Prediction And Application Of Hyperthyroidism Development Based On Machine Learning

With the development of society and economy,people are paying more and more attention to their own health while improving their living standards,they all hope to obtain convenient and high-quality healthcare services.However,due to the scarcity of high-quality medical resources and the disparity in medical resources distribution,it’s difficult for current medical services to meet the needs of patients universally.In this situation,the telemedicine service system came into being.The current telemedicine service system realized “face-to-face” consultations between experts and patients,experts and medical personnel with the help of remote communication technologies,electronic and computer technologies,which is of great significance to give full play to the technological advantages of large medical centers and improve the efficiency and effectiveness of medical treatment.However,in the online communication between doctors and patients,it often happens that patients cannot find suitable physicians because they do not understand their own condition.If we can combine the disease prediction model based on artificial intelligence technology with the traditional telemedicine service system,the intelligence and effectiveness of the current medical service system will be further improved.As a chronic disease with high incidence,hyperthyroidism has become one of the major research projects in modern endocrinology.During the time of clinical diagnosis and treatment,many patients with hyperthyroidism delayed their best treatment time because they didn’t understand their condition in time.In order to help patients check their own condition in time and prevent the worsening of hyperthyroidism,it is extremely necessary to carry out prediction research on the development of hyperthyroidism.At present,in the field of thyroid medicine,there are many research results that use artificial intelligence techniques to risk evaluation and disease prediction,but none of these studies provided a service application platform that corresponds to the model.Most of the research data are mainly based on ultrasound images,pathological diagnosis and medication,which often ignored the influence of blood test indicators on patient’s condition.Most of the models are based on linear regression and deep learning algorithms,which cannot classify and predict the development of patient’s condition.In view of the above problems,this paper proposes a hyperthyroidism staging prediction model and a hyperthyroidism development prediction model based on the change rules of blood indicators,and studies how to implement a model-based medical service prototype system.First,this paper proposed a hyperthyroidism staging prediction model.According to the clinical medical research of hyperthyroidism,this paper established the classification rules which based on the change laws of blood indicator FT3 FT4 TSH and TRAB,summed up the dependencies between these phases.Further,based on the rules and linkages,this paper designed a staging prediction algorithm for hyperthyroidism and established a hyperthyroidism staging prediction model used patient’s blood indicators value as input,which laid a foundation for the subsequent physician recommendation algorithm.Secondly,this paper achieved a hyperthyroidism development prediction model based on the blood indicator data of patients.By analyzing the clinical data of a comprehensive Grade-A tertiary hospital over the past 10 years,this paper extracted 37-dimensional features which contained the rules and characteristics of hyperthyroidism patient’s disease development,and established a classify prediction model based on SVM support vector machine and K nearest neighbor algorithm.The model use the basic information of hyperthyroidism patients and the blood indicator values during the first six-months to predict patient’s future development of hyperthyroidism,which has provided the technical support for assist doctor to give accurate clinical diagnosis.Finally,this paper implemented a medical service prototype system based on the open source Rocket.Chat remote communication platform.This paper combined the traditional medical service system with Rocket.Chat to realize the online interaction between doctors and patients,and embedded the hyperthyroidism development prediction model into the prototype system,realized the model-based disease prediction services.This paper designed a physician recommendation algorithm based on the hyperthyroidism staging prediction model,which fairly assigned the physician resources to the patients in different degrees,balanced the matching ratio between doctors and patients,and improved the efficiency of hospital visits.