ObjectiveTo address the challenges of unstructured classical Chinese expressions， nested entity relationships， and limited annotated data in famous traditional Chinese medicine（TCM） case records， this study proposes a joint relation extraction framework that integrates data augmentation and entity mapping， aiming to support the construction of TCM diagnostic knowledge graphs and clinical pattern mining. MethodsWe developed an annotation structure for entities and their relationships in TCM case texts and applied a data augmentation strategy by incorporating multiple ancient texts to expand the relation extraction dataset. A cascade binary tagging framework for relation triple extraction（CasRel） model for TCM semantics was designed， integrating a pre-trained bidirectional encoder representations from transformers（BERT） layer for classical TCM texts to enhance semantic representation， and using a head entity-relation-tail entity mapping mechanism to address entity nesting and relation overlapping issues. ResultsExperimental results showed that the CasRel model， combining data augmentation and entity mapping， outperformed the pipeline-based Bert-Radical-Lexicon（BRL）-bidirectional long short-term memory（BiLSTM）-Attention model. The overall precision， recall， and F₁-score across 12 relation types reached 65.73%， 64.03%， and 64.87%， which represent improvements of 14.26%， 7.98%， and 11.21% compared to the BRL-BiLSTM-Attention model， respectively. Notably， the F₁-score for tongue syndrome relations increased by 22.68%（69.32%）， and the prescription-syndrome relations performed the best with the F₁-score of 70.10%. ConclusionThe proposed framework significantly improves the semantic representation and complex dependencies in TCM texts， offering a reusable technical framework for structured mining of TCM case records. The constructed knowledge graph can support clinical syndrome differentiation， prescription optimization， and drug compatibility， providing a methodological reference for TCM artificial intelligence research.

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*

Objective To systematically evaluate influence factors hospitalization infections in multiple myeloma(MM)patients after chemotherapy.Methods Computer searches were conducted on relevant literature in CNKI,China Biology Medicine disc,VIP,Wanfang Data Knowledge Service Platform,PubMed,Web of Science,Embase,Cochrane Library and CINAHL from the database inception until December 16,2024.Two researchers independently screened and assessed the quality of the literature,obtained the necessary information,and a Meta-analysis of risk factors was conducted by using RevMan 5.4 software.Results 19 articles were included in total.Meta-analysis results showed that high body mass index,length of stay,smoking history,Eastern Cooperative Oncology Group(ECOG)score,granulocyte deficiency,neutropenia,Durie-Salmon stage,international staging system(ISS)stage and combined with diabetes,renal insufficiency,anemia,hypoalbuminemia were the risk factors for hospitalization infections in patients with MM after chemotherapy(P＜0.05).Conclusion This study provides a reference for intervening in the risk factors of hospitalization infections in MM patients after chemotherapy.Medical staff should prevent infections early based on relevant factors,identify high-risk populations,and maximize the protection of patient health outcomes and good prognosis.

Objective To systematically evaluate the risk factors for metabolic syndrome in hospitalized patients with schizophrenia in China.Methods Relevant observational studies were retrieved for Chinese schizophrenia patients with metabolic syndrome in Databases,with a retrieval period from the database establishment date to January 12 2025.Two researchers independently screened the literature,extracted data and evaluated the quality of the studies,and a total of 16 articles were included for Meta analysis.Results Age,body mass index,smoking history,disease duration,family history of metabolic syndrome,diabetes history,hypertension history,chlorantraniliprole use,olanzapine use,interleukin-6 levels,leptin levels,triglyceride levels,and high-density lipoprotein cholesterol levels were significant risk factors for metabolic syndrome in hospitalized Chinese schizophrenia patients(P＜0.05).Moderate recreational exercise served as a protective factor(P＜0.05).Conclusion There are many influencing factors for the association of metabolic syndrome in hospitalized patients with schizophrenia in China,and moderate exercise is a protective factor for the association of metabolic syndrome in schizophrenia.In clinical practice,high-risk groups of metabolic syndrome can be actively screened according to relevant risk factors.

Objective To investigate the clinical characteristics and prognostic risk factors in elderly patients with acute myeloid leukemia(AML).Methods A retrospective analysis was conducted on the clinical data of 101 elderly AML patients admitted to Affiliated Hospital of Jiaxing University from January 2022 to December 2024.All patients were treated with azacitidine+venetoclax regimen.The clinical characteristics of patients and the risk factors related to prognosis were explored.Results The median follow-up was 14 months.Among the 101 patients,74 achieved complete remission or complete remission with incomplete hematological recovery.The median overall survival(OS)of patients with aged ≥70 years,white blood cell count＞50 × 109/L,TP53 mutation,complex chromosomes,and high-risk European leukemia net(ELN)risk stratification was significantly shortened(P＜0.05).Multivariate analysis showed that age(HR=0.125,95％CI:0.023-0.662,P=0.015),white blood cell count(HR=0.145,95％CI:0.032-0.662,P=0.013),and ELN risk stratification(HR=100.397,95％CI:14.395-700.207,P＜0.001)were all independent influencing factors for OS in elderly AML patients.Conclusion Age,white blood cell count and ELN risk stratification are all independent influencing factors affecting OS in elderly AML patients.

Objective:To establish a rapid and accurate duplex real-time fluorescent reverse transcription-recombinase polymerase amplification/recombinase-aided amplification（RT-RPA/RAA）detection method for identification and differentiation of Nipah virus（NiV），Langya virus（LayV），Mojiang virus（MoJV），and Cedar virus（CedV）.Methods:First，specific primers and probes were designed targeting the conserved L gene regions of NiV and LayV，as well as the conserved N gene regions of MoJV and CedV，respectively. The four viruses were divided into two groups for duplex detection. Subsequently，the optimal primer and probe combinations were screened by comparing the amplification efficiency of different primer pair combinations（F1/R1，F1/R2，F2/R1，F2/R2）. The reaction temperature was optimized through temperature gradient settings from 37 ℃ to 42 ℃，and the amounts of primers and probes were optimized to establish the duplex real-time fluorescent RT-RPA/RAA detection system. Finally，the detection performance was evaluated through specificity，sensitivity，and stability tests，as well as clinical sample validation.Results:The selected primer pairs（NiV primer pair F2/R1，LayV primer pair F1/R1，MoJV primer pair F2/R2，and CedV primer pair F2/R2）all demonstrated optimal amplification efficiency when combined with their corresponding probes. The optimal annealing temperature was 39 ℃，and the minimum detection limit was 101-102 copies/μl. The method could effectively distinguish target viruses from other non-target viruses，and repeated experiments showed good stability（ R2> 0.90）. Additionally，detection results for Malaysian NiV strains and various clinical samples were consistent with the Taqman multiplex qRT-PCR method. Conclusion:The duplex real-time fluorescent RT-RPA/RAA detection method successfully established in this study can rapidly and accurately identify and differentiate four important henipavirus-like viruses：NiV，LayV，MoJV，and CedV. It features simple operation，rapid reaction，high specificity，and good stability，providing an effective molecular detection tool for rapid field diagnosis，surveillance，and control of these zoonotic viruses.

Objective:To visually analyze the research literature on the analysis of TCM compounds; To explore the research hotspots and trends in this field.Methods:The literature related to the disassembly analysis of TCM compounds was retrieved from CNKI, Chongqing VIP, Wanfang Data, SinoMed, and China Medical Journal Full-text Database from January 1, 1981 to December 31, 2024. CiteSpace 6.4.R1 software was used to visualize the number of articles, authors, institutions and keywords, and to draw the cooperation network diagram of authors and institutions, keyword co-occurrence, clustering, timeline and burst map.Results:A total of 1 728 Chinese articles were included, and the number of publications showed a fluctuating upward trend. A comparatively high number of publications was in 2007 and 2016, followed by a slight decline but maintained at a high level. There is a trend of recovery in 2024. The author with the highest number of articles was Professor Fang Zhaoqin, and the institution with the highest number of articles was Beijing University of Chinese Medicine. High frequency keywords included rats, compatibility, experimental research, cell apoptosis, TCM compound, and a total of 19 clusters and 25 emergent keywords were formed.Conclusions:The research contents and methods of the research on the disassembly of TCM compounds are relatively rich, and there are many explorations on classical prescriptions. The study of disassembled prescriptions has played a driving role in the modernization of TCM compounds. In the future, high-quality cooperation between regions, institutions, and authors, combination with modern medicine and scientific methods, will further improve the quality of research in this field.

Objective:To investigate the expression of heat shock factor binding protein 1 (HSPB1) in endometrial carcinoma and its clinical significance.Methods:The pan-cancer dataset after standardization and unification was downloaded from the University of California Santa Cruz (UCSC) Genome database (updated to December 6, 2019), and the expression of HSPB1 in pan-cancer was analyzed. The transcriptome data of endometrial carcinoma of the uterus from the Cancer Genome Atlas (TCGA) database were downloaded (updated to July 21, 2016), including 552 cases of endometrial carcinoma and 35 cases of corresponding adjacent tissue samples. The clinical data of 543 patients with endometrial cancer were obtained. The differences in the expression levels of HSPB1 in patients with different clinicopathological features were compared. R 4.3.1 software maxstat was used to calculate the optimal critical value (>46.30) of HSPB1 expression, and the patients were divided into HSPB1 low expression group (<46.30) and HSPB1 high expression group (≥46.30). Kaplan-Meier method was used to analyze the difference in prognosis between the 2 groups, and log-rank test was performed. The top 50 genes with positive and negative correlation with HSPB1 were screened by LinkedOmics database. Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis were performed on HSPB1. The interaction network of HSPB1 protein was analyzed by STRING database and Cytoscape 3.9.1 software. The correlation between HSPB1 expression and various immune cell infiltration levels was analyzed by using the TIMER2.0 database.Results:The expression of HSPB1 in 27 kinds of tumor tissues was higher than that in paracancerous tissues, and the expression of HSPB1 in 2 kinds of tumor tissues was lower than that in paracancerous tissues (all P < 0.05). In the transcriptome data of 552 cases of endometrial cancer and 35 cases of corresponding paracancerous tissues in the TCGA database, the relative expression level of HSPB1 in endometrial cancer tissues was higher than that in corresponding paracancerous tissues ( t = -2.90, P = 0.005). The result of the comparison of relative expression level of HSPB1 in endometrial cancer patients with different clinicopathological features showed that patients aged < 65 years had higher expression level compared to those aged ≥ 65 years, patients at clinical stage Ⅰ-Ⅱ had higher expression level compared to those at stage Ⅲ-Ⅳ, patients with Grade grading G 1-G 2 had higher expression level compared to those with G 3, and patients with pathological type I had higher expression level compared to those with type Ⅱ (all P < 0.05). Of the 543 patients, 2 were lost to follow-up, and the overall survival of the remaining 541 patients with high HSPB1 expression was better than that of those with the low expression ( HR = 0.532, 95% CI: 0.333-0.849, P = 0.008). HSPB1 and its related genes were mainly involved in estrogen signaling, p53 signaling and other pathways; HSPB1 was involved in cysteine-type endopeptidase inhibitor activity and calcium-dependent protein binding. The top 10 genes with the strongest correlation with HSPB1 in protein-protein interaction analysis were DSG3, EVPL, PKP1, DSC3, PKP3, PPL, KRT5, IVL, TGM1 and CSTA. The expression of HSPB1 was negatively correlated with tumor purity ( r = -0.025, P < 0.01), and positively correlated with CD4 + T cells ( r = 0.204, P < 0.01), CD8 + T cells ( r = 0.225, P < 0.01), B cells ( r = 0.285, P < 0.01), NK cells ( r = 0.269, P < 0.01), macrophages ( r = 0.234, P < 0.01) and dendritic cells ( r = 0.354, P < 0.01). Conclusions:The high expression of HSPB1 is associated with clinicopathological features, prognosis and immune infiltration in patients with endometrial carcinoma. It may be one of the reference indexes for predicting the prognosis of patients with endometrial cancer.

OBJECTIVE To explore the epidemiological characteristics of human immunodeficiency virus(HIV)/ac-quired immune deficiency syndrome(AIDS)complicated with anorectal diseases and analyze the clinical character-istics of the HIV-positive patients with perianal abscess.METHODS Totally 152 HIV/AIDS patients who were complicated with anorectal diseases and were treated in The Eighth Hospital of Wuhan from Jan.2020 to Dec.2023 were recruited as the research subjects,36 of whom had perianal abscess.The manifestations of anorec-tal diseases,treatment status and incidence rates of the diseases among the patients with different clinical charac-teristics in the different years were statistically analyzed.The clinical characteristics of the HIV-positive patients with perianal abscess were observed.RESULTS Of the 152 HIV/AIDS patients complicated with anorectal disea-ses,51.97％had mixed hemorrhoid,and 23.68％had perianal abscess.The number of patients with HIV/AIDS complicated with anorectal diseases gradually increased from 2020 to 2023;while the incidence rate of male,aged between 30 and 40 years old,married individuals,those with primary school education or below and famers were relatively high.Among the HIV/AIDS patients complicated with anorectal diseases,139 underwent surgical pro-cedures,and 23 did not.Among the 36 HIV-positive patients with perianal abscess,97.22％were male,50.00％were aged between 20 and 30 years old,and 44.44％were married.CONCLUSIONS The number of HIV/AIDS patients complicated with anorectal diseases is gradually increased from 2020 to 2023;the incidence rates are rela-tively high among the patients who are male,aged between 30 and 40 years old,married,with primary school edu-cation or below and famers.Perianal abscess and mixed hemorrhoid are the major types of anorectal diseases.Sur-gical procedures are recommended for HIV/AIDS patients complicated with anorectal diseases,while the HIV-positive patients with perianal abscess are dominated by the population of male,aged between 20 and 30 years old and married.It is necessary to attach great importance to the detection of pathological characteristics of the HIV-positive patients with perianal abscess so as to take appropriate treatment measures to improve the therapeutic effect.