search

sgadgil6 / cnslab_fmri

CNS (Computational Neuroscience) Lab project for age/sex classification of fMRI scans
80 stars 17 forks source link

How to get the dataset? #2

Closed MengzhangLI closed 3 years ago

MengzhangLI commented 4 years ago

Hi, thanks for sharing your excellent work of MICCAI 2020.

I want to re-implement this project, but I don't know how to get the data used in paper.

Will you share or plan to share the data related with this paper?

Thank you!

sgadgil6 commented 4 years ago

Hello, For HCP, you should be able to download the data "Resting State fMRI 1 Preprocessed" from https://db.humanconnectome.org/. The GIFTI images were then parcellated using https://balsa.wustl.edu/QnXj

Here are the final processed BOLD signal files: https://drive.google.com/file/d/1c4UYvp089KwqAllAHdHMHGjq8y5x882D/view. I will add that to the README.

If you need the NCANDA data, you should directly contact last author Dr. Kilian Pohl (kilian.pohl@stanford.edu) for data usage.

MengzhangLI commented 4 years ago

Hello, For HCP, you should be able to download the data "Resting State fMRI 1 Preprocessed" from https://db.humanconnectome.org/. The GIFTI images were then parcellated using https://balsa.wustl.edu/QnXj

Here are the final processed BOLD signal files: https://drive.google.com/file/d/1c4UYvp089KwqAllAHdHMHGjq8y5x882D/view. I will add that to the README.

If you need the NCANDA data, you should directly contact last author Dr. Kilian Pohl (kilian.pohl@stanford.edu) for data usage.

Hi, Soham: Thanks for your nice reply, can't wait to re-implement this project.

Best,

MengzhangLI commented 3 years ago

Hello, For HCP, you should be able to download the data "Resting State fMRI 1 Preprocessed" from https://db.humanconnectome.org/. The GIFTI images were then parcellated using https://balsa.wustl.edu/QnXj

Here are the final processed BOLD signal files: https://drive.google.com/file/d/1c4UYvp089KwqAllAHdHMHGjq8y5x882D/view. I will add that to the README.

If you need the NCANDA data, you should directly contact last author Dr. Kilian Pohl (kilian.pohl@stanford.edu) for data usage.

Hi Soham: I download HCP data from this google drive. It is a folder with 1096 cases each of them is a numpy array (22* 1200 just as your paper shows). However, in your code it is:

train_data = np.load('data/train_data_1200_1.npy') train_label = np.load('data/train_label_1200_1.npy') test_data = np.load('data/test_data_1200_1.npy') test_label = np.load('data/test_label_1200_1.npy')

What should I do if I want to run your model on HCP data？

Best,

Mengzhang Li

MengzhangLI commented 3 years ago

Sorry for my questions.

I forgot to see "preprocessing.py".