Method of providing diagnostic information on Alzheimer's disease using brain network
The present invention relates to a method of providing diagnostic information for Alzheimer's disease using a brain network.
1. A method of providing diagnostic information for classifying of Alzheimer's disease progression, including;
1) extracting a first feature;
2) constructing a brain network graph using graph theory;
3) converting the brain network graph to a feature vector using node2vec graph embedding;
4) selecting a second feature;
5) classifying Alzheimer's disease progression; and
6) evaluating the classification result,
wherein the first feature is extracted from one selected from the group consisting of electroencephalography (EEG), functional magnetic resonance imaging (fMRI), single-photon emission computed tomography (SPECT) and positron emission tomography (PET),
the graph embedding includes sampling, skip-gram, and computing embedding, and
the selection is selected from the group consisting of support vector machine-recursive feature elimination (SVM-RFE), least absolute shrinkage and selection operator (LASSO), feature selection with adaptive structure learning (FSASL), local learning and clustering based feature selection (LLCFS) and pairwise correlation based feature selection (CFS).
2. The method of claim 1 , wherein the evaluation is verified with one or more selected from the group consisting of accuracy (ACC), sensitivity (SEN), and specificity (SPE).
3. The method of claim 1 , wherein the classification uses one or more selected from the group consisting of regularized extreme learning machine (RELM) and linear support vector machine (LSVM).
4. The method of claim 1 , wherein Alzheimer's disease progression is one selected from the group consisting of healthy control, mild cognitive impairment and Alzheimer's disease.