ObjectiveTo investigate the influencing factors for recompensation in patients with decompensated liver cirrhosis， as well as the impact of recompensation on the prognosis of such patients， and to provide a basis for early identification of high-risk patients in clinical practice. MethodsA retrospective analysis was performed for the clinical data of patients who attended The Third People’s Hospital of Kunming from January 2016 to December 2022 and were diagnosed with decompensated liver cirrhosis due to hepatitis B， hepatitis C， alcoholic hepatitis， and autoimmune hepatitis， and they were divided into recompensation group and persistent decompensation group. To control for confounding factors， whether recompensation occurred was used as the rouping variable，and BMI， alcohol consumption history， HIV infection history， TG， CHOL， LDL， and HDL were used as covariates. The propensity score was calculated， and 1：1 nearest neighbor matching was performed with a caliper value of 0.1. After propensity score matching， the recompensation group and the persistent decompensation group with relatively balanced covariates were obtained. Univariate and multivariate Cox proportional-hazards regression model analyses were used to investigate the influencing factors for recompensation； the “rms” package was used to establish a nomogram； the receiver operating characteristic （ROC） curve was plotted to calculate the area under the ROC curve （AUC）； the Hosmer-Lemeshow test was used to assess the goodness of fit of the model； the “Calibration Curves” package was used to plot calibration curves for model assessment. The Kaplan-Meier method was used to plot survival curves， and the Log-rank test was used for comparison of survival curves. ResultsAmong the 863 patients with decompensated liver cirrhosis， 305 experienced recompensation， resulting in an incidence rate of 35.3%. After PSM， 610 cases were successfully matched， with 305 cases in each group. The univariate and multivariate Cox regression analyses showed that etiology （hepatitis C： hazard ratio［HR］=0.288， P=0.002）； male（HR=0.701， P=0.016）， age（HR=0.988， P=0.047）， hemoglobin （HGB）（HR=1.006， P=0.017）， and CD4 T cell（HR=1.001，P=0.047）， TIPS procedure （HR=1.808，P=0.042） were independent influencing factors for recompensation in patients with decompensated liver cirrhosis. During follow-up， 116 patients died of liver disease-related causes， with 27 patients （8.85%） in the recompensation group and 89 （15.95%） in the persistent decompensation group； 109 patients developed HCC， with 23 patients （7.54%） in the recompensation group and 86 （15.41%） in the persistent decompensation group. The Kaplan-Meier survival curves showed significant separation between the patients with different states of compensation in terms of liver disease-related mortality rate and the incidence rate of HCC， and the Log-rank test showed that there were significant differences between the two groups in liver disease-related mortality rate （χ²=9.023， P=0.003） and the incidence rate of HCC （χ²=10.526， P=0.001）. ConclusionEtiology，sex，age，TIPS，HGB，and CD4 T cell are independent influencing factors for recompensation in patients with decompensated liver cirrhosis. There is a significant difference in the incidence rate of recompensation between decompensated liver cirrhosis patients with different etiologies， and female patients and patients with a younger age，a history of TIPS， a higher HGB level， and a higher CD4 lymphocyte count are more likely to experience recompensation. Recompensation is the key to improving the long-term prognosis of patients and can significantly reduce long-term liver disease-related mortality rate and the incidence rate of HCC.

ObjectiveTo investigate the value of albumin-bilirubin （ALBI）， easy albumin-bilirubin （EZ-ALBI）， and platelet-albumin-bilirubin （PALBI） scores in predicting 2-year survival in patients with HCV-associated hepatocellular carcinoma （HCV-HCC）. MethodsA retrospective analysis was performed for the clinical data of 174 patients with HCV-HCC who were admitted to The Third People’s Hospital of Kunming from January 2020 to January 2022， and the patients were followed up till 2 years after admission. According to the follow-up results， the patients were divided into survival group with 95 patients and death group with 79 patients. The independent-samples t test or the Mann-Whitney U 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. Univariate and multivariate Cox proportional-hazards regression model analyses were used to investigate the influencing factors for the survival of HCV-HCC patients. The Kaplan-Meier method was used to plot survival curves and analyze the 2-year survival rate of HCV-HCC patients with different EZ-ALBI grades， and the log-rank test was used for comparison between groups. ResultsThere were significant differences between the survival group and the death group in platelet count， aspartate aminotransferase （AST）， total bilirubin， albumin （Alb）， alpha-fetoprotein （AFP）， prealbumin， prothrombin time， international normalized ratio， PALBI score， ALBI score， EZ-ALBI score， Model for End-Stage Liver Disease （MELD） score， HCV genotype， peritoneal effusion， and vascular invasion （all P<0.05）. The univariate Cox regression analysis showed that AST， Alb， AFP， ALBI score， EZ-ALBI score， PALBI score， MELD score， Barcelona Clinic Liver Cancer Staging， and peritoneal effusion were influencing factors for the survival of patients （all P<0.05）， and the multivariate Cox regression analysis showed that EZ-ALBI score （hazard ratio ［HR］=1.850， 95% confidence interval ［CI］： 1.054 — 3.247， P=0.032） and peritoneal effusion （HR=1.993， 95%CI： 1.030 — 3.858， P=0.041） were independent risk factors for the survival of HCV-HCC patients. The survival curve analysis showed that the patients with EZ-ALBI grade 1/2/3 had a 2-year survival rate of 90.9%， 60.2%， and 32.2%， respectively， and there was a significant difference in cumulative survival rate between the patients with different EZ-ALBI grades （χ²=26.294， P<0.001）. ConclusionEZ-ALBI score and the presence or absence of peritoneal effusion can be used as predictors of the survival of HCV-HCC patients.

ObjectiveTo investigate the value of noninvasive imaging detection （FibroScan）， two serological models of gamma-glutamyl transpeptidase-to-platelet ratio （GPR） score and S index， and two inflammatory factors of interleukin-6 （IL-6） and tumor necrosis factor-α （TNF-α） in predicting liver fibrosis in patients with HBeAg-positive chronic hepatitis B （CHB）， as well as the consistency of liver biopsy in pathological staging， and to provide early warning for early intervention of CHB. MethodsA retrospective analysis was performed for 131 HBeAg-positive CHB patients who underwent liver biopsy in The Third People’s Hospital of Kunming from January 2019 to December 2023. The results of liver biopsy were collected from all patients， and related examinations were performed before liver biopsy， including total bilirubin， alanine aminotransferase， platelet count， gamma-glutamyl transpeptidase， albumin， IL-6， TNF-α， liver stiffness measurement （LSM）， and abdominal ultrasound. An analysis of variance was used for comparison of normally distributed continuous data between groups， and the Kruskal-Wallis H test was used for comparison of non-normally distributed continuous data between groups； the chi-square test was used for comparison of categorical data between groups. A Kappa analysis was used to investigate the consistency between LSM noninvasive histological staging and pathological staging based on liver biopsy， and the Spearman analysis was used to investigate the correlation between each variable and FibroScan in the diagnosis of liver fibrosis stage. The Logistic regression analysis was used to construct joint predictive factors. The receiver operating characteristic （ROC） curve was used to evaluate the value of each indicator alone and the joint predictive model in the diagnosis of liver fibrosis， and the Delong test was used for comparison of the area under the ROC curve （AUC）. ResultsIn the consistency check， inflammation degree based on liver biopsy had a Kappa value of 0.807 （P<0.001）， and liver fibrosis degree based on liver biopsy had a Kappa value of 0.827 （P<0.001）， suggesting that FibroScan noninvasive histological staging and liver biopsy showed good consistency in assessing inflammation degree and liver fibrosis stage. Age was positively correlated with LSM， GPR score， S index， IL-6， and TNF-α （all P<0.05）， and GPR score， S index， IL-6， and TNF-α were positively correlated with LSM （all P<0.05）. GPR score， S index， IL-6， and TNF-α were all independent risk factors for diagnosing significant liver fibrosis （≥S2） and progressive liver fibrosis （≥S3） （all P<0.05）. As for each indicator alone， GPR score had the highest value in the diagnosis of significant liver fibrosis （≥S2）， followed by S index， IL-6， and TNF-α， while S index had the highest value in the diagnosis of progressive liver fibrosis （≥S3）， followed by GPR score， TNF-α， and IL-6. The joint model had a higher predictive value than each indicator alone （all P<0.05）. ConclusionThere is a good consistency between FibroScan noninvasive histological staging and pathological staging based on liver biopsy. GPR score， S index， IL-6， and TNF-α are independent risk factors for evaluating different degree of liver fibrosis in CHB， and the combined prediction model established by them can better diagnose liver fibrosis.

ObjectiveTo investigate the influencing factors for hepatic hydrothorax （HH） before intervention for primary hepatic carcinoma （PHC）， and to construct and assess the nomogram risk prediction model. MethodsA retrospective analysis was performed for the clinical data of 353 hospitalized patients who attended the Third People’s Hospital of Kunming for the first time from October 2012 to October 2021 and there diagnosed with PHC， and according to the presence or absence of HH， they were divided into HH group with 153 patients and non-HH group with 200 patients. General data and the data of initial clinical testing after admission were collected from all PHC patients. 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 or the Fisher’s exact test was used for comparison of categorical data between groups. After the multicollinearity test was performed for the variables with statistical significance determined by the univariate analysis， the multivariate Logistic regression analysis was used to identify independent influencing factors. The “rms” software package was used to construct a nomogram risk prediction model， and the Hosmer-Lemeshow test and the receiver operating characteristic （ROC） curve were used to assess the risk prediction model； the “Calibration Curves” software package was used to plot the calibration curve， and the “rmda” software package was used to plot the clinical decision curve and the clinical impact curve. ResultsAmong the 353 patients with PHC， there were 153 patients with HH， with a prevalence rate of 43.34%. Child-Pugh class B （odds ratio ［OR］=2.652， 95% confidence interval ［CI］： 1.050 — 6.698， P=0.039）， Child-Pugh class C （OR=7.963， 95%CI： 1.046‍ ‍—‍ ‍60.632， P=0.045）， total protein （OR=0.947， 95%CI： 0.914‍ ‍—‍ ‍0.981， P=0.003）， high-sensitivity C-reactive protein （OR=1.007， 95%CI： 1.001‍ ‍—‍ ‍1.014， P=0.025）， and interleukin-2 （OR=0.801， 95%CI： 0.653‍ ‍—‍ ‍0.981， P=0.032） were independent influencing factors for HH before PHC intervention， and a nomogram risk prediction model was established based on these factors. The Hosmer-Lemeshow test showed that the model had a good degree of fitting （χ²=5.006， P=0.757）， with an area under the ROC curve of 0.752 （95%CI： 0.701‍ ‍—‍ ‍0.803）， a sensitivity of 78.40%， and a specificity of 63.50%. The calibration curve showed that the model had good consistency in predicting HH before PHC intervention， and the clinical decision curve and the clinical impact curve showed that the model had good clinical practicability within a certain threshold range. ConclusionChild-Pugh class， total protein， interleukin-2， and high-sensitivity C-reactive protein are independent influencing factors for developing HH before PHC intervention， and the nomogram model established based on these factors can effectively predict the risk of developing HH.

ObjectiveTo investigate the influencing factors for recompensation in patients with decompensated hepatitis C cirrhosis， and to establish a predictive model. MethodsA total of 217 patients who were diagnosed with decompensated hepatitis C cirrhosis and were admitted to The Third People’s Hospital of Kunming l from January， 2019 to December， 2022 were enrolled， among whom 63 patients who were readmitted within at least 1 year and had no portal hypertension-related complications were enrolled as recompensation group， and 154 patients without recompensation were enrolled as control group. Related clinical data were collected， and univariate and multivariate analyses were performed for the factors that may affect the occurrence of recompensation. The independent-samples t test was used for comparison of normally distributed measurement data between two groups， and the Mann-Whitney U test was used for comparison of non-normally distributed measurement data between two groups； the chi-square test or the Fisher’s exact test was used for comparison of categorical data between two groups. A binary Logistic regression analysis was used to investigate the influencing factors for recompensation in patients with decompensated hepatitis C cirrhosis， and the receiver operating characteristic （ROC） curve was used to assess the predictive performance of the model. ResultsAmong the 217 patients with decompensated hepatitis C cirrhosis， 63 （29.03%） had recompensation. There were significant differences between the recompensation group and the control group in HIV history （χ²=4.566， P=0.034）， history of partial splenic embolism （χ²=6.687， P=0.014）， Child-Pugh classification （χ²=11.978， P=0.003）， grade of ascites （χ²=14.229， P<0.001）， albumin （t=4.063， P<0.001）， prealbumin （Z=-3.077， P=0.002）， high-density lipoprotein （t=2.854， P=0.011）， high-sensitivity C-reactive protein （Z=-2.447， P=0.014）， prothrombin time （Z=-2.441， P=0.015）， carcinoembryonic antigen （Z=-2.113， P=0.035）， alpha-fetoprotein （AFP） （Z=-2.063， P=0.039）， CA125 （Z=-2.270， P=0.023）， TT3 （Z=-3.304， P<0.001）， TT4 （Z=-2.221， P=0.026）， CD45⁺ （Z=-2.278， P=0.023）， interleukin-5 （Z=-2.845， P=0.004）， tumor necrosis factor-α （Z=-2.176， P=0.030）， and portal vein width （Z=-5.283， P=0.005）. The multivariate analysis showed that history of partial splenic embolism （odds ratio ［OR］=3.064， P=0.049）， HIV history （OR=0.195， P=0.027）， a small amount of ascites （OR=3.390， P=0.017）， AFP （OR=1.003， P=0.004）， and portal vein width （OR=0.600， P<0.001） were independent influencing factors for the occurrence of recompensation in patients with decompensated hepatitis C cirrhosis. The ROC curve analysis showed that HIV history， grade of ascites， history of partial splenic embolism， AFP， portal vein width， and the combined predictive model of these indices had an area under the ROC curve of 0.556， 0.641， 0.560， 0.589， 0.745， and 0.817， respectively. ConclusionFor patients with decompensated hepatitis C cirrhosis， those with a history of partial splenic embolism， a small amount of ascites， and an increase in AFP level are more likely to experience recompensation， while those with a history of HIV and an increase in portal vein width are less likely to experience recompensation.

ObjectiveTo investigate the efficacy and safety of the direct-acting antiviral agents coblopasvir hydrochloride/sofosbuvir （CLP/SOF） regimen used alone or in combination with ribavirin （RBV） in the treatment of patients with genotype 3 hepatitis C virus （HCV） infection in terms of virologic response rate， liver function recovery， improvement in liver stiffness measurement （LSM）， and adverse drug reactions， and to provide a reference for clinical medication. MethodsA total of 98 patients with genotype 3 HCV infection who attended The Third People’s Hospital of Kunming from January 2022 to December 2023 were enrolled， and according to the treatment method， the patients were divided into CLP/SOF+RBV treatment group with 55 patients and CLP/SOF treatment group with 43 patients. The patients were observed in terms of rapid virologic response at week 4 （RVR4）， sustained virologic response （SVR）， previous treatment experience， underlying diseases， laboratory and imaging indicators， and adverse reactions during treatment. The course of treatment was 12 weeks， and the patients were followed up for 12 weeks after drug withdrawal. 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 Friedman test was used for comparison within each group at different time points， and the Bonferroni method was used for further comparison and correction of P value； the chi-square test or the Fisher’s exact test was used for comparison of categorical data between two groups. The univariate and multivariate Logistic regression analyses were used to investigate the influencing factors for SVR12. ResultsBefore treatment， there were significant differences between the CLP/SOF+RBV treatment group and the CLP/SOF treatment group in terms of LSM， total bilirubin （TBil）， gamma-glutamyl transpeptidase （GGT）， HCV genotype， and the presence or absence of liver cirrhosis and compensation （all P<0.05）. The 98 patients with genotype 3 HCV infection had an RVR4 rate of 81.6% and an SVR12 rate of 93.9%. The patients with genotype 3a HCV infection had an RVR4 rate of 84.44% and an SVR12 rate of 97.78%， while the patients with genotype 3b HCV infection had an RVR4 rate of 79.25% and an SVR12 rate of 90.57%. There were significant differences in RVR4 and SVR12 rates between the patients without hepatocellular carcinoma and those with hepatocellular carcinoma， there was a significant difference in RVR4 rate between the patients without HIV infection and those with HIV infection， and there was a significant difference in SVR12 rate between the previously untreated patients and the treatment-experienced patients （all P<0.05）. The univariate Logistic regression analysis showed that treatment history， hypertension， hepatocellular carcinoma， ascites， albumin （Alb）， and platelet count were influencing factors for SVR12 （all P<0.05）， and the multivariate Logistic regression analysis showed that hepatocellular carcinoma （odds ratio=0.034， 95% confidence interval： 0.002‍ ‍—‍‍ 0.666， P=0.026） was an independent influencing factor for SVR12. After treatment with CLP/SOF combined with RBV or CLP/SOF alone， the patients with genotype 3 HCV infection showed gradual reductions in the liver function parameters of TBil， GGT， and alanine aminotransferase （all P<0.05） and a gradual increase in the level of Alb （P<0.05）. As for renal function， there were no significant changes in blood urea nitrogen and creatinine after treatment （P>0.05）. For the patients with or without liver cirrhosis， there was a significant reduction in LSM from baseline after treatment for 12 weeks （P<0.05）. Among the 98 patients with genotype 3 HCV infection， 9 tested positive for HCV-RNA at 12 weeks after treatment， 2 showed no response during treatment， 4 showed virologic breakthrough， and 3 experienced recurrence. The overall incidence rate of adverse events during treatment was 17.35% for all patients. ConclusionCLP/SOF alone or in combination with RBV has a relatively high SVR rate in the treatment of genotype 3 HCV infection， with good tolerability and safety in patients during treatment， and therefore， it holds promise for clinical application.

ObjectiveTo investigate the therapeutic efficacy， influencing factors， and safety of a treatment regimen based on coblopasvir hydrochloride capsules/sofosbuvir tablets in patients with chronic hepatitis C virus （HCV） infection in a real-world setting. MethodsA total of 253 patients who attended The Third People’s Hospital of Kunming from September 1， 2021 to May 31， 2024 were enrolled， among whom there were 86 patients with compensated liver cirrhosis （CLC group） and 167 patients with chronic hepatitis C （CHC group）. The patients were treated with coblopasvir hydrochloride capsules （60 mg）/sofosbuvir tablets （400 mg） with or without ribavirin tablets for 12 weeks， and they were followed up for 12 weeks after drug withdrawal. The primary outcome measures were the rate of sustained virologic response at week 12 after treatment （SVR12） and safety， and the secondary outcome measures were the changes in liver function， renal function， blood routine， and liver stiffness measurements （LSM） after 4 weeks of treatment， after 12 weeks of treatment， and at 12 weeks after drug withdrawal. The independent-samples t test and the Mann-Whitney U test were used for comparison of continuous data between two groups， and the Friedman test was used for comparison between multiple groups， while the Bonferroni method was used for paired comparison within each group； the chi-square test was used for comparison of categorical data between two groups. The Logistic analysis was used to investigate related influencing factors. ResultsThe 253 patients with chronic HCV infection had a mean age of 49.38±8.65 years， and there were 151 male patients （59.7%）. Of all patients， 33.99% （86/253） had liver cirrhosis， 25.69% （65/253） had hypertension， 10.67% （27/253） had HIV infection， 8.70% （22/253） had diabetes， 3.95% （10/253） had liver cancer， 1.98% （5/253） had chronic hepatitis B， and 7.91% （20/253） were treatment-experienced patients. As for genotype distribution， 2.77% （7/253） had genotype 1， 12.65% （32/253） had genotype 2， 66.01% （167/253） had genotype 3， 16.60% （42/253） had genotype 6， and 1.98% （5/253） had unknown genotype. The patients had an overall SVR12 rate of 92.09%， with an SVR12 rate of 93.02% in the CLC group and 91.02% in the CHC group. The multivariate logistic regression analysis showed that age （odds ratio ［OR］=1.086， 95% confidence interval ［CI］： 1.007 — 1.170， P=0.032） and HCC （OR=9.178， 95%CI： 1.722 — 48.912， P=0.009） were independent influencing factors for sustained virologic response. Compared with baseline data， the CLC group had significant reductions in alanine aminotransferase （ALT） （χ²=107.103， P0.05）， aspartate aminotransferase （AST） （χ²=90.602， P0.05）， and LSM （χ²=42.235， P0.05） after 12 weeks of treatment， while the CHC group had significant reductions in total bilirubin （χ²=15.113， P0.05）， ALT （χ²=202.237， P0.05）， AST （χ²=161.193， P0.05）， and LSM （χ²=37.606， P0.05）. The incidence rate of serious adverse events was 1.58%， and none of the patients withdrew from drug therapy； the patients with such events were relieved after active symptomatic treatment. The incidence rate of all adverse events was 23.72%， among which fatigue （17.39%） and nausea （2.37%） were the most common adverse events， and these events often disappeared within 2 weeks or were gradually relieved after symptomatic treatment. ConclusionCoblopasvir hydrochloride capsules/sofosbuvir tablets with or without ribavirin tablets has good efficacy and safety in the treatment of chronic HCV infection.

ObjectiveTo investigate the association between albumin （Alb） and recompensation by comparing recompensation rate between hepatitis B/C virus-related decompensated liver cirrhosis patients with different Alb levels， and to provide guidance for the identification and management of high-risk patients in clinical practice. MethodsRelated clinical data were collected from 734 patients with hepatitis B/C virus-related decompensated liver cirrhosis who attended The Third People’s Hospital of Kunming from January 1， 2016 to December 31， 2022， and they were divided into three groups based on the level of Alb. The linear regression analysis and chi-square test were used for trend tests. The Kaplan-Meier curve was plotted for the cumulative incidence rate of recompensation in the three groups， and the log-rank test was used for comparison between groups. A Cox proportional-hazards regression model analysis was used to investigate the association between Alb and recompensation in patients with hepatitis B/C virus-related decompensated liver cirrhosis. ResultsAmong the 734 patients with hepatitis B/C virus-related decompensated liver cirrhosis， 270 achieved recompensation， with a recompensation rate of 36.8%. All patients had a median Alb level of 29.90 （25.90 — 34.80） g/L on admission， and according to the level of Alb， they were divided into <25.9 g/L group with 177 patients， 25.9 — 34.8 g/L group with 377 patients， and >34.8 g/L group with 180 patients； 36 patients （20.3%） in the <25.9 g/L group， 138 （36.6%） in the 25.9 — 34.8 g/L group， and 96 （53.3%） in the >34.8 g/L group achieved recompensation， and the recompensation rate increased with the increase in Alb level （χ²=41.730， P<0.001）. After adjustment for all confounding factors， compared with the <25.9 g/L group， there was a significant increase in the incidence rate of recompensation in the 25.9 — 34.8 g/L group （hazard ratio ［HR］=1.842， 95% confidence interval ［CI］： 1.274 — 2.663） and the >34.8 g/L group （HR=2.336， 95% CI： 1.575 — 3.463）. The Kaplan-Meier survival analysis showed that there was a significant difference in the cumulative incidence rate of recompensation between the three groups （χ²=41.632， P<0.001）. ConclusionAlb level is an influencing factor for recompensation in patients with hepatitis B/C virus-related decompensated liver cirrhosis， and the recompensation rate increases with the increase in Alb level.

Objective:To provide evidence for medical institutions to explore the standardized management in Investigator Initiated Trials (IIT) through understanding the problems and providing suggestions of IIT management in the past 10 years.Methods:The publication year, region, content, existing problems, and suggestions were abstracted and analyzed by Bibliometric analysis.Results:58 studies were included in the analysis from 12 regions, and the top three regions were Shanghai, Beijing, and Guangdong Province. There were 10 items of IIT management, including management status, ethical management, process management, information management, contract management, approval management, human genetic resource management, researcher management, multi-center management, and methodology system management. The problems and suggestions of each item of IIT management were summarized respectively.Conclusions:The imperfect supervision system of IIT has brought great difficulties to the management department. Medical institutions should clarify management points and risk points from the item of IIT management and take targeted measures to actively promote high-quality IIT research.

Objective To address the challenges of poor performance and lack of interpretability in existing models,the SHAP algorithm is used to develop an interpretable machine learning model that offers a novel approach to wound age estimation,Methods Based on the previous discovery of the expression of 35 wound age healing-related genes in contused skeletal muscle,the woun age estimaton model was constructed using four algorithms,namly,Multilayer Perceptron(MLP),Random Forest(RF),LightGBM(LGBM),and Support Vector Machine(SVM).The SHAP(Shapley Additive Explanation)algorithm was used to rank the importance of genetic features,eliminate redundant attributes,and optimize the model for accurate wound age estimation.the genetic features of the optimal model were analyzed using SHAP's local interpretation capabilities.Results The best results were obtained using model of MLP(area under the curve(AUC)=0.99)The wound ages were classified into four categories:4～12 h,16～24 h,28～36 h,and 40～48 h,using only 15 gene features.According to SHAP analysis,Fam210a was identified as the most relevant gene.Local analysis revealed that high expression of Fam210a contributed to an increase in the predicted probability of 4 h～12 h,while high expression of Rae1 contributed to an increase in the predicted probability of 16 h～24 h.Additionally,low expression of Tbx18 contributed to an increase in the predicted probability of 28 h～36 h,whereas high expression of Tbx18 contributed to an increase in the predicted probability of 40 h～48 h.Conclusions The combined MLP and SHAP model can be used to predict wound age.Using the SHAP interpreter can better understand the degree of contribution of feature genes to the model prediction,and lay the foundation for further in-depth study of wound age estimation.