Objective: The purpose of graded care for chronic kidney disease is to share expert experience, so that doctors can more accurately diagnose chronic kidney disease, so that patients with chronic kidney disease can understand their condition in time and collect case data. The
collected case data is established into a data warehouse, the data quality is evaluated, and the BP neural network method is used for data mining to analyze the data. Methods: The paper studied BP neural network and probabilistic neural network (PNN), and used 75% of the samples to
compare the models. The model errors were analyzed including maximum, minimum, expectation, variance and running time to get Adaboost. The accuracy and robustness of the -PNN model and the IGABP model are good. Results: BP neural network model and probabilistic neural network method
can achieve higher application of graded care for chronic kidney disease. The method is capable of quickly predicting disease grading and providing a standardized treatment care regimen. The method realizes the main functions of querying, managing, and collecting data of medical records. Conclusion:
The external expansion function of BP neural network and probabilistic neural network can achieve accurate data analysis, which can effectively improve the diagnosis time and grade prediction accuracy of chronic kidney disease, and provide opinions for graded nursing.
Prediction of Chronic Kidney Disease Using Adaptive Hybridized Deep Convolutional Neural Network on the Internet of Medical Things Platform
