Browsing through MRA modality images in the dataset I noticed one image (SeriesUID: 1.2.826.0.1.3680043.8.498.10063454172499468887877935052136698373) that looked different from the others. I am not an expert in the field but it looks like a MIP:
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F13823597%2F1ba5611bf002465c65d765ec6783d437%2Fmra_image.png?generation=1753878620303418&alt=media\")
\nIs the presence of such images among MRA images to be expected in the test set as well? It seems like it is possible to identify such images in the DICOM metadata by looking at the ImageType tag (\"Derived\" for the image above), in which case there is only one in the train dataset, but this tag won't be available for test data according to data description.
Browsing through MRA modality images in the dataset I noticed one image (SeriesUID: 1.2.826.0.1.3680043.8.498.10063454172499468887877935052136698373) that looked different from the others. I am not an expert in the field but it looks like a MIP:
\n
\nIs the presence of such images among MRA images to be expected in the test set as well? It seems like it is possible to identify such images in the DICOM metadata by looking at the ImageType tag (\"Derived\" for the image above), in which case there is only one in the train dataset, but this tag won't be available for test data according to data description.
We aimed to eliminate all MIP and other derived images from the whole dataset, but it seems this one slipped through. We will double check that no such images are in the test set.
", "rawMarkdown": "We aimed to eliminate all MIP and other derived images from the whole dataset, but it seems this one slipped through. We will double check that no such images are in the test set. ", "votes": 3, "replies": [ { "id": 3258536, "postDate": "2025-07-30T20:29:13.110Z", "content": "Does this include subtraction images for CTA? There were several CTAs which looked like there was some kind of bone/non-vessel subtraction applied.
", "rawMarkdown": "Does this include subtraction images for CTA? There were several CTAs which looked like there was some kind of bone/non-vessel subtraction applied. ", "replies": [ { "id": 3258564, "postDate": "2025-07-30T21:32:57.170Z", "content": "One site shared some CTAs that are subtractions, which we plan to leaving in the dataset as is
", "rawMarkdown": "One site shared some CTAs that are subtractions, which we plan to leaving in the dataset as is", "votes": 1 } ] } ] } ], "comments": [ { "id": 3258260, "author_name": "Evan Calabrese", "author_url": "", "post_date": "2025-07-30T12:51:56.397000", "content": "We aimed to eliminate all MIP and other derived images from the whole dataset, but it seems this one slipped through. We will double check that no such images are in the test set.
", "votes": 3, "replies": [ { "id": 3258536, "author_name": "Ian Pan", "author_url": "", "post_date": "2025-07-30T20:29:13.110000", "content": "Does this include subtraction images for CTA? There were several CTAs which looked like there was some kind of bone/non-vessel subtraction applied.
", "votes": 0, "replies": [ { "id": 3258564, "author_name": "JeffRudie", "author_url": "", "post_date": "2025-07-30T21:32:57.170000", "content": "One site shared some CTAs that are subtractions, which we plan to leaving in the dataset as is
", "votes": 1, "replies": [] } ] } ] } ], "raw_markdown_by_id": { "3258259": "Browsing through MRA modality images in the dataset I noticed one image (**SeriesUID**: 1.2.826.0.1.3680043.8.498.10063454172499468887877935052136698373) that looked different from the others. I am not an expert in the field but it looks like a MIP:\n\nIs the presence of such images among MRA images to be expected in the test set as well? It seems like it is possible to identify such images in the DICOM metadata by looking at the **ImageType** tag (\"Derived\" for the image above), in which case there is only one in the train dataset, but this tag won't be available for test data according to data description.", "3258260": "We aimed to eliminate all MIP and other derived images from the whole dataset, but it seems this one slipped through. We will double check that no such images are in the test set. " } }