Issue in downloading the dataset.

[Asaad-Pak]

Hello, I am downloading MIMIC CXR JPG in my personal google drive(in a folder) through "wget" command. I have sufficient space to store the data. Now it has been downloading since last 6 hours and it is downloading only index.html for each study in the data and not the jpg x ray images presently. As MIMIC CXR JPG version contains only x-rays in jpg format nothing else so it is supposed to download those jpg images for each study. So why is this happening ? Will it first download the index.html files first and then jpg images?

[Asaad-Pak]

Hi, @alistairewj @tompollard if you can answer this I would appreciate that. I am downloading using collab and using "wget" command provided but it downloads the index.html files not the actual jpgs presently. Secondly "-c" parameter in this command means that it if the download is interrupted it will continue downloading from the point it left now I am using collab and the session expires after some hours and when I restart the session and start downloading again it doesn't download new files it keeps on downloading or checking index.html files that were previously downloaded. Can you guide me about this?

[tompollard]

I am downloading using collab and using "wget" command provided but it downloads the index.html files not the actual jpgs presently

@Asaad-Pak see: https://github.com/MIT-LCP/mimic-code/issues/1012.

Short answer, I think you are downloading the files but it's taking a long time. I would suggest accessing the data via the Google Cloud bucket instead. See: https://physionet.org/content/mimic-cxr-jpg/2.0.0/#files for instructions.

The following Collab notebook may also be helpful ("Training a Convolutional Neural Network to Classify Chest X-rays"): https://github.com/MIT-LCP/2019-hst-953/blob/master/tutorials/mimic-cxr/mimic-cxr-train.ipynb