Abstract Usually, the kidneys can be affected by renal masses or space-occupying lesions (LOE). When reference is made to the term renal mass, all benign and malignant processes that occupy, distort and affect the renal parenchyma and its environment are included, regardless of etiology, shape and volume. Therefore, renal masses include all cystic formations (abscesses), calculi, pseudotumors, neoplasms, inflammatory diseases and traumatic lesions. Thus, for the evaluation of cystic renal masses in medical imaging, according to their characteristics such as their wall (thin, irregular, thickened), septa (thin, irregular, thickened), borders (defined or not) and size, classifications such as Bosniak's classification shown in Table 1 are used, which classifies renal cysts into five categories based on the appearance of the image, to help predict whether it is a benign or malignant tumor.

Abstract Objective Accurate preoperative differential diagnosis of fat⁃poor angiomyolipoma (fp⁃AML) and clear cell renal cell carcinoma (ccrcc) is essential for proper treatment planning. In order to increase the accuracy of discrimination of fp⁃AML from ccrcc, we develop a classification model based on radiomics technology. Methods The study retrospectively collected CT images of 18 cases with fp⁃AML and 42 cases with ccrcc from department of radiology, the Second Affiliated Hospital of Suzhou University. Firstly, 430 radiomics features were extracted from CT images. Then, the feature selection was carried by three steps: Pearson’s correlation matrices were calculated to remove redundant features, Welch’s t⁃test was utilized to determine the statistically significant features, and sequential forward floating selection method was used to select the discriminative features. Finally, k⁃nearest neighborhood, random forest, support vector machine and adaboost classifiers were built for classification. Results The model built by SVM classifier achieved the best classification performance, with accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and area under the receiver operating characteristic curves of 91.67%, 88.89%, 92.86%, 84.21%, 95.12%, and 0.9418. Conclusions The proposed model can increase the classification accuracy of discrimination of fp⁃AML from ccrcc, and has great potential in helping radiologists to discriminate fp⁃AML from ccrcc.

Abstract Objective Acute kidney injury (AKI) is one of the most common complications and fatal factors in intensive care unit (ICU). Accurate prediction of AKI risk and identification of key factors related to AKI can provide effective guidance for clinical decision-making and intervention for patients with AKI risk. Methods A total of 30 020 patients in ICU (including 17 222 AKI patients and 12 798 Non-AKI patients) were selected from the public database MIMIC-III in this study, and basic information, physiological and biochemical indicators, drug use, and comorbidity during their stay in ICU were collected. All patients were randomly divided into training sets and independent testing sets according to the ratio of 4:1, and logistic regression, random forest, and lightgbm were applied to construct models for AKI predication in three time points including 24 h, 48 h and 72 h, respectively. The 10-fold cross validation was used to train and validate various models to predict the occurrence of AKI, and obtain important features. Furthermore, 24 h prediction models were used to predict AKI every 24 h during the 7-day window. Results lightgbm achieved the best performance with AUC values of 0.90, 0.88, 0.87 for 24 h, 48 h, and 72 h prediction, respectively, and F1 values were 0.91, 0.88, and 0.86. In prediction of every 24 h, the success rates of identifying AKI patients were 89%, 83%, and 80% in one day, two days and three days in advance, respectively. It was found that the length of stay in ICU, body weight, albumin, systolic blood pressure, bicarbonate, glucose, white blood cell count, body temperature, diastolic blood pressure and blood urea nitrogen played vital roles in predicting AKI for ICU patients. Using only 24 important features, the models could still achieve prominent prediction performance. Conclusions Based on basic information, physiological and biochemical indicators, drug use, and comorbidity, machine learning methods can be adopted to effectively predict AKI risk for ICU patients at several time points, and determine the dominant factors relative to AKI.

Abstract Objective To investigate the thickness of glomerular basement membrane (GBM) in patients withminimal change disease (MCD). Methods Select MCD patients aged 41 to 50 years without hema-turia, hypertension, diabetes, or hereditary kidney disease, and initially treated. Renal biopsy specimens were fixed as routine, embedded ultrathin sectioned, and observed under transmission electron microscope. For each sample, we took 10 to 15 photos according to the principle of equidistant curve movement. Test lines were randomly set. The vertical lines through the points where the test lines intersect with the podocyte side of GBM were made. The distance between the inner and outer intersections of the basement membrane was the section width of the GBM which represents the thickness of the GBM. Results The thickness of the GBM for 30 samples was 314.21±42.22 (256.97~393.51) nm, among which male was 360.30±47.0 (256.97~452.43) nm and female was 314.21±42.22 (256.97 ~393.51) nm. There was significant difference between the two groups (P = 0.009). The correlation be-tween GBM thickness and age was not significant. Conclusions The thickness of the GBM in newly diag-nosed MCD patients aged 41 to 50 years is constant, thicker in males than in females.

Abstract Objective: To analyze the impact of “Internet Plus”oriented continuous nursing intervention on hemodialysis self-management ability (HSMA) of patients with uremia and its countermeasures.Methods: 60 uremia patients admitted to hemodialysis in the hospital from January to December 2018 were selected as the control group (using routine continuous nursing intervention); 60 uremia patients admitted to hemodialysis from January to December 2019 were also selected as the observation group (using "Internet Plus"oriented continuous nursing intervention); the changes in the score values of the two groups of patients according to the self-management scale (SMSH) and chronic disease health literacy after intervention respectively. Results: After intervention，the self-management score of the patients in the observation group in terms of problem solving，emotional processing and self-care was higher than that of the control group (P<0.05)，and the score value in terms of information acquisition ability，improvement of health willingness, communication and interaction ability was higher than that of the control group (P<0).0.05).Conclusion:the continuity of nursing intervention based on “Internet plus”self-management enhances the self-management ability of patients with uremic hemodialysis and improves their health literacy.

Abstract Introduction: There is currently an increase in urinary tract infections in kidney transplant recipients due to multidrug-resistant organisms (MRO), which have become a medical challenge. Objective: To describe the prevalence of urinary tract infection (UTI) due to RMO in hospitalized renal transplant patients (PTxR), their risk factors, treatment and evolution at 1 year. Material and methods: medical records and cultures of hospitalized PTxR infectious with OMR in the period between 1/1/2016 and 31/12/2017 were reviewed. Risk factors such as: gender, advanced age, prolonged presence of double J catheter, surgical complications and prolonged hospitalization and renal function at hospitalization, at discharge and at one year and the occurrence of rejections at one year were evaluated. Results: The presence of multidrug-resistant germs was found in 58 PTxR (31.18%) who presented 105 episodes of UTI, 36 had a single infection and 22 P had more than one. 55.17% (32) were male and the mean age was 50.52 ±14.24 years. Of the total number of patients, 43 (74.15%) had risk factors such as: late removal of the double J catheter in 8 (13.8%), surgical complications in 11 (18.9%), prolonged inter- nation in 12 (20.7%) and 18 (31.03%) were older than 60 years. Nine patients required dialysis, 4 of whom recovered renal function. Creatinine at hospitalization in patients who did not require dialysis was 1.8 (1.39 - 3.01) mg/dl; at discharge 1.5 (1.1 - 2.1) mg/dl (p=0.025) and at one year it was 1.5 (1.18 - 2.1) mg/dl with no significant difference with respect to that at discharge (p=0.089). In the annual follow-up 5 patients died and 5 lost the graft. The incidence of rejection was 15.51%. The germs rescued were 13 A. baumanii cpx. (ABA) (11.92%), E. coli (ECO) 24 (22.01%), Enterobacter spp. 4 (3.66%), Enterococ- cus spp. 3 (2.75%), Klebsiella spp. 58 (53.21%), Serratia spp. 5 (4.58%), Proteus spp. 1 (0.91%) and Pseudomonas aeruginosa (PAE)1 (0.91). Of the 105 episodes of UTI, 79 were treated with monotherapy: 57 with carbapenem (54.28%), 10 with Colistin (9.51%), 4 with Linezolid (3.8%), 4 with Piperacillin+Tazobactam (3.8%), 3 with Ciprofloxacin (2.85%) and 1 with Nitrofurantoin (0.95%). In 26 episodes combined therapies of Carbapenem were used in 21 cases, colis- tin in 14, amikacin in 13, fosfomycin in 2 and tigecycline in 1 and ciprofloxacin in another. Conclusion: ORM UTIs were frequent and similar to those described in other series. No differences were found in the evolution of renal function, in rejections, in mortality in ORM UTIs with or without associated risk factors, nor was there any influence of recurrent or recurrent UTIs. Further studies with a larger number of patients are needed to evaluate the prognosis and evolution of patients with these infections.

Abstract Medical devices are playing a significant role in modern medicine. Devices are divided into four classes according to their risk levels. Class 1 are general medical devices whose risk levels are extremely low, such as in vitro diagnostic devices. Class 2 are controlled medical devices whose risks are relatively low such as MRI devices or electronic endoscopes. Class 3 are specially controlled medical devices whose risk levels are relatively high such as dialyzer. And Class 4 are invasive devices threatening patients’ lives.