ObjectiveTo investigate whether anti-programmed death-1 （PD-1） monoclonal antibody can enhance the efficacy and safety of argon-helium cryoablation combined with targeted therapy in patients with unresectable hepatocellular carcinoma （uHCC） aged 60 years or older. MethodsA retrospective analysis was performed for the clinical data of 124 patients with advanced uHCC aged 60 years or older who were treated at The Fifth Medical Center of Chinese PLA General Hospital from January 2013 to September 2024. After propensity score matching， 57 patients received cryoablation combined with targeted therapy （double combination group）， while 57 received cryoablation combined with targeted therapy and anti-PD-1 monoclonal antibody （triple combination group）. The indicators for efficacy assessment included objective response rate （ORR）， disease control rate （DCR）， progression-free survival （PFS）， overall survival （OS）， and the incidence rate of adverse events. The Mann-Whitney U test was used for comparison of continuous data between two groups， and the chi-square test or the Fisher’s exact test was used for comparison of categorical data between two groups. The Kaplan-Meier method was used to plot survival curves， and the Log-rank test was used for comparison between groups. A Cox proportional-hazards regression model analysis was used to investigate the influencing factors for survival prognosis. ResultsThe triple combination group had a significantly higher ORR than the double combination group （59.6% vs 29.8%， χ²=9.083， P=0.003）， while there was no significant difference in DCR between the two groups （87.7% vs 77.2%， χ²=1.516， P=0.218）， and compared with the double combination group， the triple combination group had significantly longer median PFS （9.1 months vs 4.8 months， χ²=7.813， P=0.005） and median OS （26.1 months vs 13.6 months， χ²=14.199， P<0.001）. The multivariate Cox proportional-hazards regression model analysis showed that triple combination treatment was an independent influencing factor for PFS （hazard ratio ［HR］=0.52， 95% confidence interval ［CI］： 0.35 — 0.78， P=0.001） and OS （HR=0.32， 95%CI： 0.20 — 0.51， P<0.001）. There was no significant difference in the incidence rate of adverse events between the two groups （P>0.05）. ConclusionTriple combination treatment with argon-helium cryoablation， targeted therapy， and anti-PD-1 monoclonal antibody can significantly improve survival benefits in uHCC patients aged 60 years or older， with a controllable safety profile.

With the emergence of deep learning techniques based on convolutional neural networks, artificial intelligence (AI) has driven transformative developments in the field of medical image analysis. Recently, large language models (LLMs) such as ChatGPT have also started to achieve distinction in this domain. Increasing research shows the undeniable role of AI in reshaping various aspects of medical image analysis, including processes such as image enhancement, segmentation, detection in image preprocessing, and postprocessing related to medical diagnosis and prognosis in clinical settings. However, despite the significant progress in AI research, studies investigating the recent advances in AI technology in the aforementioned aspects, the changes in research hotspot trajectories, and the performance of studies in addressing key clinical challenges in this field are limited. This article provides an overview of recent advances in AI for medical image analysis and discusses the methodological profiles, advantages, disadvantages, and future trends of AI technologies.
Artificial Intelligence ; Humans ; Image Processing, Computer-Assisted/methods* ; Neural Networks, Computer ; Deep Learning ; Diagnostic Imaging/methods*

Tuberculosis is a zoonotic disease posing a substantial public health threat.Immunological diagnosis and vaccine im-munization are both necessary to control tuberculosis prevalence.However,the identical antigenic components in diagnostic reagents and vaccines hinder the use of animal vaccines and limit the specificity of clinical diagnosis in humans.Differentiating infected from vaccinated animals can overcome these problems.This article reviews the progress in differential diagnosis research from three as-pects:the diagnostic effects of antigens,methods for discovering new antigens,and screening of new host immune markers,to provide a theoretical basis for future research.

Tuberculosis is a zoonotic disease posing a substantial public health threat.Immunological diagnosis and vaccine im-munization are both necessary to control tuberculosis prevalence.However,the identical antigenic components in diagnostic reagents and vaccines hinder the use of animal vaccines and limit the specificity of clinical diagnosis in humans.Differentiating infected from vaccinated animals can overcome these problems.This article reviews the progress in differential diagnosis research from three as-pects:the diagnostic effects of antigens,methods for discovering new antigens,and screening of new host immune markers,to provide a theoretical basis for future research.

ObjectiveTo investigate whether anti-PD-1 monoclonal antibody can improve the efficacy and safety of cryoablation combined with lenvatinib in the treatment of unresectable hepatocellular carcinoma （HCC）. MethodsA retrospective analysis was performed for 232 patients with unresectable HCC who were treated at The Fifth Medical Center of Chinese PLA General Hospital from January 2018 to December 2022， among whom 128 received cryoablation combined with lenvatinib （double combination） and 104 received cryoablation combined with lenvatinib and anti-PD-1 monoclonal antibody （triple combination）. Propensity score matching was performed at a ratio of 1∶1， and finally there were 86 patients in each group. The two groups were evaluated in terms of objective response rate （ORR）， disease control rate （DCR）， overall survival （OS）， progression-free survival （PFS）， and adverse events （AEs）. The independent-samples t test was used for comparison of normally distributed continuous data between two groups， and the Mann-Whitney U test was used for comparison of non-normally distributed continuous data between two groups； the chi-square test was used for comparison of categorical data between two groups. Survival curves were plotted， and the Kaplan-Meier method was used to calculate the survival rate of patients in both groups， while the log-rank test was used for comparison between the two groups. The Cox regression model was used to calculate hazard ratio （HR） and 95% confidence interval （CI） and perform the univariate and multivariate analyses of influencing factors for prognosis. ResultsThe median follow-up time was 28 months， and there were 33 deaths （38.0%） in the triple combination group and 40 deaths （46.0%） in the double combination group. Compared with the double combination group， the triple combination group had significantly higher ORR （35.6% vs 14.5%， P=0.008） and DCR （86.1% vs 64.1%， P=0.003）. OS and PFS in the triple combination group were significantly higher than those in the double combination group （P=0.045 and 0.026）. The univariate and multivariate Cox proportional-hazards regression model analyses showed that treatment regimen （HR=0.60， P=0.038） and alpha-fetoprotein level （HR=2.37， P=0.001） were independent risk factors for OS， and treatment regimen （HR=0.65， P=0.025）， diabetes mellitus （HR=1.94， P=0.005）， whether or not to have received local treatment （HR=0.63， P=0.014）， and distant metastasis （HR=0.58， P=0.009） were independent risk factors for PFS. There was no significant difference in the incidence rate of AEs between the two groups （P>0.05）. ConclusionFor patients with unresectable HCC， the triple combination of cryoablation， lenvatinib， and anti-PD-1 monoclonal antibody significantly improves the treatment outcome and survival of patients compared with the double combination of cryoablation and lenvatinib， without increasing AEs， which provides a clinical basis for optimizing the treatment regimen for unresectable HCC.

Objective To investigate the expression of Midkine(MDK)in cholangiocarcinoma(CCA)and its value in predicting the prognosis of CCA,as well as the potential mechanism of the effect of MDK on the progression of CCA.Methods The data of CCA samples were obtained from TCGA database to analyze the difference in the expression of MDK between cancer tissue and paracancerous tissue and its association with clinical features,and the data collected from GEO database and 11 CCA patients who underwent surgical resection in The Fifth Medical Center of Chinese PLA General Hospital from June 2018 to September 2021 were used for validation.STRING and Cytoscape were used to construct a protein-protein interaction network,and gene ontology(GO)and Kyoto Encyclopedia of Genes and Genomes(KEGG)enrichment analyses were used to investigate the biological functions and tumor-related pathways involving MDK-related genes.In addition,TIMER and TISIDB databases were used to analyze the correlation between MDK expression and immune cell infiltration in CCA tissue.The independent-samples t test or the Mann-Whitney U test was used for comparison of continuous data between two groups,and the Fisher's exact test was used for comparison of categorical data between two groups.The Kaplan-Meier method was used to plot survival curves,and the Log-rank test was used for comparison between groups.The Spearman correlation analysis was used to investigate the correlation between two variables.Results The expression level of MDK in cancer tissue and paracancerous tissue of CCA patients was compared based on TCGA database,and the results of the non-paired and paired analyses showed that the expression level of MDK in CCA tumor tissue was significantly higher than that in paracancerous tissue(P<0.001).Transcriptome sequencing was performed for the tumor tissue and its corresponding paracancerous tissue from 11 CCA patients,and the results showed that the expression level of MDK in CCA tumor tissue was significantly higher than that in corresponding paracancerous tissue(P<0.01).High expression of MDK was associated with lymph node metastasis(P=0.045)and vascular invasion(P=0.044).Survival analysis showed that compared with the CCA patients with low MDK expression,the CCA patients with high MDK expression had significantly shorter overall survival time(χ2=5.30,P=0.028)and disease-specific survival time(χ2=6.25,P=0.019).The GO and KEGG enrichment analyses showed that the 30 MDK-related genes were closely associated with ubiquitin-mediated proteolysis and affected the prognosis of CCA patients.The TIMER analysis showed that the expression level of MDK was positively correlated with the infiltration of B cells(r=0.356,P=0.035 6)and dendritic cells(r=0.409,P=0.014 7)in tumor microenvironment of CCA;the TISIDB analysis showed that the expression level of MDK was positively correlated with CXCL16(r=0.465,P=0.004 67)and was negatively correlated with CXCL12(r=-0.389,P=0.019 7)and CXCR5(r=-0.393,P=0.018 5),and it was also negatively correlated with the immune checkpoint regulators VTCN1(r=-0.393,P=0.018 3),LTA(r=-0.380,P=0.022 7),and PVR(r=-0.350,P=0.037 3).Conclusion High expression of MDK is associated with poor prognosis in CCA patients,and MDK has the potential of being used as a molecular marker for predicting the prognosis of CCA.MDK may promote the development and progression of CCA by regulating ubiquitin-mediated proteolysis and the infiltration of B cells and dendritic cells.

Objective To investigate the expression of disulfidptosis-related genes in hepatocellular carcinoma(HCC)and the prognostic value of disulfidptosis in HCC,to construct a prognostic model,and to analyze its impact on the biological processes of HCC and sorafenib resistance.Methods The TCGA-LIHC database was used to collect the mRNA expression profiles and corresponding clinical data of HCC patients,and the LASSO-Cox regression algorithm was used to construct a four-gene predictive model for prognosis in the TCGA cohort.The external datasets ICGC and GSE14520 were used to validate the prognostic efficacy of the model,and the Cancer Drug Sensitivity Genomics(GDSC)data were used to investigate the value of the disulfidptosis model in predicting sorafenib treatment response,and gene ontology(GO)and Kyoto Encyclopedia of Genes and Genomes(KEGG)analyses were used to investigate the biological functions of disulfidptosis-related genes.The independent-samples t test was used for comparison of continuous data between two groups,and the chi-square test was used for comparison of categorical data between two groups.The Kaplan-Meier curve and the log-rank test were used to evaluate the difference in prognosis,and univariate and multivariate Cox regression analyses were used to investigate whether risk score was an independent influencing factor for patient prognosis.Results The univariate Cox regression analysis in the TCGA cohort showed that seven known disulfidptosis-related genes were significantly associated with overall survival(OS)in HCC(all P<0.05).The LASSO-Cox regression analysis was used to construct a prognostic model based on disulfidptosis-related genes(DRG),and the risk score RS-DRG was calculated as RS-DRG=(0.061 6)×GYS1 expression level+(0.152 8)×LRPPRC expression level+(0.268 3)×RPN1 expression level+(0.183 5)×SLC7A11 expression level.The log-rank test showed that the patients with a high risk score based on the disulfidptosis model had a significantly lower OS than those with a low risk score(P<0.001).Based on the results of the multivariate Cox regression analysis,risk score was an independent predictive factor for OS in both TCGA and ICGC cohorts(TCGA:hazard ratio[HR]=1.869,P=0.002;ICGC:HR=3.469,P=0.004).The Spearman correlation analysis showed that RS-DRG was significantly positively correlated with the infiltration level of various immune cells(including B lymphocytes,CD4+T lymphocytes,neutrophils,macrophages,and dendritic cells)in tumor microenvironment(all P<0.05).The patients in the high-risk score group had a significantly lower IC50 value of sorafenib and were more sensitive to sorafenib(P<0.001).The KEGG/GO enrichment analysis showed that the differentially expressed disulfidptosis-related genes were significantly enriched in various mitosis-related molecular functions.Conclusion This study constructed a novel prognostic model based on disulfidptosis-related genes,which has a potential clinical value in predicting the prognosis of HCC,and targeting disulfidptosis-related genes may provide a promising approach for HCC treatment.

When performing eye movement pattern classification for different tasks, support vector machines are greatly affected by parameters. To address this problem, we propose an algorithm based on the improved whale algorithm to optimize support vector machines to enhance the performance of eye movement data classification. According to the characteristics of eye movement data, this study first extracts 57 features related to fixation and saccade, then uses the ReliefF algorithm for feature selection. To address the problems of low convergence accuracy and easy falling into local minima of the whale algorithm, we introduce inertia weights to balance local search and global search to accelerate the convergence speed of the algorithm and also use the differential variation strategy to increase individual diversity to jump out of local optimum. In this paper, experiments are conducted on eight test functions, and the results show that the improved whale algorithm has the best convergence accuracy and convergence speed. Finally, this paper applies the optimized support vector machine model of the improved whale algorithm to the task of classifying eye movement data in autism, and the experimental results on the public dataset show that the accuracy of the eye movement data classification of this paper is greatly improved compared with that of the traditional support vector machine method. Compared with the standard whale algorithm and other optimization algorithms, the optimized model proposed in this paper has higher recognition accuracy and provides a new idea and method for eye movement pattern recognition. In the future, eye movement data can be obtained by combining it with eye trackers to assist in medical diagnosis.
Animals ; Support Vector Machine ; Whales ; Eye Movements ; Algorithms

Objective To investigate the role of P-I-R classification and Laennec grading in evaluating histological changes in patients with hepatitis B cirrhosis after receiving antiviral therapy, as well as the association of these two evaluation systems with clinical prognosis. Methods A total of 218 patients from 14 centers were consecutively screened from October 2013 to October 2014, and these patients were diagnosed with liver cirrhosis based on pathology (Ishak score ≥5), received antiviral therapy for 72 weeks, completed two liver biopsies, and met the P-I-R classification criteria. The 218 patients were divided into non-hepatocellular carcinoma (HCC) group with 186 patients and HCC group with 32 patients. The chi-square test and the Fisher's exact test were used for comparison of categorical data between groups. For the comparison of HCC after antiviral therapy, the non-parametric Mann-Whitney U test was used for continuous variables, and for the comparison of P-I-R classification and Laennec grading, the non-parametric Kruskal-Wallis H test was used for continuous variables. Univariate and multivariate Cox regression analyses were used to calculate hazard ratio ( HR ) and 95% confidence interval ( CI ), and the Kaplan-Meier method was used to calculate the cumulative incidence rate of HCC. Results After 72 weeks of antiviral therapy, there was a significant difference in P-I-R classification between the non-HCC group and the HCC group ( P < 0.001). There were significant differences in the distribution of Laennec grading and P-I-R classification before and after antiviral therapy ( P < 0.001). After antiviral therapy, the 218 patients were divided into 4A group with 33 patients, 4B group with 71 patients, and 4C group with 114 patients according to Laennec grading, and there were significant differences between these three groups in platelet count (PLT) ( H =36.429, P < 0.001), liver stiffness measurement (LSM) ( H =13.983, P =0.004), Ishak score ( χ 2 =23.060, P < 0.001), and HAI score ( P < 0.001). After antiviral therapy, the 218 patients were divided into R group with 70 patients, I group with 52 patients, and P group with 96 patients according to P-I-R classification, and there were significant differences between these three groups in PLT ( H =7.193, P =0.028), LSM ( H =6.238, P =0.045), Ishak score ( χ 2 =7.986, P < 0.001), HAI score ( P =0.002), and HCC ( P < 0.001). There was a significant difference in the incidence rate of HCC between the P and R groups based on P-I-R classification ( HR =24.21, 95% CI : 0.46-177.99, P =0.002). After adjustment for other confounding factors, P-I-R classification was an independent predictive factor for HCC ( HR =12.69, 95% CI : 4.63-34.80, P =0.002). Conclusion Both P-I-R classification and Laennec grading can reflect the features and changes of fibrosis before and after antiviral therapy, and P-I-R classification is more sensitive to fibrosis changes after antiviral therapy. P-I-R classification (after treatment) can be used to assess the risk of HCC in patients after antiviral therapy.

Objective:To investigate the use of discriminant analysis to predict the risk of nosocomial mortality in patients with traumatic hemorrhagic shock.Methods:The clinical data of 238 patients with traumatic hemorrhagic shock admitted to Peking University People's Hospital from Sep 2013 to Aug 2020 were retrospectively analyzed. Patients were divided into survival group (214 cases) and death group (24 cases). Stepwise discriminant analysis was used to establish a discriminant model.Results:The difference of history of stroke (9.8% vs. 25.0%), main site of bleeding (extremities)(58.9% vs. 29.2%), APACHEⅡ score (16.4±5.1 vs. 23.2±6.1), blood lactic acid [2.1(1.1-3.5) mmol/L vs. 4.9(2.0-13.4) mmol/L] and surgery (92.5% vs. 58.3%) between the two groups was all statistically significant (all P<0.05). Finally, There are five indicators that entered the discriminant model: history of stroke, main site of bleeding (extremities), blood lactic acid, APACHE Ⅱ score and surgery. The area under the ROC curve for predicting the risk of mortality in patients with traumatic hemorrhagic shock was 0.857, 95% CI 0.754-0.959. Conclusions:The established discriminant model has a high accuracy in predicting the risk of in-hospital mortality in patients with traumatic hemorrhagic shock.