Prediction System Of Medical Cost For Gastric Cancer Based On Process Mining

At present,disputes caused by medical costs are widespread.How to use information means to provide accurate prediction of medical costs for serious diseases has become a research hotspot.As a common disease,gastric cancer has high medical costs and low cure rate.Using process mining method to mine and analyze the medical logs of gastric cancer can accurately predict the medical cost of gastric cancer.Using process mining technology to predict the medical cost of gastric cancer,we are faced with the following challenges:(1)the medical process is dynamic,the traditional workflow net can not reflect the updates and the number of repeated execution for nodes in the medical path.(2)The existing process mining algorithms regard the repeated nodes as the same nodes,and only count once.However,to present the medical scheme and predict medical cost of gastric cancer,it has to consider the repeated activities in the medical path of gastric cancer,such as multiple radiotherapy and chemotherapy,so it needs a new algorithm to solve this problem.(3)The traditional medical scheme evaluation system needs to be updated,because its evaluation factors are not suitable for the evaluation of gastric cancer medical pathway and cost prediction.In order to tackle these challenges,this paper has carried out the following works:(1)On the basis of Petri net,a dynamic medical path net is proposed to replace,split and merge medical activities,reflecting the execution of repeated nodes.(2)After preprocessing the medical log of gastric cancer,αtj algorithm mines the order of medical activities in the medical log,finds the repeated sequences,and then uses TNC algorithm to count the repeated execution times of the repeated sequences.Build a dynamic medical path net model for medical process about gastric cancer,and then the performance of the algorithms above is measured by consistency detection criteria.(3)Obtain the unit price of gastric cancer medical activities and the reimbursement ratio of "three directories" of medical insurance,and then construct the unit price data table of medical activities.This paper proposes a comprehensive evaluation model for medical schemes of gastric cancer based on four evaluation indexes.Traverse the medical path net of gastric cancer to obtain path about each optional scheme.Combined with the unit price data table of medical activities,the cost calculation formula and the type of medical insurance held by patients,it forecasts the cost of each candidate medical scheme,and recommends the comprehensive best scheme for patients by using the scheme evaluation formula(4)To sum up,we design and implement a cost prediction system of gastric cancer based on process mining.Input the basic information of patients,judge the type of patients,mine corresponding logs,conduct medical process modeling and evaluate of medical schemes,so as to predict the treatment cost of each alternative scheme and recommend the best scheme for patients.