Search

Background/Aims: Renal biopsy is the gold standard to determine the pathologic type of primary nephrotic syndrome, which is critical for diagnosis, choice of treatment and evaluation of prognosis. However, in some cases, renal biopsy cannot be performed. Methods: To explore the possibility of predicting the histology type of primary nephrotic syndrome without the need for biopsy, we trained and validated a machine learning algorithm using data from 222 patients with biopsy-confirmed primary nephrotic syndrome treated at our hospital between May 2008 and January 2016. The model was then tested prospectively on another sample of 63 patients with biopsy-confirmed primary nephrotic syndrome. Results: Overall accuracy of prediction from the retrospective set of 222 patients was 62.2% across all types of nephrotic syndrome. The accuracy of model prediction for the prospectively collected dataset of 63 patients was 61.9%. The algorithm identified 17 of 33 variables as contributing strongly to type of renal pathology. Conclusion: To our knowledge, this is the first such application of machine learning to predict the pathologic type of primary nephrotic syndrome, which may be clinically useful by itself as well as helpful for guiding future efforts at machine learning-based prediction in other disease contexts.

Background: Recently, research has linked Helicobacter pylori ( H. pylori ) stomach infection to colonic inflammation, mediated by toxin production, potentially impacting colorectal cancer occurrence., Aim: To investigate the risk factors for post-colon polyp surgery, H. pylori infection, and its correlation with pathologic type., Methods: Eighty patients who underwent colon polypectomy in our hospital between January 2019 and January 2023 were retrospectively chosen. They were then randomly split into modeling ( n = 56) and model validation ( n = 24) sets using R. The modeling cohort was divided into an H. pylori -infected group ( n = 37) and an H. pylori -uninfected group ( n = 19). Binary logistic regression analysis was used to analyze the factors influencing the occurrence of H. pylori infection after colon polyp surgery. A roadmap prediction model was established and validated. Finally, the correlation between the different pathological types of colon polyps and the occurrence of H. pylori infection was analyzed after colon polyp surgery., Results: Univariate results showed that age, body mass index (BMI), literacy, alcohol consumption, polyp pathology type, high-risk adenomas, and heavy diet were all influential factors in the development of H. pylori infection after intestinal polypectomy. Binary multifactorial logistic regression analysis showed that age, BMI, and type of polyp pathology were independent predictors of the occurrence of H. pylori infection after intestinal polypectomy. The area under the receiver operating characteristic curve was 0.969 [95% confidence interval (95%CI): 0.928-1.000] and 0.898 (95%CI: 0.773-1.000) in the modeling and validation sets, respectively. The slope of the calibration curve of the graph was close to 1, and the goodness-of-fit test was P > 0.05 in the two sets. The decision analysis curve showed a high rate of return in both sets. The results of the correlation analysis between different pathological types and the occurrence of H. pylori infection after colon polyp surgery showed that hyperplastic polyps, inflammatory polyps, and the occurrence of H. pylori infection were not significantly correlated. In contrast, adenomatous polyps showed a significant positive correlation with the occurrence of H. pylori infection., Conclusion: Age, BMI, and polyps of the adenomatous type were independent predictors of H. pylori infection after intestinal polypectomy. Moreover, the further constructed column-line graph prediction model of H. pylori infection after intestinal polypectomy showed good predictive ability., Competing Interests: Conflict-of-interest statement: The authors have no conflicts of interest to declare., (©The Author(s) 2024. Published by Baishideng Publishing Group Inc. All rights reserved.)

Background/Aims: Renal biopsy is the gold standard to determine the pathologic type of primary nephrotic syndrome, which is critical for diagnosis, choice of treatment and evaluation of prognosis. However, in some cases, renal biopsy cannot be performed. Methods: To explore the possibility of predicting the histology type of primary nephrotic syndrome without the need for biopsy, we trained and validated a machine learning algorithm using data from 222 patients with biopsy-confirmed primary nephrotic syndrome treated at our hospital between May 2008 and January 2016. The model was then tested prospectively on another sample of 63 patients with biopsy-confirmed primary nephrotic syndrome. Results: Overall accuracy of prediction from the retrospective set of 222 patients was 62.2% across all types of nephrotic syndrome. The accuracy of model prediction for the prospectively collected dataset of 63 patients was 61.9%. The algorithm identified 17 of 33 variables as contributing strongly to type of renal pathology. Conclusion: To our knowledge, this is the first such application of machine learning to predict the pathologic type of primary nephrotic syndrome, which may be clinically useful by itself as well as helpful for guiding future efforts at machine learningbased prediction in other disease contexts. [ABSTRACT FROM AUTHOR]

Objective: The purpose of this study was to describe the ultrasonographic findings of angioleiomyoma based on pathological subtypes., Materials and Methods: Thirty-nine patients with subcutaneous angioleiomyomas in the extremities were retrospectively reviewed by two radiologists and a pathologist. Sonographic images were analyzed to evaluate each tumor's anatomic location, size, shape, margin, heterogeneity, echogenicity, associated findings, and vascularity., Results: Angioleiomyomas were divided into 3 subtypes: capillary (n = 16), venous (n = 22), and cavernous (n = 1). The one cavernous angioleiomyoma was a hypoechoic mass with rich vascularity. Hypoechogenicity was more frequently observed for venous tumors (77.3%) than for capillary tumors (43.8%), and isoechogenicity was more frequently observed for capillary tumors (56.2%) than for venous tumors (22.7%). Moderate vascularity was more frequently observed for venous tumors (59.1%) than for capillary tumors (12.5%), and little vascularity was more frequently observed for capillary tumors (62.5%) than for venous tumors (13.6%). The aforementioned findings including echogenicity ( p = 0.034) and vascularity ( p = 0.003) were statistically significant., Conclusion: Awareness of sonographic findings of angioleiomyomas based on pathologic subtypes could be helpful for diagnosing angioleiomyoma and could increase diagnostic accuracy for superficial soft-tissue masses in our practice.

Background/aims: Renal biopsy is the gold standard to determine the pathologic type of primary nephrotic syndrome, which is critical for diagnosis, choice of treatment and evaluation of prognosis. However, in some cases, renal biopsy cannot be performed., Methods: To explore the possibility of predicting the histology type of primary nephrotic syndrome without the need for biopsy, we trained and validated a machine learning algorithm using data from 222 patients with biopsy-confirmed primary nephrotic syndrome treated at our hospital between May 2008 and January 2016. The model was then tested prospectively on another sample of 63 patients with biopsy-confirmed primary nephrotic syndrome., Results: Overall accuracy of prediction from the retrospective set of 222 patients was 62.2% across all types of nephrotic syndrome. The accuracy of model prediction for the prospectively collected dataset of 63 patients was 61.9%. The algorithm identified 17 of 33 variables as contributing strongly to type of renal pathology., Conclusion: To our knowledge, this is the first such application of machine learning to predict the pathologic type of primary nephrotic syndrome, which may be clinically useful by itself as well as helpful for guiding future efforts at machine learning-based prediction in other disease contexts., (© 2017 The Author(s). Published by S. Karger AG, Basel.)

Background: Breast cancer is the most common cancer among women worldwide. Based on the latest Iranian national cancer department report, the total number of women registered with breast cancer was 6976 cases during 2007. Five year survival is one of the indicators used for evaluation of the quality for care to different types of malignancies including breast cancer., Objectives: The aim of this study was to estimate survival rate of breast cancer in 6147 Iranian patients at a national level in different geographic regions., Materials and Methods: 6147 cases of breast cancer, which had telephone number and were diagnosed between 2001-2006, were called to obtain information about their life status. Survival estimates were calculated using the Kaplan-Meier method, and the survival probability was calculated for the overall cohort and in different categories of gender, age and pathologic type of tumor. Hazard ratios (HR) according to demographic and risk variables were calculated by Cox's proportional hazard model., Results: The overall 5-year survival rate was 71.0%. The mean survival time was different between men and women, which was statistically significant. The number of men involved with breast cancer was 172 (2.8%) of all cases. The 5-year survival rate for patients in age group 41-50 years was significantly higher than other age groups (P = 0.001). The likelihood of death was higher in patients with 61 years old or more years rather than those below forty years old (HR = 1.31; 95% CI: 1.12-1.55)., Conclusions: THE FINDINGS OF THIS STUDY MIGHT HELP IRANIAN HEALTH MANAGERS: 1) to be more conscious about geographical and regional determinants which will affect overall survival rate. 2) To carry preventive activities such as public education particularly in Iranian men. 3) To think about screening and early detection of breast cancer.