Application Of Machine Learning In Disease Diagnosis

With the development and progress of society,people pay more and more attention to their own health.However,due to the pollution of the global environment and the accelerated pace of life,many diseases have followed,which has seriously endangered human health.Therefore,it is very practical to study a stable,reliable and efficient disease-assisted diagnosis system.The purpose of machine learning is to use the prior experience and available data to improve the performance of the machine.It can actively learn from the sample information and then make high-accuracy decisions,so it has been widely studied and applied in disease diagnosis.This paper mainly discusses the application of machine learning in disease diagnosis,and provides some practical reference significance for disease-assisted diagnosis system and medical intelligence diagnosis.In this paper,three case data and four common machine learning algorithms are selected for discussion.The algorithms include logistic regression,support vector machine,adaptive boosting algorithm and extreme gradient boosting algorithm.The diagnostic model was constructed by using the above three algorithms in three case data,and the performance of each model was evaluated according to the evaluation method and evaluation indicator.The evaluation method uses K-fold cross-validation and is also used to select model parameters.In addition,the evaluation indicators refer to the classification accuracy,error rate,precision,recall rate and F₁-score of the model.By comprehensively comparing the results of each evaluation,it is found that the diagnostic model based on the extreme gradient boosting algorithm has the best robustness,the robustness of the other three algorithms is poor,and the evaluation results between the three case data are very different.