myopic group comprised 203 participants(31.9%), including 100 males and 103 females, with a mean age of 8.82±1.98 years old. The myopic group comprised 434 participants(68.1%), including 213 males and 221 females, with a mean age of 9.10±1.95 years old. The myopia prevalence rates in the 6-7, 8-9, and 10-12 age groups were 37.8%(76/201), 71.9%(169/235), and 94.0%(189/201), respectively(P<0.001). Regarding the knowledge and practice of myopia prevention, the overall awareness rate in the non-myopic group(59.7%±9.7%)was significantly higher than that in the myopic group(48.7%±8.5%; P<0.001). Additionally, the non-myopic group scored higher on the key practice of “regular eye examinations”(4.27±0.96)compared to the myopic group(4.10±1.05; P<0.05). Logistic regression analysis indicated that age was the primary risk factor for myopia occurrence.CONCLUSION: Age is the dominant factor in the onset of myopia, and there is a phenomenon of “knowledge-practice gap”; the traditional health education model has limitations, and a precise prevention and control system based on developmental patterns should be established.

OBJECTIVE To investigate the therapeutic effect and mechanism of Qiwei dongqingye powder （QDP） on bronchial asthma in mice. METHODS The mice were divided into blank group （normal saline）， model group （normal saline）， dexamethasone group （2 mg/kg）， and QDP low-， medium-， and high-dose groups （200， 400， 800 mg/kg）， with 14 mice in each group. Except for the blank group， mice in all other groups were given ovalbumin via intraperitoneal injection followed by aerosol inhalation to induce a bronchial asthma model. During the modeling process， mice in each group were administered corresponding drug solutions or normal saline intragastrically/intraperitoneally. After the last medication， the number of cells in the bronchoalveolar lavage fluid （BALF） of the mice was observed and counted； the pathological changes of the bronchus and lung tissue were observed； the levels of malondialdehyde （MDA）， nitric oxide （NO）， total superoxide dismutase （T-SOD）， and glutathione peroxidase （GSH-Px） in the lung tissue of the mice were determined， and the level of interleukin-17 （IL-17） in the BALF and serum was determined. Transcriptomics was employed to predict and validate the mechanism of action of QDP against bronchial asthma. RESULTS Compared with the model group， the total cell count， neutrophil count， lymphocyte count， and macrophage counts in the BALF of the QDP high-dose group were all significantly reduced （ P ＜0.05）； the levels of MDA and NO in the lung tissue， and the levels of IL-17 in the BALF and serum were all decreased significantly （ P ＜0.05）； the levels of T-SOD and GSH-Px were significantly increased （ P ＜0.05）； the arrangement of lung tissue cells tended to normalize， with reduced infiltration of inflammatory cells and decreased exfoliation of bronchial simple columnar epithelial cells. The transcriptomic results revealed that the differentially expressed genes were B-cell receptor signaling pathway， nuclear factor κB （NF-κB） signaling pathway， ferroptosis signaling pathway， and others. Further validation revealed that， compared with the model group， the expression levels of NF-κB p65 and chemokine ligand 20， as well as the phosphorylation level of NF-κB inhibitor protein α， were significantly decreased in the lung tissues of the mice in all QDP groups （ P ＜0.05）. Conversely， the protein expression of nuclear factor erythroid 2-related factor 2 （Nrf2） and heme oxygenase 1 （HO-1） were significantly increased （ P ＜0.05）. CONCLUSIONS QDP can effectively alleviate bronchial asthma by inhibiting the NF-κB signaling pathway， activating the Nrf2/HO-1 signaling pathway， regulating oxidative stress， and reducing inflammatory responses.

BACKGROUND:Rapid and accurate segmentation of brain tissue in medical images is of great significance for three-dimensional biomechanical modeling and diagnosis of craniocerebral injuries.Currently,artificial intelligence(AI)-based baseline models exhibit excellent generalization capabilities on large-scale datasets.However,due to the specificity and complexity of craniocerebral tissues,these models have certain limitations in their application to craniocerebral tissue segmentation.Additionally,the scarcity of craniocerebral tissue samples makes it difficult for baseline models to achieve precise segmentation results through fine-tuning.OBJECTIVE:To construct a craniocerebral tissue segmentation model based on texture feature retrieval enhancement to improve segmentation accuracy under a small number of samples.METHODS:Segment Anything in Medical Images(MedSAM)model was selected as the basic framework,and texture features were combined with deep learning to build a brain tissue segmentation model based on texture feature retrieval enhancement(DP-MedSAM).Dice Coefficient and mean intersection over union(MIoU)were selected to evaluate the efficiency of image segmentation results.In comparison with the original MedSAM model,the ablation experiment systematically evaluated the influence of key components on the model performance.The sensitivities of MedSAM,the Segment Anything Model(SAM)for medical image segmentation(SAM-Med2D)and DP-MedSAM in the mandible,left optic nerve,and left parotid gland were compared.RESULTS AND CONCLUSION:(1)By verifying the impact of the number of point prompts on segmentation results on the HaN-Seg dataset,the experimental results indicated that the optimal Dice score was achieved with the addition of three points.(2)DP-MedSAM demonstrated performance improvements compared with MedSAM and SAM-Med2D on two datasets(HaN and Public Domain Database for Computational Anatomy).Especially on the Public Domain Database for Computational Anatomy dataset,in terms of the MIoU metric,DP-MedSAM outperformed MedSAM by 6.59％and SAM-Med2D by 37.35％;in terms of the Dice metric,DP-MedSAM outperformed MedSAM and SAM-Med2D by 4.34％and 25.32％,respectively.(3)The ablation experiment results showed that removing the texture feature extraction module in the DP-MedSAM model,relying solely on original image features,led to a significant decrease in results on the test set.Furthermore,removing the vector cache database and its retrieval enhancement function from the model,which deprived the ability of the model to perform similarity retrieval using an external knowledge base,further reduced model performance.(4)Under conditions of limited data resources,the DP-MedSAM model outperformed the other two models in all evaluation metrics.The DP-MedSAM model performed excellently when processing simple and moderately difficult samples,demonstrating a clear advantage over the other two models and indicating good generalization ability.Processing the fine structures of difficult samples placed higher demands on the model's segmentation capabilities.Although the performance of the DP-MedSAM model declined slightly,it still outperformed the other two models.(5)This study proposes an innovative craniocerebral tissue segmentation model,DP-MedSAM,which improves the baseline model's performance in capturing local details and global structural information in medical images by introducing target region texture feature extraction.Through vector similarity retrieval technology,DP-MedSAM can retrieve the feature vector most similar to the current target region from a pre-constructed vector database,providing more precise guiding information for the segmentation process.

BACKGROUND:Inter-limb asymmetry is a common phenomenon observed during human growth and development.Prolonged specialized training can lead to specific adaptations in inter-limb asymmetry among athletes.OBJECTIVE:To review the formation causes,manifestations,and impacts of inter-limb asymmetry on sports performance,and provide an overview of the relevant assessment methods and intervention strategies.METHODS:A literature search was conducted in the CNKI,WanFang,PubMed,and Web of Science databases from their inception to September 2024.The search terms included"asymmetry,asymmetries,asymmetric,asymmetrical,imbalance,strength,power,force,jump,sprint,athletic performance,anthropometry,injury"in English and Chinese.After excluding duplicate publications,irrelevant content,and conference papers,a total of 131 articles were finally included for analysis.RESULTS AND CONCLUSION:(1)Inter-limb asymmetry can be influenced by various factors including genetics,task demands,training regimens,injuries,fatigue,and limb preference.These factors lead to being primarily manifested in anatomical structure,strength performance,and task-specific asymmetry.(2)An increase in inter-limb asymmetry can result in impaired performance in bilateral in-phase symmetric movements.However,the relationship between increased inter-limb asymmetry and bilateral out-of-phase symmetric movements remains unclear and requires further investigation.(3)Training interventions have been shown to effectively mitigate inter-limb asymmetry,with unilateral training demonstrating superior outcomes compared with bilateral training.The choice of training methods and content should be tailored to meet the specific demands of the sport.(4)To further clarify the relationship between inter-limb asymmetry and athletic performance,it is recommended that future research adopt the concept of"task specificity"in inter-limb asymmetry.This includes standardizing study designs,selecting sensitive testing methods and indicators,unifying calculation methods to provide more high-quality evidence,and establishing categorized warning threshold standards for inter-limb asymmetry in different sports.

Bile-processed Coptidis Rhizoma（B-pCR）， first documented in Shengji Zonglu， is a unique processed products of Coptidis Rhizoma（CR） characterized by "mutual enhancement processing" and "enhancing the cold property of cold-natured herbs". Pig bile can enhance the bitter and cold properties of CR， yielding potent effects in purging excess fire from the liver and gallbladder. The processing increases the dissolution of alkaloids such as berberine， coptisine， and palmatine， while introducing bile acids from pig bile， including taurine-type and glycine-type cholic acids. This enhances its pharmacological effects， such as antipyretic activity， regulation of glucose and lipid metabolism disorders， and intestinal absorption. Traditional processing techniques and quality standards for B-pCR are outlined in the Shanghai Traditional Chinese Medicine（TCM） Decoction Pieces Processing Standard and the Gansu TCM Processing Standard. However， incomplete specifications for critical process parameters and quality criteria significantly impact its production and clinical application. A review of research over the past two decades on the processing history， process optimization， quality evaluation， material basis， and changes in pharmacological effects and properties of B-pCR reveals that the pretreatment method and dosage of pig bile， and processing temperature are key factors influencing its quality. Furthermore， current quality standards lack specific indicators. Additionally， the enhancement of the cold property and medicinal efficacy direction of B-pCR is not only associated with changes in alkaloid groups but also depend on the synergistic effects of bile acids. This review can provide insights for improving the quality evaluation system of B-pCR.

Retinal neovascular diseases represent a critical subset of retinal diseases that severely impair vision and can lead to blindness. In recent years, artificial intelligence(AI)has demonstrated breakthrough applications in the medical field, particularly in ophthalmology, leveraging its robust capabilities in image recognition and data analysis. Machine learning and deep learning, as core AI technologies, enable precise feature extraction from vast volumes of medical imaging data and the construction of predictive models, offering novel approaches for the auxiliary diagnosis and prognosis of retinal neovascular diseases. This review synthesizes the latest advancements in AI applications for neovascular retinal diseases, including diabetic retinopathy, retinal vein occlusion, retinopathy of prematurity, and age-related macular degeneration. It further discusses the limitations and challenges in clinical implementation. Through a comprehensive summary and analysis, this review aims to provide insights for advancing AI-driven diagnosis and treatment strategies, ultimately facilitating early detection and predictive management of these vision-threatening diseases.