An intelligent data-driven model for disease diagnosis based on machine learning theory

摘要: In the era of data, major decisions are determined by massive especially in healthcare industry. this paper, an intelligent data-driven model is proposed based on machine learning theory, specifically, support vector (SVM) and random forest (RF). The then applied to a case disease diagnosis, cough variant asthma (CVA). data 137 samples with 12 attributes collected for experiments. results show that achieves better prediction performance than single SVM RF. Besides, order identify key medical indicators enhance diagnosis accuracy efficiency, most important factors affecting CVA generated model, including FENO, EOS%, MMEF75/25, FEV1/FVC, PEF, etc. Meanwhile, it demonstrated could be user-friendly tool improve diagnosis.