ObjectiveTo investigate the application value of a machine learning model based on preoperative clinical indicators in predicting the risk of hypoalbuminemia after partial hepatectomy. MethodsA retrospective analysis was performed for the clinical data of 700 patients who underwent partial hepatectomy in Nanfang Hospital， Southern Medical University， from January 2018 to January 2023， including demographic data， history of underlying diseases， tumor characteristics， preoperative laboratory markers， and perioperative indicators. The research data were divided into a training set and a test set at a ratio of 7∶3. The two-independent-samples t test was used for comparison of normally distributed continuous data between two groups； the two-independent-samples Wilcoxon rank-sum test was used for comparison of continuous data with skewed distribution between two groups； the chi-square test or the Fisher’s exact test was used for comparison of categorical data between two groups. The least absolute shrinkage and selection operator （LASSO） regression analysis was used to identify characteristic variables， and 7 machine learning algorithms were used to construct predictive models， i.e.， logistic regression， decision tree， artificial neural network， K-nearest neighbors （KNN）， support vector machine， eXtreme gradient boosting， and light gradient boosting machine. The receiver operating characteristic （ROC） curve and the area under the ROC curve （AUC） were used to assess the discriminatory ability of models， and the DeLong test was used for comparison of AUC. The calibration curve and decision curve analysis were used to assess the calibration and clinical practicability of models， and the models were compared with albumin-bilirubin （ALBI） score and Model for End-Stage Liver Disease （MELD） score. SHapley Additive exPlanations （SHAP） were used to interpret the key influencing factors for the optimal model. ResultsA total of 700 patients were finally enrolled， 283 （40.42%） developed hypoalbuminemia after surgery. The LASSO regression analysis identified 8 predictive factors of age， hepatitis B， fatty liver， blockade time， preoperative albumin （Alb）， time of operation， intraoperative blood loss， and preoperative aspartate aminotransferase （AST）. Among the 7 machine learning models， the KNN model showed the best overall predictive performance， with an AUC of 0.835 （95% confidence interval： 0.781 — 0.889）， a sensitivity of 84.0%， and a specificity of 65.5% in the test set. ALBI and MELD scores had an AUC of 0.652 and 0.524， respectively， and the KNN model had a better predictive performance than these two scores （Z=5.309 and 8.945， both P <0.001）. The calibration curve showed good consistency between predicted probabilities and actual incidence rates， and the decision curve analysis showed that the KNN model had net clinical benefit across a wide threshold range. The SHAP analysis showed that preoperative Alb， hepatitis B， time of operation， and age were the most significant influencing factors， and a synergistic effect was observed between hepatitis B and age/time of operation. ConclusionThe KNN machine learning model constructed based on preoperative clinical indicators can effectively predict the risk of hypoalbuminemia after partial hepatectomy and has a better performance than traditional scoring models， which provides a reference for the early identification of high-risk patients in clinical practice.

Objective:To explore potential categories of disease progression fear and self-management ability in patients with maintenance hemodialysis (MHD) and their influencing factors.Methods:Convenience sampling was used to select MHD patients in the Blood Purification Center of the Affiliated Nantong Hospital of Nanjing University of Chinese Medicine, the Affiliated Suzhou Hospital of Nanjing University Medical School, and the Affiliated Huaian First People's Hospital of Nanjing Medical University from November 2022 to May 2023 as study subjects. General Information Questionnaire, Fear of Progression Questionnaire-short form (FoP-Q-SF), Hemodialysis Self-management Instrument (HDSMI), and Perceived Social Support Scale (PSSS) were used to conduct the survey. Mplus 8.7 software was applied to conduct latent profile analysis, and multiple Logistic regression was used to explore the factors influencing the disease progression fear and self-management ability of different MHD patients.Results:A total of 380 questionnaires were distributed, and 364 valid questionnaires were recovered, with an effective recovery rate of 95.79%. The mean scores of FoP-Q-SF and HDSMI in 364 MHD patients were (2.53±0.84) and (2.81±0.54), respectively. MHD patients' fear of disease progression and self-management ability were manifested in three different potential categories, named high fear-general self-management (59.34%, 216/364), moderate fear-low self-management (24.45%, 89/364), and low fear-good self-management (16.21%, 59/364). Multiple Logistic regression showed that age, education level, marital status, comorbidities, duration on dialysis, and social support were influencing factors in the potential categories of fear of disease progression and self-management ability in MHD patients, and the difference was statistically significant ( P<0.05) . Conclusions:Healthcare professionals should give appropriate interventions to MHD patients with different types of fear of progression and self-management ability in order to reduce patients' fear of disease progression and improve their self-management ability.

Objective:To explore the trajectories of kinesiophobia in initial hemodialysis patients and their influencing factors.Methods:A convenience sampling method was used to recruit patients undergoing their first hemodialysis treatment at Hemodialysis Center, the Affiliated Huai'an No.1 People's Hospital of Nanjing Medical University from June to December 2023. Baseline data were collected using a general information questionnaire, the Tampa Scale for Kinesiophobia (TSK), the Dialysis Patient Symptom Burden Index (DFSSBI), the Fear of Progression Questionnaire-Short Form (FoP-Q-SF), the Multidimensional Fatigue Inventory (MFI-20), and the Pittsburgh Sleep Quality Index (PSQI). The TSK was used to measure kinesiophobia levels at 1, 3, and 6 months post-dialysis. Data were analyzed using a latent class growth mixture model, univariate analysis, and multinomial Logistic regression.Results:A total of 251 patients were surveyed, and 239 completed follow-ups. Three latent classes of kinesiophobia trajectories were identified: persistently high kinesiophobia group (32.22%, 77/239), moderate kinesiophobia-declining group (28.45%, 68/239), and persistently low kinesiophobia group (39.33%, 94/239). Multinomial Logistic regression analysis indicated that age, marital status, presence of sleep disorders, history of falls in the past year, symptom burden, fear of disease progression, and fatigue significantly influenced kinesiophobia trajectory classes ( P<0.05) . Conclusions:Kinesiophobia in initial hemodialysis patients exhibits distinct trajectories. Healthcare professionals should implement targeted interventions based on the influencing factors of kinesiophobia trajectory development.

Objective To explore the development trajectory and root cause of kinesiophobia in patients with diabetic peripheral neuropathy.Methods By the convenience sampling method,240 diabetic peripheral neuropathy patients who were hospitalized in the Department of Endocrinology of a tertiary A hospital in Jiangsu Province from May 2022 to November 2022 were selected.The baseline data of the patients are collected using a general questionnaire,the Tampa Scale for Kinesiophobia,the Chinese version of the Multidimensional Fatigue Inventory-20,the Pittsburgh Sleep Quality Index,the Social Support Rate Scale,and the Numerical Rating Scale for Pain.At the time of discharge,and at 3,6,and 12 months post-discharge,the level of kinesiophobia in the patients is assessed.The growth mixture model was used to classify the trajectory of kinesiophobia and analyze its root causes.Results Ultimately,227 patients in this study completed the survey.The trajectories of kinesiophobia in patients with diabetic peripheral neuropathy were divided into 3 latent classes:persistent high kinesiophobia group(22.5％),moderate kinesiophobia decline group(44.9％)and persistent low kinesiophobia group(32.6％).The results of multivariate Logistic regression analysis showed that age,frequency of hypoglycemia,complications,sleep quality,fall history within a year,pain and fatigue score were influencing factors of the latent classes of kinesiophobia trajectories in patients with DPN(P＜0.05).Conclusion Within 12 months after discharge,the kinesiophobia of patients with diabetic peripheral neuropathy shows population heterogeneity.Medical staff should give targeted interventions according to the developmental trajectories of kinesiophobia and root cause analysis.

Objective:To explore potential categories of disease progression fear and self-management ability in patients with maintenance hemodialysis (MHD) and their influencing factors.Methods:Convenience sampling was used to select MHD patients in the Blood Purification Center of the Affiliated Nantong Hospital of Nanjing University of Chinese Medicine, the Affiliated Suzhou Hospital of Nanjing University Medical School, and the Affiliated Huaian First People's Hospital of Nanjing Medical University from November 2022 to May 2023 as study subjects. General Information Questionnaire, Fear of Progression Questionnaire-short form (FoP-Q-SF), Hemodialysis Self-management Instrument (HDSMI), and Perceived Social Support Scale (PSSS) were used to conduct the survey. Mplus 8.7 software was applied to conduct latent profile analysis, and multiple Logistic regression was used to explore the factors influencing the disease progression fear and self-management ability of different MHD patients.Results:A total of 380 questionnaires were distributed, and 364 valid questionnaires were recovered, with an effective recovery rate of 95.79%. The mean scores of FoP-Q-SF and HDSMI in 364 MHD patients were (2.53±0.84) and (2.81±0.54), respectively. MHD patients' fear of disease progression and self-management ability were manifested in three different potential categories, named high fear-general self-management (59.34%, 216/364), moderate fear-low self-management (24.45%, 89/364), and low fear-good self-management (16.21%, 59/364). Multiple Logistic regression showed that age, education level, marital status, comorbidities, duration on dialysis, and social support were influencing factors in the potential categories of fear of disease progression and self-management ability in MHD patients, and the difference was statistically significant ( P<0.05) . Conclusions:Healthcare professionals should give appropriate interventions to MHD patients with different types of fear of progression and self-management ability in order to reduce patients' fear of disease progression and improve their self-management ability.

Objective:To explore the trajectories of kinesiophobia in initial hemodialysis patients and their influencing factors.Methods:A convenience sampling method was used to recruit patients undergoing their first hemodialysis treatment at Hemodialysis Center, the Affiliated Huai'an No.1 People's Hospital of Nanjing Medical University from June to December 2023. Baseline data were collected using a general information questionnaire, the Tampa Scale for Kinesiophobia (TSK), the Dialysis Patient Symptom Burden Index (DFSSBI), the Fear of Progression Questionnaire-Short Form (FoP-Q-SF), the Multidimensional Fatigue Inventory (MFI-20), and the Pittsburgh Sleep Quality Index (PSQI). The TSK was used to measure kinesiophobia levels at 1, 3, and 6 months post-dialysis. Data were analyzed using a latent class growth mixture model, univariate analysis, and multinomial Logistic regression.Results:A total of 251 patients were surveyed, and 239 completed follow-ups. Three latent classes of kinesiophobia trajectories were identified: persistently high kinesiophobia group (32.22%, 77/239), moderate kinesiophobia-declining group (28.45%, 68/239), and persistently low kinesiophobia group (39.33%, 94/239). Multinomial Logistic regression analysis indicated that age, marital status, presence of sleep disorders, history of falls in the past year, symptom burden, fear of disease progression, and fatigue significantly influenced kinesiophobia trajectory classes ( P<0.05) . Conclusions:Kinesiophobia in initial hemodialysis patients exhibits distinct trajectories. Healthcare professionals should implement targeted interventions based on the influencing factors of kinesiophobia trajectory development.

Objective To explore the development trajectory and root cause of kinesiophobia in patients with diabetic peripheral neuropathy.Methods By the convenience sampling method,240 diabetic peripheral neuropathy patients who were hospitalized in the Department of Endocrinology of a tertiary A hospital in Jiangsu Province from May 2022 to November 2022 were selected.The baseline data of the patients are collected using a general questionnaire,the Tampa Scale for Kinesiophobia,the Chinese version of the Multidimensional Fatigue Inventory-20,the Pittsburgh Sleep Quality Index,the Social Support Rate Scale,and the Numerical Rating Scale for Pain.At the time of discharge,and at 3,6,and 12 months post-discharge,the level of kinesiophobia in the patients is assessed.The growth mixture model was used to classify the trajectory of kinesiophobia and analyze its root causes.Results Ultimately,227 patients in this study completed the survey.The trajectories of kinesiophobia in patients with diabetic peripheral neuropathy were divided into 3 latent classes:persistent high kinesiophobia group(22.5％),moderate kinesiophobia decline group(44.9％)and persistent low kinesiophobia group(32.6％).The results of multivariate Logistic regression analysis showed that age,frequency of hypoglycemia,complications,sleep quality,fall history within a year,pain and fatigue score were influencing factors of the latent classes of kinesiophobia trajectories in patients with DPN(P＜0.05).Conclusion Within 12 months after discharge,the kinesiophobia of patients with diabetic peripheral neuropathy shows population heterogeneity.Medical staff should give targeted interventions according to the developmental trajectories of kinesiophobia and root cause analysis.

Objective To explore the potential categories of self-management behavior in prediabetes patients and analyze the influencing factors of different potential categories.Methods A convenient sampling method was used to select pre-diabetic patients treated in physical examination centers and endocrinology outpatient clinics in 2 tertiary A hospitals in Jiangsu Province from September 2022 to June 2023.General data questionnaire,Prediabetic Disease Knowledge Questionnaire,Prediabetes Self-management Scale,Diabetes Self-efficacy Scale and Social Support Rating Scale were used for investigation.Results A total of 445 subjects were included in this study,with an average age of(44.92±4.98)years.4 potential categories of self-management behavior were identified,namely low self-management(16.4％),static self-management(19.1％),dynamic self-management(25.6％)and high self-management(38.9％).Occupational status,marital status,family per capita monthly income,family history of diabetes,disease knowledge,self-efficacy and social support were the influencing factors of self-management behavior in prediabetes patients(P＜0.05).Conclusion There are significant category characteristics of self-management behavior in patients with prediabetes.Medical staff can identify the category characteristics and influencing factors early,and provide bases for medical staff to formulate targeted clinical intervention measures to help patients improve their self-management level.

Objective To explore the potential categories of fatigue trajectory in patients with diabetic peripheral neuropathy and analyze the influencing factors.Methods Convenience sampling method was used to select 426 patients with diabetic peripheral neuropathy who were hospitalized in a tertiary hospital in Jiangsu Province from April 2022 to April 2023.General information and disease related data of the patients were collected,and the Chinese version of Multi-dimensional Fatigue Scale,Pittsburgh Sleep Quality Index Scale,Social support Rating Scale,pain number rating scale and Tampa scale of kinetophobia were used to investigate at discharge(T1).The Chinese version of multi-dimensional fatigue scale was used to investigate 3 months after discharge(T2),6 months after discharge(T3),and 12 months after discharge(T4).A mixed growth model was used to identify the potential categories of fatigue trajectory and logistic regression were employed to analyze the influencing factors of fatigue.Results 414 patients finished the study.Three categories of fatigue trajectory in the patients with diabetic peripheral neuropathy were identified:significantly increasing group(19.6%),slowly increasing group(47.8%),and decreasing group(32.6%).Logistic regression analysis showed that age,sleep,social support,number of comorbidities,body mass index(BMI),pain and fear of exercise were the factors influencing the potential categories of fatigue trajectory in the DPN patients.Conclusions There are three potential categories of fatigue trajectories in DPN patients.Age,sleep disorders,social support,comorbidities,BMI,pain score,and exercise fear were influencing factors of fatigue in patients with DPN.

Objective:To investigate the mid-to-long term therapeutic effects of allogeneic bone composite prosthesis reconstruction in patients with large bone defects after the resection of distal femoral tumors.Methods:From June 2013 to December 2018, a total of 19 patients with malignant tumors of the distal femur who underwent reconstruction with allogeneic bone composite prosthesis in the Department of Bone Tumor, Shanghai General Hospital were retrospectively collected. There were 10 males and 9 females, aged 22.3±11 years (range, 11-42 years). The mean body mass index was 19.3±3.4 kg/m 2 (range, 14-27 kg/m 2). There were 18 cases of osteosarcoma and 1 case of Ewing's sarcoma. According to Ennecking staging, there were 17 cases of stage IIB and 2 cases of stage III. The intraoperative blood loss and operation time were recorded, and the prosthesis and patient survival conditions and postoperative complications were observed. The limb function was evaluated by the Musculoskeletal Tumor Society (MSTS) 93 function score. Results:All patients successfully completed the operation. The operation time was 187.3±39.8 min (range, 110-260 min), the intraoperative blood loss was 284.9±87.0 ml (range, 200-500 ml), and the blood transfusion volume was 327±213 ml (range, 100-800 ml). The remaining length of the proximal femur was 153.7±26.6 mm (range, 93-190 mm), and the length of allogeneic bone was 84.1±24.6 mm (range, 39-134 mm). Among the 19 patients, 9 patients (47%) achieved bony union with an average healing time of 16.7±4.8 months (range, 10-25 months), and 7 patients had delayed healing with an average healing time of 18.4±4.0 months (range, 15-25 months). The remaining 10 cases were nonunion between allogeneic bone and host bone. All patients were followed up for 80.7±20.2 months (range, 56-121 months). During the follow-up, 3 cases died due to pulmonary metastasis of bone tumors, and the time of death was 57 months, 63 months, and 59 months after surgery, respectively. At the last follow-up, the patient survival rate was 84% (16/19), and the MSTS 93 function score of the 16 patients was (24.3±2.4) points (range, 21-28 points), with an excellent rate of 100% (16/16). Seven patients underwent revision surgery, 3 cases were aseptic loosening, 3 cases were prosthesis stem fracture at the junction of the allograft bone and the host bone, and 1 case was periprosthetic infection, among which the patient with periprosthetic infection had poor local soft tissue conditions due to preoperative radiotherapy, and the infection was controlled after two revision surgeries. Five cases were revised with allogeneic bone composite prosthesis, and 2 cases were revised with short-stem giant prosthesis with cortical steel plate or locking nail. After revision, the remaining length of the proximal femur was 143.4±31 mm (range, 91-175 mm), and the length of the allograft bone was 92.6±26.6 mm (range, 61-123 mm). The 7 revised patients were still in follow-up. There were no cases of pulmonary infection, nerve injury, deep vein thrombosis or other complications after surgery.Conclusion:The survival period of patients after the surgery to reconstruct large bone defects following the resection of malignant tumors at the distal end of the femur using allogeneic bone composite prosthesis is satisfactory, and the limb function is good. However, the incidence of prosthesis complications is high, which can be reconstructed through revision.