Objective To construct large medical model named by "Huaxi HongYi"and explore its application effectiveness in assisting medical record generation. Methods By the way of a full-chain medical large model construction paradigm of "data annotation - model training - scenario incubation", through strategies such as multimodal data fusion, domain adaptation training, and localization of hardware adaptation, "Huaxi HongYi" with 72 billion parameters was constructed. Combined with technologies such as speech recognition, knowledge graphs, and reinforcement learning, an application system for assisting in the generation of medical records was developed. Results Taking the assisted generation of discharge records as an example, in the pilot department, after using the application system, the average completion times of writing a medical records shortened (21 min vs. 5 min) with efficiency increased by 3.2 time, the accuracy rate of the model output reached 92.4%. Conclusion It is feasible for medical institutions to build independently controllable medical large models and incubate various applications based on these models, providing a reference pathway for artificial intelligence development in similar institutions.

Colon cancer is a complex disease characterized by the impairment of body,qi and spirit,as well as the establishment of a tumor immune microenvironment(TIME)induced by chronic stress.Chronic stress is classified as a micro-level mental disorder,while TIME serves as the biological foundation for qi disorders.The observable manifestation of colon cancer is the tangible representation of physical disease.The interconnected mechanism of"chronic stress-TIME-colon cancer"aligns with the traditional Chinese medicine's understanding of disease as involving the interplay between the body,qi and spirit.In treatment,we should cooperate to improve the"regulating mind"of chronic stress and reshape the"invigorating qi"of TIME,and finally achieve the purpose of shape treatment to delay the progression of colon cancer.The paper is to provide new insights into the treatment of colon cancer with traditional Chinese medicine.

The concept of competency-based standardized residency training is gaining global popularity. However, the process of assessing, continuously evaluating, and conducting final competency evaluations remains challenging. The Milestones 2.0 system, developed by the Accreditation Council for Graduate Medical Education, provides a framework for evaluating competencies in radiation oncology residents. The core objective of this system is to assess sub-competencies within core competencies, categorizing them from novice to expert across 5 levels. Evaluation occurs every 6 months, with the expectation that all residents reach level 4 in all sub-competencies by the end of their training. This approach aims to enhance the standardization of residency evaluations across the United States. This article aims to analyze the Milestones 2.0 competency framework and explore its potential applicability and reference value for standardized radiation oncology residency training in China.

Colon cancer is a complex disease characterized by the impairment of body,qi and spirit,as well as the establishment of a tumor immune microenvironment(TIME)induced by chronic stress.Chronic stress is classified as a micro-level mental disorder,while TIME serves as the biological foundation for qi disorders.The observable manifestation of colon cancer is the tangible representation of physical disease.The interconnected mechanism of"chronic stress-TIME-colon cancer"aligns with the traditional Chinese medicine's understanding of disease as involving the interplay between the body,qi and spirit.In treatment,we should cooperate to improve the"regulating mind"of chronic stress and reshape the"invigorating qi"of TIME,and finally achieve the purpose of shape treatment to delay the progression of colon cancer.The paper is to provide new insights into the treatment of colon cancer with traditional Chinese medicine.

The concept of competency-based standardized residency training is gaining global popularity. However, the process of assessing, continuously evaluating, and conducting final competency evaluations remains challenging. The Milestones 2.0 system, developed by the Accreditation Council for Graduate Medical Education, provides a framework for evaluating competencies in radiation oncology residents. The core objective of this system is to assess sub-competencies within core competencies, categorizing them from novice to expert across 5 levels. Evaluation occurs every 6 months, with the expectation that all residents reach level 4 in all sub-competencies by the end of their training. This approach aims to enhance the standardization of residency evaluations across the United States. This article aims to analyze the Milestones 2.0 competency framework and explore its potential applicability and reference value for standardized radiation oncology residency training in China.

ObjectiveTo observe the effect of Tongxie Yaofang on the function of tumor-related natural killer （NK） cells under chronic stress and explore the possible molecular mechanism. MethodFifty SPF-grade BABL/C male mice were randomized into normal， model， and low-， medium-， and high-dose （6.825， 13.65， and 27.3 g·kg^-1， respectively） Tongxie Yaofang groups， with 10 mice in each group. Other groups except the blank group were subjected to 7 days of chronic restraint stress， and then forced swimming and tail suspension tests were carried out to evaluate the modeling performance. After the successful modeling， rats in Tongxie Yaofang groups were administrated with low-， medium-， and high-doses of Tongxie Yaofang by gavage， while those in the other groups were administrated with normal saline by gavage. After 14 days， each group of mice was inoculated with subcutaneous colon cancer to establish the model of colon cancer under chronic stress. The pathological changes of the tumor tissue in each group of mice were observed using hematoxylin-eosin （HE） staining. The content of CD49b-positive cells in the peripheral blood and tumor tissue of mice was measured by flow cytometry. Enzyme-linked immunosorbent assay （ELISA） was employed to measure the content of molecules associated with NK cell activation in the peripheral blood. Western blot was employed to determine the protein levels of major histocompatibility complex class Ⅰ polypeptide-related sequences A and B （MICA+MICB） and UL-16-binding protein 1 （ULBP1） in the tumor tissue. ResultCompared with the normal group， the model group showed a decrease in 5-hydroxytryptamine （5-HT） content and an increase in corticosterone （CORT） content in the serum （P<0.05）. Compared with the model group， Tongxie Yaofang increased the 5-HT content and decreased the CORT content （P<0.05， P<0.01）. Compared with the normal group， the modeling increased the tumor volume and weight （P<0.05）， while Tongxie Yaofang inhibited such increases with no statistical significance. The tumor cells in the model group presented neat arrangement， irregular shape， uneven size， obvious atypia， common nuclear division， and small necrotic area， and blood vessels were abundant surrounding the tumor cells. Compared with the model group， Tongxie Yaofang groups showed sparse arrangement of tumor cells， different degrees of patchy necrosis areas in the tumor， and karyorrhexis， dissolution， and nuclear debris in the necrotic part. Compared with the normal group， the model group showed reduced CD49b-positive cells in the peripheral blood and tumor tissue （P<0.01）. Compared with the model group， Tongxie Yaofang increased CD49b-positive cells （medium dose P<0.01， high dose P<0.05， P<0.01）. Compared with the normal group， the modeling lowered the serum levels of granzymes-B （Gzms-B）， perforin （PF）， interferon （IFN）-γ， and tumor necrosis factor （TNF）-α （P<0.05， P<0.01）. Compared with the model group， low-dose Tongxie Yaofang elevated the serum levels of PF， Gzms-B， and TNF-α （P<0.05， P<0.01）， and medium-dose Tongxie Yaofang elevated the serum levels of Gzms-B， PF， IFN-γ， and TNF-α （P<0.05， P<0.01）. In addition， high-dose Tongxie Yaofang elevated the serum levels of PF， IFN-γ， and TNF-α （P<0.01）. Compared with the normal group， the model group presented down-regulated protein level of ULBP1 （P<0.05）. Compared with the model group， low-， medium-， and high-dose Tongxie Yaofang up-regulated the protein level of ULBP1 （P<0.05， P<0.01）， and medium- and high-dose Tongxie Yaofang up-regulated the protein level of MICA+MICB （P<0.05， P<0.01）. ConclusionTongxie Yaofang may promote NK cell activation by up-regulating the expression of MICA+MICB and ULBP1， thereby delaying the progression of colon cancer under chronic stress.

Since the approval of gemtuzumab ozogamicin,an antibody-drug conjugate(ADC)targeting CD33 in 2000,13 ADC drugs have been approved by the FDA.Although these drugs have clearly improved the survival of patients with various types of advanced cancers,their significant toxicity has compromised their therapeutic benefits.The adverse reactions of ADC drugs are complex and include on-target and off-target toxicities,where the payload drug is a determining factor.Antibody and linker may also affect the degree of toxicity.Combination therapy becomes an important strategy in anticancer treatment because of its increased efficiency,but treatment-related adverse reactions also increase accordingly.This review comprehensively analyzes the toxicity mechanisms of current ADC drugs and proposes various optimization strategies,including but not limited to optimizing linker molecules,upgrading antibody design,and changing drug administration strategies,to improve the overall safety profile of ADC drugs.

Objective To investigate the predictive value of systemic immune-inflammation index (SII) for venous thrombo embolism (VTE) in patients with lung cancer.Methods The patients with lung cancer admitted and treated in Chengdu Municipal Second People's Hospital from March 2022 to October 2023 were selected as the study subjects and divided into the VTE group (n=33) and the non-VTE group (n=37) ac-cording to whether or not complicating VTE.The clinical characteristics and laboratory indexes were collect-ed.The multivariate logistic regress was used to analyze the influencing factors.The receiver operating charac-teristic (ROC) curve was drawn.The difference of the areas under curve (AUC) was compared.The Kaplan-Meier survival curve was used to evaluate the risk stratification capability of the related variables for VTE oc-currence in the patients with lung cancer.Results There were statistically significant differences in WBC,neutrophil,lymphocyte,albumin (ALB),D-dimer (D-D),plasma viscosity,neutrophil-to-lymphocyte ratio (NLR),platelet-to-lymphocyte ratio (PLR) and SII between the two groups (P<0.05).The multivariate lo-gistic regression analysis showed that high SII,tumor stage (Ⅲ-Ⅳ stage) and plasma viscosity were inde-pendent risk factors for VTE occurrence in the patients with lung cancer (P<0.05).The ROC curve analysis showed that AUC of the Khorana score,SII,modified Khorana score,NLR and PLR were 0.747,0.776,0.866,0.754 and 0.672,respectively.The predictive efficiency of SII for VTE occurrence in the patients with lung cancer was better than that of Khorana score,NLR and PLR.Conclusion High SII is an independent risk factor for VTE occurrence in the patients with lung cancer,its risk stratification for VTE occurrence in the pa-tients with lung cancer is conducive to early recognize the high risk population.

Objective To screen mitochondria-related genes in Crohn's disease (CD) based on the Gene Expression Omnibus (GEO) database, construct an artificial neural network diagnostic model and evaluate its performance. Methods The CD-related datasets GSE186582 and GSE102133 were downloaded from the GEO database for differential expression genes (DEGs) screening. The intersection of DEGs and mitochondrial genes from the MitoCarta 3.0 database was obtained. Least absolute shrinkage and selection operator regression and random forest algorithms were used to identify CD-specific genes and construct an artificial neural network diagnostic model. The model was further validated by the validation set GSE95095, and the diagnostic performance was evaluated by the area under the curve (AUC) of the receiver operating characteristic curve. The immune cell infiltration in CD was assessed by the CIBERSORT algorithm, and the relationship between biomarkers and infiltrated immune cells was investigated. Results A total of 551 DEGs were obtained, including 275 upregulated and 276 downregulated genes. There were 20 mitochondria-related genes associated with CD. A total of 9 mitochondria-related feature genes (SOD2, MTHFD2, BPHL, PXMP2, RMND1, AGXT2, MAOA, HMGCS2, MAOB) were screened by two machine learning algorithms. An artificial neural network diagnostic model was constructed by the selected feature genes. The values of AUC of the model in the training and validation groups were 0.956 and 0.736 respectively. Immune cell infiltration analysis showed that the feature genes were associated with resting memory CD4⁺ T cells, activated memory CD4⁺ T cells, activated dendritic cells, neutrophils, and CD8⁺ T cells. Conclusion The artificial neural network diagnostic model for CD based on 9 mitochondrial genes has good predictive performance.

Objective To construct a prognostic prediction model for hepatocellular carcinoma (HCC) based on immune and metabolism related genes, analyze the prognostic immune response of HCC patients, and screen potential drugs for HCC treatment through drug sensitivity analysis. Methods HCC expression profiling and clinical data were obtained from The Cancer Genome Atlas (TCGA) database, and a list of immune-related genes was obtained from the Immport database; the Perl language was used to extract metabolism-related pathway gene sets from the Molecular Signatures Database(MSig DB), and co-expression related genes were found through differential analysis and co-expression analysis; the univariate Cox regression analysis, the least absolute shrinkage and selection operator (LASSO) regression analysis, and multivariate Cox regression analysis were used to screen prognosis-related genes and construct a risk prognosis model for HCC, and risk scores for all HCC samples were calculated. Using the median risk score as the critical value, the reliability of the prognostic model was evaluated through risk curves, Kaplan-Meier survival analysis, receiver operating characteristic (ROC) curves, independent prognostic analysis, and Nomograms. The correlations between risk scores and pathway enrichment analysis as well as immune cell infiltration were analyzed. Drug sensitivity analysis was used to identify potential therapeutic drugs for HCC. Results Five immune and metabolic genes with independent prognostic value were obtained, and a prognostic model based on immune and metabolic genes was constructed. Survival analysis showed that in the total dataset, training group and validation group, the survival rate of the low-risk group was significantly higher than that ofthe high-risk group (P < 0.05). The areas under the ROC curves of the prognostic model for the training group at 1, 3 and 5 years were 0.780, 0.699 and 0.706 respectively. Cox regression analysis showed that grading and risk score could be used as independent prognostic factors for HCC (P < 0.05), with a concordance index of 0.734 (95%CI, 0.669 to 0.798), indicating good model performance. Immune cell infiltration results showed significant differences in resting NK cells, monocytes, M0 macrophages, and M1 macrophages between the high-risk and low-risk groups (P < 0.05). Drug sensitivity analysis screened 12 drugs that may have potential therapeutic effects in HCC patients (P < 0.01). Conclusion The prognostic model of HCC based on five immune and metabolic genes has good predictive performance, and can be used as a new indicator for prognosis evaluation; the 12 drugs screened out have potential efficacy for HCC.