Drug Information Collection And Multi Drug Combination Analysis In AI Diagnosis And Treatment Of Diseases

In recent years,with the rapid development of artificial intelligence technology,the practical application of AI diagnosis and treatment model related to diseases and drugs is emerging.At the same time,due to the increase in the number of patients with chronic diseases,especially patients with multiple diseases,the combined use of drugs has become the norm.However,due to the interaction between drugs in different degrees,the reactions caused by antagonistic or synergistic effects will have different degrees of impact on patients’ health and clinical treatment effect.Thyroid nodule(TN)is a common disease in endocrine system.With the incidence rate of thyroid nodules increasing year by year,the related treatment and research are attracting more and more attention.Drugs play an important role in the treatment and recurrence control of thyroid nodules.Therefore,it is of great clinical significance to study the correlation between drugs and thyroid nodules.However,in the actual scene,it is difficult for the hospital to track and obtain the actual medication information of patients,which brings difficulties for the analysis of combined medication.From the patient’s point of view,although the doctor will tell the patient how to use the medicine,due to the poor compliance of patients,the treatment failure or unsatisfactory effect still occur from time to time.The main manifestations of patients’ medication non-compliance are:medication time error,missed medication,premature withdrawal,medication dose error,random dressing change,etc.The primary reason for non-compliance is that patients do not get prompt or guidance in the process of self medication.In view of the above problems,this paper studies and realizes the drug information collection in AI diagnosis and treatment of diseases,on this basis,a multi drug combination analysis method for hyperthyroid nodule recurrence was designed.The main contributions of this paper are as follows:(1)The prototype APP of medication reminder and information collection based on Android is designed and implemented,including the modules of medication knowledge graph,prescription management,medication reminder and medication time planning.The problems of medication noncompliance at the patient level and medication information collection at the doctor level are dealt with.(2)In the prototype APP of medication reminder and information collection,the method of medication rationality analysis and safety time period planning is proposed.Through the analysis of medication rationality,we can find the drug overdose or interaction relationship between the entered drug components by calling the knowledge map,and automatically detect and remind the above problems.Safety time period planning,according to the constraints of the drugs to be taken,establishes the planning algorithm,plans all the medication time periods,obtains the feasible safe medication time period,and provides a scientific basis for medication reminder.(3)Based on the collection of medication information,this paper studies the multi drug combination analysis method for thyroid nodule recurrence.The screening method of clinical medication data and the matrix representation method of patients’ historical medication data were proposed.Based on the matrix representation of patients’ historical medication information,the prediction model of thyroid nodule recurrence based on Text CNN and the clustering analysis model guided by convolution neural network are realized.The statistical analysis of the clustering results shows the correlation between patients’ combined medication information and thyroid sarcoidosis recurrence,which proves the reference value of the prediction model in the medical field.