Here is a dataset with all the DICOM imagenes windowed and saved as png of 128x128 pixels.\nhttps://www.kaggle.com/guiferviz/rsna_stage1_png_128
\n\nThis notebook has more information about the process I've used to generate this dataset:\nhttps://www.kaggle.com/guiferviz/prepare-dataset-resizing-and-saving-as-png\nThe Kaggle notebook, due to runtime limitations (disk space) is running on a subsample of the data. If you want to replicate my full dataset creation use this Google Colab:\nhttps://colab.research.google.com/gist/guiferviz/50912a681776d5afe012b1a9259bd637/resize-dataset.ipynb\nIf you try to unzip the competition data in Google Colab you will also run out of space, so I've used the amazing tool fuse-zip to mount the zip and work with the files in it without extracting any of those.
\n\nOf course you can always use your own computer if you have enough capacity :)
\n\nNOTE: I've just realize that Tom already uploaded a png version of the dataset (https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/discussion/109747#latest-632186) but I've been working on this the last two days so here is my post :) The size of this images is also different.
", "messages": [ { "id": 632701, "postDate": "2019-09-24T00:13:07.310Z", "content": "Here is a dataset with all the DICOM imagenes windowed and saved as png of 128x128 pixels.\nhttps://www.kaggle.com/guiferviz/rsna_stage1_png_128
\n\nThis notebook has more information about the process I've used to generate this dataset:\nhttps://www.kaggle.com/guiferviz/prepare-dataset-resizing-and-saving-as-png\nThe Kaggle notebook, due to runtime limitations (disk space) is running on a subsample of the data. If you want to replicate my full dataset creation use this Google Colab:\nhttps://colab.research.google.com/gist/guiferviz/50912a681776d5afe012b1a9259bd637/resize-dataset.ipynb\nIf you try to unzip the competition data in Google Colab you will also run out of space, so I've used the amazing tool fuse-zip to mount the zip and work with the files in it without extracting any of those.
\n\nOf course you can always use your own computer if you have enough capacity :)
\n\nNOTE: I've just realize that Tom already uploaded a png version of the dataset (https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/discussion/109747#latest-632186) but I've been working on this the last two days so here is my post :) The size of this images is also different.
", "rawMarkdown": "Here is a dataset with all the DICOM imagenes windowed and saved as png of 128x128 pixels.\nhttps://www.kaggle.com/guiferviz/rsna_stage1_png_128\n\nThis notebook has more information about the process I've used to generate this dataset:\nhttps://www.kaggle.com/guiferviz/prepare-dataset-resizing-and-saving-as-png\nThe Kaggle notebook, due to runtime limitations (disk space) is running on a subsample of the data. If you want to replicate my full dataset creation use this Google Colab:\nhttps://colab.research.google.com/gist/guiferviz/50912a681776d5afe012b1a9259bd637/resize-dataset.ipynb\nIf you try to unzip the competition data in Google Colab you will also run out of space, so I've used the amazing tool *fuse-zip* to mount the zip and work with the files in it without extracting any of those.\n\nOf course you can always use your own computer if you have enough capacity :)\n\nNOTE: I've just realize that Tom already uploaded a png version of the dataset (https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/discussion/109747#latest-632186) but I've been working on this the last two days so here is my post :) The size of this images is also different.", "votes": 43 }, { "id": 639540, "postDate": "2019-10-03T10:18:01.870Z", "content": "I uploaded varying .png resolutions from 224x224 all the way up to 512x512. Please do make sure you check it out https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/discussion/110840#latest-638127
", "rawMarkdown": "I uploaded varying .png resolutions from 224x224 all the way up to 512x512. Please do make sure you check it out https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/discussion/110840#latest-638127", "votes": 1 }, { "id": 656304, "postDate": "2019-10-24T06:09:18.630Z", "content": "Great post that I almost missed, thanks @guiferviz and @vbookshelf (who ping me about this post)
", "rawMarkdown": "Great post that I almost missed, thanks @guiferviz and @vbookshelf (who ping me about this post)" }, { "id": 655201, "postDate": "2019-10-22T19:50:51.837Z", "content": "Have you tried using fuse-zip outside of Kaggle environment?
Thanks
", "rawMarkdown": "Thanks", "votes": 1 }, { "id": 639455, "postDate": "2019-10-03T08:22:26.040Z", "content": "It's relly helpful to me. Thanks a lot
", "rawMarkdown": "It's relly helpful to me. Thanks a lot", "votes": 1 }, { "id": 635156, "postDate": "2019-09-27T07:58:50.330Z", "content": "Thank you so much! Nice work.
", "rawMarkdown": "Thank you so much! Nice work.", "votes": 1 } ], "comments": [ { "id": 639540, "author_name": "Tim Yee", "author_url": "", "post_date": "2019-10-03T10:18:01.870000", "content": "I uploaded varying .png resolutions from 224x224 all the way up to 512x512. Please do make sure you check it out https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/discussion/110840#latest-638127
", "votes": 1, "replies": [] }, { "id": 656304, "author_name": "Neuron Engineer", "author_url": "", "post_date": "2019-10-24T06:09:18.630000", "content": "Great post that I almost missed, thanks @guiferviz and @vbookshelf (who ping me about this post)
", "votes": 0, "replies": [] }, { "id": 655201, "author_name": "William Green", "author_url": "", "post_date": "2019-10-22T19:50:51.837000", "content": "Have you tried using fuse-zip outside of Kaggle environment?
Thanks
", "votes": 1, "replies": [] }, { "id": 639455, "author_name": "Cho Royou", "author_url": "", "post_date": "2019-10-03T08:22:26.040000", "content": "It's relly helpful to me. Thanks a lot
", "votes": 1, "replies": [] }, { "id": 635156, "author_name": "Hue Nguyen", "author_url": "", "post_date": "2019-09-27T07:58:50.330000", "content": "Thank you so much! Nice work.
", "votes": 1, "replies": [] } ], "raw_markdown_by_id": { "632701": "Here is a dataset with all the DICOM imagenes windowed and saved as png of 128x128 pixels.\nhttps://www.kaggle.com/guiferviz/rsna_stage1_png_128\n\nThis notebook has more information about the process I've used to generate this dataset:\nhttps://www.kaggle.com/guiferviz/prepare-dataset-resizing-and-saving-as-png\nThe Kaggle notebook, due to runtime limitations (disk space) is running on a subsample of the data. If you want to replicate my full dataset creation use this Google Colab:\nhttps://colab.research.google.com/gist/guiferviz/50912a681776d5afe012b1a9259bd637/resize-dataset.ipynb\nIf you try to unzip the competition data in Google Colab you will also run out of space, so I've used the amazing tool *fuse-zip* to mount the zip and work with the files in it without extracting any of those.\n\nOf course you can always use your own computer if you have enough capacity :)\n\nNOTE: I've just realize that Tom already uploaded a png version of the dataset (https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/discussion/109747#latest-632186) but I've been working on this the last two days so here is my post :) The size of this images is also different.", "639540": "I uploaded varying .png resolutions from 224x224 all the way up to 512x512. Please do make sure you check it out https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/discussion/110840#latest-638127", "656304": "Great post that I almost missed, thanks @guiferviz and @vbookshelf (who ping me about this post)", "655201": "Have you tried using `fuse-zip` outside of Kaggle environment?", "632811": "", "650422": "Thanks", "639455": "It's relly helpful to me. Thanks a lot", "635156": "Thank you so much! Nice work." } }