Objective To improve the accuracy and clinical interpretability of recurrence prediction after liver transplantation in patients with hepatocellular carcinoma（HCC），and to explore the potential application of explainable machine learning models in integrating multimodal data. Methods This study included data from 138 liver transplant patients with liver cancer at the Third Medical Center of the Chinese People's Liberation Army General Hospital from December 2018 to December 2021. Preoperative contrast-enhanced CT radiomics features and clinical variables were extracted. Predictive models were developed using four machine learning methods ：LASSO regression，random forest，support vector machine（SVM），and neural network. Importance of variate was employed to identify key predictive factors. Model performance was evaluated using area under the receiver operating characteristiccurve（AUC），Brier score，and calibration curves. A nomogram was ultimately constructed based on important variables. Results The AUCs for the random forest model in predicting recurrence at 1，2，and 3 years were 0.881，0.906，and 0.915，respectively，significantly outperforming other models. The importance analysis identified five key imaging features，and the nomogram model combining these with clinical variables demonstrated good consistency and predictive capability. Conclusion Explainable machine learning models based on multimodal data can effectively improve the accuracy and transparency of recurrence prediction following liver transplantation in HCC patients. These models have strong clinical applicability and provide valuable support for individualized preoperative risk assessmentand treatment decision-making. 

Objective To explore the association between immune therapy and the risk of rejection after liver transplantation in patients with hepatocellular carcinoma （HCC）. Methods A retrospective analysis was conducted on 11 HCC patients who received immune therapy before transplantation at Beijing Friendship Hospital from 2019 to 2025， and 70 cases reported in the literature were included. The incidence of rejection and influencing factors were analyzed. Results The rejection rate in our center's cohort was 9.1% （1/11）， which was lower than that in theneoadjuvant therapy cohort （25.0%）and the post-transplantation cohort （30.8%）in the literature. The rejection rate was significantly reduced when the interval between the last dose of immune therapy and transplantation was more than two halflives. The expression of programmed death ligand 1（PD-L1）in the graft may be related to the risk of rejection and could serve as a potential biomarker for predicting rejection. Conclusion It is recommended to wait at least 8 weeks （two halflives）after immune therapy before liver transplantation. The expression levels of programmed death receptor 1（PD-1）/PDL1 in the graft tissue can be used as potential biomarkers， but further research is needed to explore the mechanism and safe time intervas. 

Objective To explore the physiological and psychological factors affecting pulmonary infection in kidney transplant recipients and to construct a risk prediction model. Methods The clinical and follow-updata of 327 recipients undergoing kidney transplantation in Shandong Provincial Qianfoshan Hospital from January2019 to January 2024 were retrospectively analyzed. The recipients were divided into pulmonary infection group（102 cases）and non-pulmonary infection group（225 cases）according to whether pulmonary infection occurred after kidney transplantation. The data of the two groups of recipients was analyzed with multivariate regression analyze，theprediction model of pulmonary infection of kidney transplant recipients was constructed，and the receiver operatingcharacteristic（ROC）curve was used to verify the predictive value. Results The rate of pulmonary infection within 1 years after kidney transplantation was 31.19%. Age，blood glucose，insomnia severe index，use of cyclosporine and mycophenolate mofetil were independent risk factors for pulmonary infection. While white blood cell count on postoperative day 5 and positive psychological capital were independent protective factors（P ＜ 0.05）.The areaunder curve（AUC）of the constructed model equation for predicting the risk of pulmonary infection was 88.7% with a Kappa value of 0.66（P ＜ 0.05）. Conclusion The prediction model of pulmonary infection in kidney transplant recipients constructed based on physiological and psychological factors has a good predictive value. 

Objective To investigate the clinical efficacy of peptide receptor radionuclide therapy （PRRT）in patients with recurrent neuroendocrine tumors （NETs） after liver transplantation. Methods A retrospective analysis was conducted on the clinical data of 5 patients with recurrent NETs after liver transplantation who underwent PRRT from April 2023 to May 2024 in Tianjin First Central Hospital. Adverse reactions and treatment efficacy were evaluated. Results The median follow-up was 15.6 months. Among the patients，1 achieved partial response （PR），3 had stable disease （SD）， and 1 experienced progressive disease （PD）. Liver function remained stable during treatment， and no rejection reactions occurred. Grade 3 neutropenia was observed in 2 cases， while grade 2 neutropenia occurred in 3 cases. Symptomatic treatment alleviated the condition in 4 cases， but 1 case had persistent severe neutropenia， leading to treatment discontinuation. Grade 1 ～ 2 proteinuria was observed in 2 cases. Conclusion PRRT is safe and effective for patients with recurrent NETs after liver transplantation. The main side effect is neutropenia， and PRRT can be considered as one of the multimodal treatment options. 

Objective To explore the molecular mechanisms and potential inflammation-related biomarkers of hepatic ischemia reperfusion injury（HIRI）process using bioinformatics techniques. Methods HIRI datasets GSE12720 and GSE14951 were obtained. Differential expressed genes（DEGs）were analyzed for functional enrichment and pathway enrichment. Hub genes were obtained using protein interactions analysis，and the correlationbetween Hub genes and inflammatory genes was calculated to finalize the key genes. The biological processes involved in these key genes were explored using GSEA analysis，and the expression of key genes was verified by mouse HIRI model. Finally，miRNAs of the key genes were predicted. Results Signaling pathways such as TNF，IL-17，MAPK and NF-κB were involved in inflammatory and immune processes during HIRI，and five key genes were closely related to inflammatory responses and their expressions are upregulated after HIRI. Conclusion This studyconfirms the central role of the inflammatory response in the process of HIRI and provides new perspectives for the discovery of the pathogenesis and potential therapeutic targets of HIRI. 

Objective To analyze the research hotspots and development trends of artificial intelligencein the field of organ transplantation. Methods Literatures on the application of artificial intelligence inthe field of organ transplantation from January 2004 to June 2024 were retrieved from the Web of ScienceCore Collection and China National Knowledge Infrastructure database. CiteSpace 6.2.R6 software was used for visual analysis. Results A total of 728 relevant literatures were retrieved，559 English literatures and 143 Chinese literatures were included after screening. The number of literatures published in the early stage was small，while the number of literatures published in recent years has increased significantly. The country with the largest number of articles was the United States，followed by China. The top three institutions in terms of foreignpublications were Mayo Clinic，University of Toronto and Harvard University. Zhejiang University，Peking UnionMedical College and the First Affiliated Hospital of Sun Yat-sen University were the most active institutions in China. The authors who have published the most articles in English were Cheungpasitporn Wisit and Thongprayoon Charat，and the author who has published the most articles in Chinese was Ren Bin. The main high-frequency keywords in English were machine learning，survival，artificial intelligence，liver transplantation，kidney transplantation，mortality，risk，model，outcome and deep learning. The main high-frequency keywords in Chinese were machine learning， liver transplantation， kidney transplantation， deep learning， artificial intelligence， hepatocellular carcinoma，radiomics，liver cancer，prognosis and tacrolimus. The first emergent keywords in English were artificial neural networks，data mining and survival analysis，and the most recent were predictive model and kidney transplant. The first emergent keywords in Chinese were neural networks，cyclosporine A，and blood concentration，and the most recent were deep learning and machine learning. Conclusion The application of artificial intelligence in thefield of organ transplantation is increasing significantly. The use of machine learning and deep learning to construct prediction models and to analyze the survival of various organ transplantation patients is a current research hotspots. In the future，exchanges and cooperation between countries and disciplines can be strengthened to promote the study and application of advanced AI technologies，so as to further promote the development of this field. 

Objective Split liver transplantation （SLT） is an important innovative technique in transplant surgery that addresses the shortage of donor livers and expands the source of donor organs. Currently， the number of hepatocellular carcinoma patients undergoing SLT is increasing year by year， but its impact on the occurrence of complications and prognosis in hepatocellular carcinoma recipients compared with whole liver transplantation （WLT） remains unclear. This study aims to explore the differences in the occurrence of complications and prognosisbetween WLT and SLT recipients with hepatocellular carcinoma， and to construct a prognostic prediction model forhepatocellular carcinoma recipients underwent SLT by analyzing the prognostic factors. Methods A retrospective selection of 3 773 adult recipients with hepatocellular carcinoma who underwent WLT and SLT registered in the ChinaLiver Transplant Registry （CLTR） from January 2015 to December 2023 was conducted. Clinical data of donors and recipients were collected to compare the clinicopathological characteristics of donors and recipients， the occurrence ofcomplications and prognosis of recipients between the two surgical methods was evaluated. Propensity score matching （PSM） at a 2 ：1 ratio was used to match the baseline clinical data with differences between the WLT and SLT groups，and the incidence of complications and prognosis in the two groups were analyzed. Univariate and multivariate analyses were performed to screen for risk factors related to the prognosis of SLT recipients， and then a prognostic prediction model

was constructed. Results There was no significant difference in the overall survival rate and recurrence-free survival rate between SLT recipients and WLT recipients （P ＞ 0.05）. However， the incidence of postoperative biliary leakage，portal vein embolism， and new-onset diabetes after surgery in SLT recipients were significantly higher than those in theWLT group （postoperative biliary leakage，P ＝ 0.0010 ； portal vein embolism，P ＝ 0.0044 ； new-onset diabetes after surgery，P ＝ 0.036）. Univariate and multivariate analyses of the prognosis of SLT recipients found that donor-recipient

ABO blood type incompatibility， recipient alpha-fetoprotein （AFP）＞ 100 ng/ml， and cumulative tumor diameter ≥ 8 cm were independent risk factors for the prognosis of SLT hepatocellular carcinoma recipients （all P ＜ 0.05）. A prognostic prediction model for SLT recipients was constructed based on the above risk factors， with AUC values of 0.561，0.698， and 0.687 for 1-year，3-year， and 5-year survival rates， respectively. Conclusion The incidence of some postoperative complications in SLT recipients is higher than that in WLT recipients， but the clinical prognosis is similar. Among them，donor-recipient ABO blood type incompatibility， recipient AFP ＞ 100 ng/ml， and cumulative tumor diameter ≥ 8 cm areindependent risk factors affecting the prognosis of SLT hepatocellular carcinoma recipients. The prediction model establishedon this basis has good predictive value for the survival rate of recipients. 

Objective Maintaining stable concentrations of anti-rejection drugs represents a critical facetof post-kidney transplantation patient care; however，achieving personalized and precise management for each patient remains challenging. This study leverages an artificial intelligence-based deep learning framework to develop a machine learning predictive model for tacrolimus concentration，with the objective of recommending optimal dosing regimens for individual kidney transplant recipients. Methods Fifty kidney transplant recipients who underwentsurgery at the Urology Department of Beijing Friendship Hospital，Capital Medical University，between January 2024 and April 2025，were enrolled in this study. Drawing on prior experience in tacrolimus dosing，we collected data on patients' gender，age，weight，comorbidities，CYP3A5 metabolic phenotypes，initial tacrolimus（Tac）doses，and FK506 levels measured on postoperative days 7，9，11，13，and 15，with subsequent dosage adjustments made according to each concentration measurement. A LightGBM regression model was employed to predict and optimize tacrolimus dosing regimens. Results Among the 50 kidney transplant recipients enrolled in this study，none developed severe complications，including delayed recovery of graft function，postoperative infections，or bleedingThe dataset was partitioned into training and validation sets using a five-fold cross-validation approach. The final model demonstrated robust predictive performance in the test set，with a mean absolute error（MAE）of 0.166，root mean square error（RMSE）of 0.227，mean absolute percentage error（MAPE）of 7.035%，P20 of 0.935，P30 of 0.97，and a coefficient of determination（R-squared）of 0.932. Conclusion The LightGBM regression model exhibited excellent performance，providing a novel and effective strategy for personalized tacrolimus dosage adjustment in kidney transplant recipients. 

Objective To explore the effect of early attribution training on liver transplant recipients after surgery. Methods A retrospective analysis was conducted on 240 liver transplant recipients who underwent livertransplantation surgery at the Second Affiliated Hospital of Guangxi Medical University from January 1，2023 toDecember 31，2024. According to the random number method，the recipients were divided into two groups，with 120cases in each group. The control group received routine nursing care，while the intervention group received attributiontraining in addition to the routine nursing care. The changes in attribution style，sleep quality，and medical coping strategies between the two groups before and after 4 weeks of intervention were compared. Results After 4 weeks，the total attribution score，internal and external factors，prevalence，and stability of intervention groups were higherthan those of the control group（P ＜ 0.05）；The scores of surrender，avoidance，and facing in medical copingstrategies were higher than those in the control group（P ＜ 0.05）. Conclusion Early attribution training for liver transplant recipients can improve their attribution style，alleviate emotional management difficulties，improve sleep quality，and enhance their medical coping strategies. 

Objective To analyze the influence of tumor pathological grade on the prognosis of patientswith urothelial carcinoma （UC） after kidney transplantation （KT）. Methods A total of 90 KT recipients whounderwent surgery and were pathologically diagnosed as UC in the Department of Urology， Beijing Friendship Hospital，Capital Medical University from January 1997 to December 2021 were included in this study. The patients' age，gender， time from transplantation to tumor occurrence， immunosuppressant use， tumor location， tumor TNM stage，tumor pathological grade， tumor multiplicity， postoperative survival and recurrence were collected. The patients in this study， the recipients who developed UC after KT were divided into three groups according to the location of the tumor ： non muscle invasive bladder cancer， upper urinary tract urothelial carcinoma， and utuc+ bladder cancer. Eachgroup was divided into high-grade tumor group and non-high-grade tumor group according to the tumor pathological grade of the patients. The main differences in clinical characteristics between the two groups were compared， and then the prognosis of KT recipients between the two groups was analyzed. Results There were 11 cases of non-muscle invasive bladder cancer and 5 cases of high-grade tumor. There were no significant differences in age （P ＝0.081）， gender （P ＝ 0.242）， time from transplantation to tumor occurrence （P ＝ 0.734） and tumor multiplicity （P ＝ 0.545） between high-grade tumor and non-high-grade tumor groups. Only one patient with non-high-grade

tumor died 85 months after the occurrence of tumor， and the rest of the recipients survived during the follow-up period. There were 49 cases of upper urinary tract urothelial carcinoma and 38 cases of high-grade tumor. There were no significant differences in age （P ＝ 0.951）， gender （P = 0.400）， time from transplantation to tumor occurrence （P ＝ 0.206）， tumor multiplicity （P ＝ 0.729）， and T stage （P = 0.073） between the two groups. There were 30 cases of upper urinary tract urothelial carcinoma + bladder cancer at the same time， including 13 cases of high-grade tumors. There were no significant differences in age （P ＝ 0.741）， gender （P ＝ 0.355）， time from transplantation to tumor occurrence（P ＝ 0.783）， and T stage （P ＝ 0.488） between the two groups. The overall survival of patients with high-grade tumors wassignificantly higher than that of patients with non high-grade tumors （P ＝ 0.006）. There was no significant difference inrecurrence between the two groups. Conclusion This study found that high-grade tumor was a high-risk factor affecting the prognosis of KT recipients only when the tumor occurred simultaneously in the bladder and upper urinary tract， while it had little impact on the prognosis of patients when the tumor occurred only in the bladder or upper urinary tract.