xiaotingsong / dGLCN

The codes are from MICCAI 2022, "Dual-graph Learning Convolutional Networks for Interpretable Alzheimer’s Disease Diagnosis"
18 stars 3 forks source link

why subjects'features is n*d #2

Open Huahuoy opened 1 year ago

Huahuoy commented 1 year ago

I have a question about your code and data, in the mat format file of the ROI extracted from fmri, the rows represent the time and the columns represent the ROI, which is supposed to be a matrix, but I see that your code in features.mat gets the features for all people, n*d, then the features for each subject is a one dimensional vector, may I ask How did you get that?

xiaotingsong commented 1 year ago

Thank you for reaching out with your question. In our dataset, the element in the ith row and jth column represents the jth ROI feature of the ith subject. The process to arrive at this feature includes several preprocessing steps such as spatial distortion correction, skull-stripping, and cerebellum removal. Subsequently, the data is segmented into gray matter, white matter, and cerebrospinal fluid. We then utilize the AAL template for further refinement. I hope this clarifies your question regarding the structure of the features.mat file. Please feel free to reach out if you have any further questions or need additional details.