Study on the Integrated Classification of Alzheimer′s Disease in Cerebral Cortical Thickness
CUI Shu-hua1，HU Bin1，2，HU Tao2
1（Department of Computer Science and Technology，Shandong Normal University，Jinan 250000，China）2（School of Information Science and Engineering，Lanzhou University，Lanzhou 730000，China）
Abstract：In this paper，the Adaboost integrated classification method was used to differentiate the patients with mild cognitive impairment and normal healthy people by using magnetic resonance image data（SMRI）based on the thickness of the cerebral cortex.In this paper，100 healthy people and 104 patients with mild cognitive impairment were selected as normal control group and experimental group，the structural magnetic resonance images of the two groups were analyzed.Firstly，the feature selection method based on two independent samples T-test is used to filter the noise characteristics，and outlier detection method is used to analyze and eliminate the abnormal data.Finally，the Adaboost integrated classification method was used to classify and the accuracy of the classification was as high as 89.37% by leaving a cross validation.Experiments show that the above method has a significant effect on distinguishing patients with mild cognitive impairment and normal healthy people，which will help to achieve early diagnosis of Alzheimer′s disease.
崔书华,胡斌,胡涛. 阿尔茨海默病在脑皮层厚度中的集成分类方法研究[J]. 小型微型计算机系统, 2017, 38(12): 2652-2657.
CUI Shu-hua,HU Bin,HU Tao. Study on the Integrated Classification of Alzheimer′s Disease in Cerebral Cortical Thickness. Journal of Chinese Computer Systems, 2017, 38(12): 2652-2657.
2017, Vol. 38 
