{ "id": 340482, "title": "RSNA-CSFD PNG/JPG Dataset", "url": "/competitions/rsna-2022-cervical-spine-fracture-detection/discussion/340482", "author_name": "Awsaf", "post_date": "2022-07-29T09:33:43.152000", "votes": 58, "comment_count": 12, "views": 0, "content": "

Hi everyone, to help you get started I simply converted .dcm files to .jpg/.png. As these datasets are quite small comparing the actual dataset, you can easily use them to fast up your training.

\n

Jpg

\n\n

Png

\n

Coming soon

", "messages": [ { "id": 1875766, "postDate": "2022-07-29T09:33:43.153Z", "content": "

Hi everyone, to help you get started I simply converted .dcm files to .jpg/.png. As these datasets are quite small comparing the actual dataset, you can easily use them to fast up your training.

\n

Jpg

\n\n

Png

\n

Coming soon

", "rawMarkdown": "Hi everyone, to help you get started I simply converted `.dcm` files to `.jpg`/`.png`. As these datasets are quite small comparing the actual dataset, you can easily use them to fast up your training.\n## Jpg\n* `256x256`: [RSNA-CSFD: 256x256 Jpg Dataset](https://www.kaggle.com/datasets/awsaf49/rsna-csfd-256x256-jpg-dataset) (~18GB).\n* `512x512`: [RSNA-CSFD: 512x512 Jpg Dataset](https://www.kaggle.com/datasets/awsaf49/rsna-csfd-512x512-jpg-Dataset) (~55GB)\n## Png\nComing soon", "votes": 58 }, { "id": 1877235, "postDate": "2022-07-30T12:47:14.263Z", "content": "

@awsaf49 given your experience from the competitions, do you think converting images from 16-bit Dicom to JPEG or PNG with an 8-bit depth will distort the image's quality and characteristics?

", "rawMarkdown": "@awsaf49 given your experience from the competitions, do you think converting images from 16-bit Dicom to JPEG or PNG with an 8-bit depth will distort the image's quality and characteristics?", "votes": 5 }, { "id": 1877090, "postDate": "2022-07-30T10:42:25.643Z", "content": "

I also made a kernel for PNG conversion:

\n
\n

https://www.kaggle.com/code/jirkaborovec/spine-fracture-load-convert-dicom-to-png

\n
", "rawMarkdown": "I also made a kernel for PNG conversion:\n>https://www.kaggle.com/code/jirkaborovec/spine-fracture-load-convert-dicom-to-png", "votes": 6, "replies": [ { "id": 1881091, "postDate": "2022-08-02T08:45:51.207Z", "content": "

I also created a dataset from the kernel output with some denoising, so check also:

\n
\n

https://www.kaggle.com/datasets/jirkaborovec/cervical-spine-fracture-detection-equalized-png

\n
", "rawMarkdown": "I also created a dataset from the kernel output with some denoising, so check also:\n\n>https://www.kaggle.com/datasets/jirkaborovec/cervical-spine-fracture-detection-equalized-png", "votes": 2 } ] }, { "id": 1876108, "postDate": "2022-07-29T15:25:23.877Z", "content": "

Thanks for sharing, it will be quite useful!
\nDid you use interpolation for resizing from 512 x 512 to 256 x 256?

", "rawMarkdown": "Thanks for sharing, it will be quite useful! \nDid you use interpolation for resizing from 512 x 512 to 256 x 256?", "votes": 1, "replies": [ { "id": 1880097, "postDate": "2022-08-01T13:15:33.620Z", "content": "

I used inter_area* interpolation.

", "rawMarkdown": "I used *inter_area** interpolation.", "votes": 2 } ] }, { "id": 3229704, "postDate": "2025-06-22T01:47:43.777Z", "content": "

![](url to embed)info_popup.9.png

", "rawMarkdown": "[![](url to embed)](url)info_popup.9.png" }, { "id": 1988772, "postDate": "2022-10-15T14:33:52.733Z", "content": "

Anyone compares training CV loss between the original dataset and this Jpg dataset?

", "rawMarkdown": "Anyone compares training CV loss between the original dataset and this Jpg dataset? " }, { "id": 1964131, "postDate": "2022-09-30T15:33:53.507Z", "content": "

Since jpg is a lossy storage format, it seems possible that small/hairline fractures will be removed from the image. Especially since these small features will show up in the high-frequency spectrum of the images DCT, which could be removed by the jpg conversion. I'd be interested to see the results of training with raw and jpg formats

", "rawMarkdown": "Since jpg is a lossy storage format, it seems possible that small/hairline fractures will be removed from the image. Especially since these small features will show up in the high-frequency spectrum of the images DCT, which could be removed by the jpg conversion. I'd be interested to see the results of training with raw and jpg formats" }, { "id": 1956074, "postDate": "2022-09-26T09:10:33.343Z", "content": "

May I ask how did you overcome disk space issue? I tried to create my own dataset but ran into full disk issue, even I save images into /tmp path

", "rawMarkdown": "May I ask how did you overcome disk space issue? I tried to create my own dataset but ran into full disk issue, even I save images into /tmp path" }, { "id": 1875844, "postDate": "2022-07-29T10:45:03.650Z", "content": "

Thank for sharing!!!!

", "rawMarkdown": "Thank for sharing!!!! ", "votes": 1 }, { "id": 1875983, "postDate": "2022-07-29T13:40:56.353Z", "content": "

Thanks for sharing ..

", "rawMarkdown": "Thanks for sharing .." }, { "id": 1875931, "postDate": "2022-07-29T12:51:40.550Z", "content": "

Thanks, this will help me!

", "rawMarkdown": "Thanks, this will help me!" } ], "comments": [ { "id": 1877235, "author_name": "Mohammad Dehghanmanshadi", "author_url": "", "post_date": "2022-07-30T12:47:14.263000", "content": "

@awsaf49 given your experience from the competitions, do you think converting images from 16-bit Dicom to JPEG or PNG with an 8-bit depth will distort the image's quality and characteristics?

", "votes": 5, "replies": [] }, { "id": 1877090, "author_name": "Jirka", "author_url": "", "post_date": "2022-07-30T10:42:25.643000", "content": "

I also made a kernel for PNG conversion:

\n
\n

https://www.kaggle.com/code/jirkaborovec/spine-fracture-load-convert-dicom-to-png

\n
", "votes": 6, "replies": [ { "id": 1881091, "author_name": "Jirka", "author_url": "", "post_date": "2022-08-02T08:45:51.207000", "content": "

I also created a dataset from the kernel output with some denoising, so check also:

\n
\n

https://www.kaggle.com/datasets/jirkaborovec/cervical-spine-fracture-detection-equalized-png

\n
", "votes": 2, "replies": [] } ] }, { "id": 1876108, "author_name": "Augusto Sarquis Serpa", "author_url": "", "post_date": "2022-07-29T15:25:23.877000", "content": "

Thanks for sharing, it will be quite useful!
\nDid you use interpolation for resizing from 512 x 512 to 256 x 256?

", "votes": 1, "replies": [ { "id": 1880097, "author_name": "Awsaf", "author_url": "", "post_date": "2022-08-01T13:15:33.620000", "content": "

I used inter_area* interpolation.

", "votes": 2, "replies": [] } ] }, { "id": 3229704, "author_name": "tamamlife support", "author_url": "", "post_date": "2025-06-22T01:47:43.777000", "content": "

![](url to embed)info_popup.9.png

", "votes": 0, "replies": [] }, { "id": 1988772, "author_name": "Pierre Tisseur", "author_url": "", "post_date": "2022-10-15T14:33:52.733000", "content": "

Anyone compares training CV loss between the original dataset and this Jpg dataset?

", "votes": 0, "replies": [] }, { "id": 1964131, "author_name": "Cole Harlow", "author_url": "", "post_date": "2022-09-30T15:33:53.507000", "content": "

Since jpg is a lossy storage format, it seems possible that small/hairline fractures will be removed from the image. Especially since these small features will show up in the high-frequency spectrum of the images DCT, which could be removed by the jpg conversion. I'd be interested to see the results of training with raw and jpg formats

", "votes": 0, "replies": [] }, { "id": 1956074, "author_name": "Phat Tran", "author_url": "", "post_date": "2022-09-26T09:10:33.343000", "content": "

May I ask how did you overcome disk space issue? I tried to create my own dataset but ran into full disk issue, even I save images into /tmp path

", "votes": 0, "replies": [] }, { "id": 1875844, "author_name": "IMvision12", "author_url": "", "post_date": "2022-07-29T10:45:03.650000", "content": "

Thank for sharing!!!!

", "votes": 1, "replies": [] }, { "id": 1875983, "author_name": "Kishore M", "author_url": "", "post_date": "2022-07-29T13:40:56.353000", "content": "

Thanks for sharing ..

", "votes": 0, "replies": [] }, { "id": 1875931, "author_name": "andrew_tep", "author_url": "", "post_date": "2022-07-29T12:51:40.550000", "content": "

Thanks, this will help me!

", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "1875766": "Hi everyone, to help you get started I simply converted `.dcm` files to `.jpg`/`.png`. As these datasets are quite small comparing the actual dataset, you can easily use them to fast up your training.\n## Jpg\n* `256x256`: [RSNA-CSFD: 256x256 Jpg Dataset](https://www.kaggle.com/datasets/awsaf49/rsna-csfd-256x256-jpg-dataset) (~18GB).\n* `512x512`: [RSNA-CSFD: 512x512 Jpg Dataset](https://www.kaggle.com/datasets/awsaf49/rsna-csfd-512x512-jpg-Dataset) (~55GB)\n## Png\nComing soon", "1877235": "@awsaf49 given your experience from the competitions, do you think converting images from 16-bit Dicom to JPEG or PNG with an 8-bit depth will distort the image's quality and characteristics?", "1877090": "I also made a kernel for PNG conversion:\n>https://www.kaggle.com/code/jirkaborovec/spine-fracture-load-convert-dicom-to-png", "1876108": "Thanks for sharing, it will be quite useful! \nDid you use interpolation for resizing from 512 x 512 to 256 x 256?", "3229704": "[![](url to embed)](url)info_popup.9.png", "1988772": "Anyone compares training CV loss between the original dataset and this Jpg dataset? ", "1964131": "Since jpg is a lossy storage format, it seems possible that small/hairline fractures will be removed from the image. Especially since these small features will show up in the high-frequency spectrum of the images DCT, which could be removed by the jpg conversion. I'd be interested to see the results of training with raw and jpg formats", "1956074": "May I ask how did you overcome disk space issue? I tried to create my own dataset but ran into full disk issue, even I save images into /tmp path", "1875844": "Thank for sharing!!!! ", "1875983": "Thanks for sharing ..", "1875931": "Thanks, this will help me!" } }