search

tjiagoM / adni_phenotypes

Identifying healthy individuals with Alzheimer neuroimaging phenotypes in the UK Biobank
Other
4 stars 0 forks source link

Questions about data preprocessing #2

Open 2001923 opened 1 month ago

2001923 commented 1 month ago

Hi, thank you for sharing your work! I’m currently working with the ADNI dataset and encountered an issue during the preprocessing steps. I have access to the dataset, but I'm struggling to find or generate the 'raw_collated_freesurfer.csv' file. Could you please provide more details on how to obtain or derive this specific file? Additionally, if there are any prerequisites or specific steps I need to follow, I’d appreciate your guidance on that as well.

Thanks!

tjiagoM commented 1 month ago

Hi! Unfortunately, as we say in our repository and paper, we are not allowed to publicly release the data used from ADNI, NACC, and the UK Biobank.

In the paper, you can see where we asked to access the data in the Data availability section. Furthermore, in our paper Method's ADNI preprocessing section, we describe what steps we took using Freesurfer to generate the variables we used in the model.

In the file datasets.py you can see which cortical and volume variables we used from processing the datasets (i.e., variables FEATURES_CORTICAL, FEATURES_VOLUME, and FEATURES_VOLUME_EXTRA).

Inside this repository's data/ folder you can see how we preprocessed each dataset. For instance, in Preprocessing ADNI data.ipynb you can see how we preprocessed the data we got from ADNI. The file raw_collated_freesurfer.csv is thus just the collated variables resulting from the FreeSurfer steps.

I hope this helps and thanks for your interest in our work!