Method of AAV different treatment regimen prognosis prediction model based on AI technology

Abstract

The invention discloses a method of an AAV different treatment regimen prognosis prediction model based on an AI technology. The method comprises the following steps: S1, collecting kidney survival prognosis condition data of existing ANCA related nephritis patients after treatment by adopting different treatment regimens; S2, constructing a prognosis prediction model of each treatment regimen ina treatment regimen set by adopting a Cox regression analysis method and an XGBoost machine learning algorithm, training the constructed prognosis prediction model of each treatment regimen by adopting the kidney survival prognosis condition data after the ANCA related nephritis patients are treated by adopting different treatment regimens to obtain a high-quality Cox regression model and a high-quality XGBoost model for effective prognosis prediction capable of judging the kidney survival prognosis condition of each treatment regimen; S3, respectively inputting the medical characteristic dataof a to-be-evaluated patient into the high-quality Cox regression model and the high-quality XGBoost model of each treatment regimen in the treatment regimen set; and S4, selecting a proper treatmentregimen according to the kidney survival probability of each treatment regimen in the treatment regimen set and the reference value of whether the kidney of each treatment regimen in the treatment regimen set survives and the prognosis risk. According to the invention, the purpose of guiding a doctor to adopt an ACNC related nephritis treatment regimen with the lowest prognosis risk is achieved.

Description

technical field [0001] The present invention relates to the field of medical plan evaluation system, specifically, based on artificial intelligence technology, a method and model for constructing prognosis prediction models of different treatment plans for ANCA-related nephritis, which are used to guide clinicians to judge the prognosis of the disease and provide reference for selecting treatment plans . Background technique [0002] ANCA-associated nephritis (AAV for short) is mainly a kidney disease caused by ANCA-associated vasculitis. The related manifestations are low-grade fever, fatigue, arthralgia, etc. The related renal pathological manifestations are proteinuria, hematuria, and edema. In severe cases, there may be Hypertension and renal failure, etc., lung-related pathological manifestations include coughing up phlegm, coughing up blood, severe respiratory infection, etc., because ANCA-associated vasculitis is a systemic autoimmune disease that can affect multiple ...

AI Technical Summary

Problems solved by technology

There are differences in renal survival after doctors choose different treatment options for patients with ANCA-associated nephritis, and different treatment options will lead to different treatment outcomes and different renal survival rates. It is irreversible, so it is very necessary to effectively evaluate the treatment plan, and there is currently no effective model that can predict the prognosis of renal survival in different treatment plans based on the clinical characteristics of the patient at admission

Images