A Fact-Finding Procedure Integrating Machine Learning and AHP Technique to Predict Delayed Diagnosis of Bladder Patients with Hematuria

A new method for disease diagnosis based on hierarchical BRB with power set

Disease diagnosis occupies an important position in the medical field. The diagnosis of the disease is the basis for choosing the right treatment plan. Doctors must first diagnose what the patient has based on the clinical characteristics of various diseases, and then they can administer the right medicine. When building models for disease diagnosis, models are required to be able to handle various uncertainty information. The belief rule base (BRB) can effectively handle various information under uncertainty by introducing belief distributions. However, in current research, BRB-based disease diagnosis models still have problems of combinatorial rule explosion and inability to deal with local ignorance effectively. Therefore, a hierarchical BRB with power set (H-BRBp)-based disease diagnosis model is proposed in this paper. First, the physiological indexes and data of the patients were analyzed, and the data were preprocessed using the principal component regression (PCR) algorithm. Second, the H-BRBp disease diagnosis model was constructed to solve the deficiencies in the above BRB disease diagnosis model. Finally, the validity and advantages of the model were verified by experiments on lumbar spine disease diagnosis and a large number of comparison experiments.

Clinicopathological Characteristics and Prognostic Factors of Primary Bladder Signet Ring Cell Carcinoma

Introduction

The aim of this study is to examine the treatment pattern and predictors of long-term survival of patients with primary signet ring cell carcinoma (PSRCC) of the urinary bladder based on the analysis of the SEER database.

Methods

The 3-year and 5-year overall survival (OS) and cancer-specific survival (CSS) were calculated using the Kaplan–Meier method. Then, we compared the CSS curves by the log-rank test. The independent risk factors were determined using univariate and multivariate Cox regression.

Results

The 3-year OS and CSS rates for PSRCC of the bladder were 25.3% and 33.3%. The 5-year OS and CSS rates for the entire cohort were 16.4% and 25.2%. The CSS rates, respectively, were 0, 25.0, 66.7, 33.2, 42.4, and 31.7% at 3 years and 0, 25.0, 34.3, 24.1, 27.2, and 31.7% at 5 years for none, transurethral resection of the bladder (TURB), partial cystectomy, radical cystectomy with reconstruction, pelvic exenteration, and other surgeries (P = 0.001). Multivariate analyses showed independent risk factors only including T stage, M stage, lymph node removal, and surgical approach.

Conclusions

T stage, M stage, lymph node removal, and surgical approach are independent risk factors of PSRCC of the urinary bladder. TURB and radical cystectomy with reconstruction appear to provide a better outcome.