Combining Multiple Clinical Neuroimaging Characteristics to Diagnose Early Alzheimers Disease:Validated Evaluation Using Random Forests
With the increase of the aging population,there are more and more people suffering from dementia.Given the importance of identifying dementia prodromes for future treatment efforts,two methods,including resting-state functional MRI(rs-fMRI)and voxel-based morphometry(VBM),are proposed to discriminate two types of mild cognitive impairment(MCI): amnestic MCI(aMCI)subtypes and dysexecutive MCI(dMCI)subtypes.Although two methods are analyzed and compared in specific data using the statistical approach,these outcomes may not be applicable in all clinical data.In order to verify the applicability of two methods,random forests(RF)were performed to classify different testing data.With the hybrid framework,we can achieve classification accuracies of 77.42%(AUC = 0.8101)between aMCI and healthy controls,and 82.14%(AUC = 0.8473)between dMCI and healthy controls.If comparing two MCI subtypes against each other,the accuracy can reach 79.57%(AUC = 0.8410).The present results suggest that multi-modality methods of pattern matching reach a clinically relevant accuracy for the a priori diagnosis in MCI subtypes.
Mild cognitive impairment resting state functional magnetic resonance imaging functional connectivity voxel-based morphometry random forests
Hong-Yuan Tzou Shih-Ting Yang Jiann-Der Lee
Department of Electrical Engineering Chang Gung University Taoyuan,Taiwan,333,R.O.C.
国际会议
The 2014 ICME International Conference on Complex Medical Engineering (CME2014)ICME复合医学工程国际会议
台北
英文
250-255
2014-06-26（万方平台首次上网日期，不代表论文的发表时间）