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: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.

ObjectiveTo investigate the accuracy of multiple discriminant analysis （MDA） versus fibrosis-4 （FIB-4） in assessing liver fibrosis degree in patients with HBV infection， as well as the possibility of MDA as an indicator for disease progression. MethodsA total of 263 patients with HBV infection who underwent liver biopsy in The First Affiliated Hospital of Guangxi Medical University from April 2010 to April 2024 were included， and their clinical data were collected. According to the results of pathological examination， they were divided into non-significant fibrosis group （F<2） with 126 patients and significant fibrosis group （F≥2） with 137 patients. The correlation of MDA and FIB-4 with liver fibrosis degree was analyzed， and MDA and FIB-4 were compared in terms of their accuracy in assessing significant liver fibrosis. A total of 62 patients completed follow-up， and according to the presence or absence of progression to liver cirrhosis at the last follow-up visit， they were divided into progressive group with 21 patients and non-progressive group with 41 patients； the efficacy of MDA and FIB-4 in diagnosing disease progression was analyzed and compared. The independent-samples t test was used for comparison of normally distributed continuous data between groups， and the Mann-Whitney U test was used for comparison of non-normally distributed continuous data between groups； the Kruskal-Wallis H test was used for comparison between multiple groups， and the Bonferroni method was used for further comparison between two groups. The chi-square test was used for comparison of categorical data. The Spearman’s correlation coefficient was used for correlation analysis. The Wilcoxon signed rank sum test was used for the analysis of baseline data and data at the end of follow-up， and the binary Logistic regression analysis was used to investigate the influencing factors for progression to liver cirrhosis. The receiver operating characteristic （ROC） curve was used to investigate the diagnostic efficacy of indicators， the Z-test was used for comparison of the area under the ROC curve （AUC）， and the paired chi-square test was used for comparison of the sensitivity， specificity， and accuracy of the two indicators. ResultsThe correlation coefficient between FIB-4 and liver fibrosis degree was 0.378， while the correlation coefficient between MDA and liver fibrosis degree was -0.325 （both P<0.001）. FIB-4 had an AUC of 0.688， a sensitivity of 64.96%， a specificity of 68.87%， a positive predictive value of 67.42%， a negative predictive value of 63.36%， an accuracy of 65.40%， and a cut-off value of 1.01， while MDA had an AUC of 0.653， a sensitivity of 52.55%， a specificity of 78.57%， a positive predictive value of 72.73%， a negative predictive value of 60.37%， an accuracy of 65.02%， and a cut-off value of 0.29， suggesting that compared with FIB-4， MDA had a lower sensitivity （P=0.004） and a higher specificity （P=0.001）. The progressive group had a significantly higher age than the non-progressive group at baseline （t=2.611， P=0.011）. For the progressive group， there was an increase in FIB-4 and a reduction in MDA from baseline to the end of follow-up （both P<0.001）， while the non-progressive group showed no significant changes （both P>0.05）. The multivariate Logistic regression analysis showed that aspartate aminotransferase （odds ratio ［OR］=0.940， 95% confidence interval ［CI］： 0.885‍ ‍—‍ ‍0.998， P<0.05） and MDA （OR=0.445， 95%CI： 0.279‍ ‍—‍ ‍0.710， P<0.001） were independent influencing factors for disease progression. MDA had an AUC of 0.893 and an optimal cut-off value of -0.01 in diagnosing the disease progression of liver cirrhosis. ConclusionMDA has a comparable accuracy to FIB-4 in the diagnosis of significant liver fibrosis， and MDA<-0.01 has a high accuracy in diagnosing the progression of liver fibrosis to liver cirrhosis， which can help to reduce the need for liver biopsy in clinical practice.

Quantum dots (QDs), nanoscale semiconductor crystals, have emerged as a revolutionary class of nanomaterials with unique optical and electrochemical properties, making them highly promising for applications in disease diagnosis and treatment. Their tunable emission spectra, long-term photostability, high quantum yield, and excellent charge carrier mobility enable precise control over light emission and efficient charge utilization, which are critical for biomedical applications. This article provides a comprehensive review of recent advancements in the use of quantum dots for disease diagnosis and therapy, highlighting their potential and the challenges involved in clinical translation. Quantum dots can be classified based on their elemental composition and structural configuration. For instance, IB-IIIA-VIA group quantum dots and core-shell structured quantum dots are among the most widely studied types. These classifications are essential for understanding their diverse functionalities and applications. In disease diagnosis, quantum dots have demonstrated remarkable potential due to their high brightness, photostability, and ability to provide precise biomarker detection. They are extensively used in bioimaging technologies, enabling high-resolution imaging of cells, tissues, and even individual biomolecules. As fluorescent markers, quantum dots facilitate cell tracking, biosensing, and the detection of diseases such as cancer, bacterial and viral infections, and immune-related disorders. Their ability to provide real-time, in vivo tracking of cellular processes has opened new avenues for early and accurate disease detection. In the realm of disease treatment, quantum dots serve as versatile nanocarriers for targeted drug delivery. Their nanoscale size and surface modifiability allow them to transport therapeutic agents to specific sites, improving drug bioavailability and reducing off-target effects. Additionally, quantum dots have shown promise as photosensitizers in photodynamic therapy (PDT). When exposed to specific wavelengths of light, quantum dots interact with oxygen molecules to generate reactive oxygen species (ROS), which can selectively destroy malignant cells, vascular lesions, and microbial infections. This targeted approach minimizes damage to healthy tissues, making PDT a promising strategy for treating complex diseases. Despite these advancements, the translation of quantum dots from research to clinical application faces significant challenges. Issues such as toxicity, stability, and scalability in industrial production remain major obstacles. The potential toxicity of quantum dots, particularly to vital organs, has raised concerns about their long-term safety. Researchers are actively exploring strategies to mitigate these risks, including surface modification, coating, and encapsulation techniques, which can enhance biocompatibility and reduce toxicity. Furthermore, improving the stability of quantum dots under physiological conditions is crucial for their effective use in biomedical applications. Advances in surface engineering and the development of novel encapsulation methods have shown promise in addressing these stability concerns. Industrial production of quantum dots also presents challenges, particularly in achieving consistent quality and scalability. Recent innovations in synthesis techniques and manufacturing processes are paving the way for large-scale production, which is essential for their widespread adoption in clinical settings. This article provides an in-depth analysis of the latest research progress in quantum dot applications, including drug delivery, bioimaging, biosensing, photodynamic therapy, and pathogen detection. It also discusses the multiple barriers hindering their clinical use and explores potential solutions to overcome these challenges. The review concludes with a forward-looking perspective on the future directions of quantum dot research, emphasizing the need for further studies on toxicity mitigation, stability enhancement, and scalable production. By addressing these critical issues, quantum dots can realize their full potential as transformative tools in disease diagnosis and treatment, ultimately improving patient outcomes and advancing biomedical science.

Malignant tumors(commonly referred to as cancer)represent a major global public health challenge and contribute significantly to the worldwide disease burden.Early screening plays a critical role in improving detection rates,enabling timely intervention,and enhancing pa-tient survival rates.However,current cancer screening guidelines primarily focus on site-specific screening,which may not fully address the need for comprehensive early detection.A scientifical-ly rational,multi-cancer screening approach offers several advantages:it optimizes the use of bio-logical samples,reduces time costs for participants,enhances the efficiency and comprehensive-ness of screening,and minimizes overall expenses.Such an approach also facilitates the rational allocation of healthcare resources,ultimately helping to reduce the societal burden of cancer.To address this need,the Cancer Epidemiology Committee of the Chinese Anti-Cancer Association has developed the Expert Consensus on Combined Screening for Common Cancers in China.This consensus integrates multidisciplinary expertise and synthesizes the latest domestic and interna-tional researches on cancer screening,early detection,and treatment for prevalent malignancies.Drawing upon China's unique demographic and healthcare context,as well as practical screening experiences,the consensus provides evidence-based recommendations on target populations,screening technologies,and procedural workflows for multi-cancer screening.These guidelines align with the principles and methodologies established by the World Health Organization(WHO),aiming to:enhance the effectiveness of combined cancer screening in China,improve early detec-tion rates,and provide a scientific foundation for national cancer prevention and control strategies.

Objective To elucidate the anti-aging effect of β-sitosterol(BS),an important component in the fruits of Alpinia oxyphylla Miq.,in C.elegans and its regulatory effect on ETS-5 gene to modulate ferroptosis.Methods C.elegans treated with 10 μg/mL BS were monitored for survival time and changes in body length,motility,and reproductive function.The effect of ETS-5 gene knockdown on survival time of C.elegans was observed,and the changes in fat accumulation and lipid redox homeostasis in the transfected C.elegans were assessed using Oil Red O staining and by detecting MDA levels and the GSH/GSSG ratio.The mRNA expression levels of ferroptosis-related genes(FTN-1,GPX-1 and AAT-9)were detected using qPCR.The effects of BS treatment and ETS-5 knockdown on AAT-9 enzyme activity in C.elegans were examined.The effect of BS on nuclear localization of FEV(the human homolog of ETS-5)was validated in cultured human umbilical venous endothelial cells(HUVECs).Results Both BS treatment and ETS-5 knockdown significantly prolonged the lifespan,promoted lipid accumulation and reduced lipid peroxidation in C.elegans.ETS-5 knockdown resulted in upregulated expressions of the ferroptosis repressors GPX-1,AAT-9 and FTN-1 and increased the GSH/GSSG ratio in C.elegans.Conclusion BS inhibits ferroptosis in C.elegans by suppressing the expression of ETS-5 transcription factor and hence the activity of AAT-9 enzyme,a key gene for ferroptosis,which in turn prolongs the lifespan of C.elegans.

Background: The association of changes in metabolic syndrome (MetS) with cognitive function remains unclear. We explored this association using prospective and Mendelian randomization (MR) studies. Methods: MetS components including high-density lipoprotein cholesterol (HDL-C), systolic blood pressure (SBP), waist circumference (WC), fasting plasma glucose (FPG), and triglycerides were measured at baseline and two follow-ups, constructing a MetS index. Immediate, delayed memory recall, and cognitive function along with its dimensions were assessed by immediate 10- word recall test (IWRT) and delayed 10-word recall test (DWRT), and mini-mental state examination (MMSE), respectively, at baseline and follow-ups. Linear mixed-effect model was used. Additionally, the genome-wide association study (GWAS) of MetS was conducted and one-sample MR was performed to assess the causality between MetS and cognitive function. Results: Elevated MetS index was associated with decreasing annual change rates (decrease) in DWRT and MMSE scores, and with decreases in attention, calculation and recall dimensions. HDL-C was positively associated with an increase in DWRT scores, while SBP and FPG were negatively associated. HDL-C showed a positive association, whereas WC was negatively associated with increases in MMSE scores, including attention, calculation and recall dimensions. Interaction analysis indicated that the association of MetS index on cognitive decline was predominantly observed in low family income group. The GWAS of MetS identified some genetic variants. MR results showed a non-significant causality between MetS and decrease in DWRT, IWRT, nor MMSE scores. Conclusion Our study indicated a significant association of MetS and its components with declines in memory and cognitive function, especially in delayed memory recall.

Objective To explore the clinical value of energy-spectrum CT single-energy imaging in enhancing the image quality and stent display of stent placement CT angiography(CTA)in lower extremity atherosclerotic occlusive disease.Methods Twenty patients[mean(65.61±9.65)years;male/female,16/4]who underwent stenting for chronic occlusive disease of the lower extremity arteries by lower extremity arterial energetic spectral CTA were retrospectively recruited at our hospital.The original images were reconstructed into seven sets of single energy(40-100 keV),120 kVp,virtual unenhanced images(VUE)and metal artifact reduction(MAR)technique images.Images were debossed and then scaffolded for display with volumetric reconstruction(VR),maximum density projection(MIP)and curve planar reconstruction(CPR),and were objectively and subjectively assessed and compared using one-way analysis of variance(ANOVA).Results The 80 keV and MAR images had the highest scores compared to the other reconstruction group images(P＜0.01).Conclusion 80 keV single-energy imaging and de-metallization artifacts MAR improve the image quality of lower extremity arterial stent lumen and structure display;therefore,they have higher diagnostic value for clinicians.

An automated identification method for adolescent schizophrenia based on brain electroencephalogram(EEG)entropy features is proposed for further improving the diagnostic accuracy of adolescent schizophrenia.The raw EEG signals are decomposed into 5 commonly used rhythm bands:Delta,Theta,Alpha,Beta,and Gamma.The permutation entropy,fuzzy entropy,and sample entropy are extracted from each rhythm band and then organized into a feature matrix structured by electrode location×frequency band.Finally,an ECA-CNN model integrating efficient channel attention(ECA)and convolutional neural network(CNN)is constructed for feature classification and realizing the automated identification of adolescent schizophrenia.The results demonstrate that the proposed ECA-CNN model has higher recognition accuracy than the traditional machine learning models,achieving an accuracy of 99.08%,a sensitivity of 99.27%,a specificity of 98.85%,a precision of 99.01%,a F1 score of 99.14%,and a Kappa coefficient of 0.9814.This study provides a new idea and method for the diagnosis of adolescent schizophrenia.

Objective To validate the diagnostic performance and risk stratification ability of an AI-based recognition system(PE-AI)for pulmonary embolism(PE)using computed tomography pulmonary angiography(CTPA)so as to analyze its diagnostic value in clinical practice.Methods A total of 416 patients with suspected PE who underwent CTPA from January 1,2023 to December 10,2023 at our hospital were included in this study.Two junior radiologists and PE-AI separately detected and diagnosed emboli in the collected cases by double-blind method,and recorded the diagnosis time respectively.Three senior radiologists reviewing with clinical follow-up results were used as the gold standard in this study.Diagnostic performance was evaluated by using the receiver operating characteristic(ROC)curve analysis and Delong-t test.For positive cases,the pulmonary artery obstruction index(PAOI)calculated by AI and manually were collected respectively and consistency analysis was performed.Results The area under the curve(AUC)of PE-AI,manual and combined diagnosis was 85.6％,90.8％and 95.1％,respectively,which differed significantly(P＜0.05).The reading time of PE-AI[(0.16±0.07)min]was significantly lower than the time of manual[(4.42±1.85)min,P＜0.001]and combined diagnosis[(4.58±1.84)min,P＜0.001].The PAOI measured by PE-AI and manually had high consistency(intraclass correlation efficient,ICC=0.80)in the subgroup analysis of confirmed cases.Conclusion AI can quickly identify pulmonary artery emboli in a short time and assist radiologists to improve diagnostic efficiency.At the same time,through the intelligent detection of PAOI,it is helpful for the risk stratification of patients with PE and optimizing the diagnosis and treatment pathway for pulmonary embolism.