Against the backdrop of rapid development in vascular science,automated diagnosis of carotid arteries has gained widespread attention from all sectors of society.However,in the face of complex environment,conventional methods will produce er-rors that cannot be ignored.In order to solve this problem,the feature set method is added to the extreme gradient lifting algorithm,and the weight factor is used to optimize the data mining technology to generate a fusion algorithm.The binary tree rule is added to the algorithm to generate a fusion algorithm.Finally,the experiment is carried out on Sclero data set and compared with three systems,such as Golden Sine.In one day,the power consumption of the converged system is 0.21 kW*h,which is the lowest among the four systems.After one month's diagnosis,the patients'carotid atherosclerosis scores were 2.8,3.0,3.1 and 3.4,respectively,which indicated that the proposed method had the best curative effect,and its blood flow rate score was 3.2,which indicated that the method had the highest adaptability to patients.The experimental results show that the fusion system proposed in this study has achieved the best results in experimental accuracy and diagnosis of patients'carotid hardness,and is suitable for diagnosis of patients with carotid atherosclerosis.
关键词
数据挖掘/权重因子/极限梯度提升算法/特征集合/二叉树规则/颈动脉硬化/自动诊断
Key words
data mining/weight factor/limit gradient lifting algorithm/feature set/bifurcation tree rule/carotid atherosclero-sis/automatic diagnosis
维普
万方数据