{ "id": 438690, "title": "Data DICOM to 3 different types of Image (Grayscale) ", "url": "/competitions/rsna-2023-abdominal-trauma-detection/discussion/438690", "author_name": "Minh-Long Pham", "post_date": "2023-09-12T09:28:33.927000", "votes": 17, "comment_count": 6, "views": 0, "content": "

Hello everyone, I've converted the DICOM files to 3 different types of Image (Grayscale)

\n\n

Here is an example of 3 different types:

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
BasedScaledWindowed
\"\"\"\"\"\"
\n

Due to the Kaggle limit, I have divided the dataset into sub-parts (the whole dataset is divisible by 7). Each part is a version of the data. So, please look at the version notes to choose the right part.
\nLet me know if there are any issues, and I'll fix them as soon as possible.
\nPlease upvote if you find it useful! Thank you all!

", "messages": [ { "id": 2434399, "postDate": "2023-09-12T09:28:33.927Z", "content": "

Hello everyone, I've converted the DICOM files to 3 different types of Image (Grayscale)

\n\n

Here is an example of 3 different types:

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
BasedScaledWindowed
\"\"\"\"\"\"
\n

Due to the Kaggle limit, I have divided the dataset into sub-parts (the whole dataset is divisible by 7). Each part is a version of the data. So, please look at the version notes to choose the right part.
\nLet me know if there are any issues, and I'll fix them as soon as possible.
\nPlease upvote if you find it useful! Thank you all!

", "rawMarkdown": "Hello everyone, I've converted the DICOM files to 3 different types of Image (Grayscale) \n\n- Based convert: [RSNA_2023_ATD_base](https://www.kaggle.com/datasets/hundred3421/rsna-2023-atd-base)\n- Scaled (High contrast): [RSNA_2023_ATD_scale](https://www.kaggle.com/datasets/hundred3421/rsna-2023-atd-scale)\n- Windowed base on [abdomen soft tissues](https://radiopaedia.org/articles/windowing-ct): [RSNA_2023_ATD_Window](https://www.kaggle.com/datasets/hundred3421/rsna-2023-atd-window)\n\nHere is an example of 3 different types:\n| Based | Scaled | Windowed |\n| --- | --- | --- |\n| ![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F5888135%2Ffaff803df8a3ee1cdf8219bd3784c468%2F1000_base.png?generation=1695001326442443&alt=media) | ![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F5888135%2Fa15b016033c415b1eee917da0c928e2c%2F1000_scale.png?generation=1695001368715283&alt=media) | ![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F5888135%2F815a5e42dbc90d0eb7e0865a39ebc12e%2F1000_window.png?generation=1695001386916434&alt=media) | \n\n\nDue to the Kaggle limit, I have divided the dataset into sub-parts (the whole dataset is divisible by 7). Each part is a version of the data. So, please look at the version notes to choose the right part.\nLet me know if there are any issues, and I'll fix them as soon as possible. \nPlease upvote if you find it useful! Thank you all!\n", "votes": 17 }, { "id": 2444217, "postDate": "2023-09-18T08:02:10.473Z", "content": "

Hey,
\nI think some further description would be useful.
\n-What does scaled mean?
\n-Maybe you could share the code for creating the dataset
\nAlso I can only see several versions in the 'based'. In the other datasets I can only see one version

", "rawMarkdown": "Hey,\nI think some further description would be useful.\n-What does scaled mean?\n-Maybe you could share the code for creating the dataset\nAlso I can only see several versions in the 'based'. In the other datasets I can only see one version", "votes": 1, "replies": [ { "id": 2445637, "postDate": "2023-09-19T02:14:09.657Z", "content": "

Hi, I have updated versions for other datasets. Please have a look if it is okay

\n", "rawMarkdown": "Hi, I have updated versions for other datasets. Please have a look if it is okay\n- `What does scaled mean?` I have made it 'high contrast' by splitting the **image pixel values** into groups and then scaling it to values between 0 and 1.\n- Here is the notebook [RSNA_2023_ATD:data test](https://www.kaggle.com/code/hundred3421/rsna-2023-atd-data-test/notebook) for convenience.", "votes": 3, "replies": [ { "id": 2446228, "postDate": "2023-09-19T10:28:19.413Z", "content": "

Hey, thank you very much!
\nWhat do you mean by splitting the data into groups? Didn't you do this image by image? (both the scaling and the windowing)

", "rawMarkdown": "Hey, thank you very much!\nWhat do you mean by splitting the data into groups? Didn't you do this image by image? (both the scaling and the windowing)" }, { "id": 2447293, "postDate": "2023-09-20T01:50:28.393Z", "content": "

Hi, I mean the pixel values of each image (in both scaling and windowing). I have edited it. Thank you

", "rawMarkdown": "Hi, I mean the pixel values of each image (in both scaling and windowing). I have edited it. Thank you", "votes": 1 }, { "id": 2448031, "postDate": "2023-09-20T11:40:17.617Z", "content": "

Thanks for the clarification. Can you tell me why you have split it into groups? Could have you just scaled the all the pixel values in the image from 0 to 1?

", "rawMarkdown": "Thanks for the clarification. Can you tell me why you have split it into groups? Could have you just scaled the all the pixel values in the image from 0 to 1?" } ] } ] }, { "id": 2739027, "postDate": "2024-04-06T19:18:07.017Z", "content": "

thanks for you effort but I found scan 51093 almost completely black
\n

", "rawMarkdown": "thanks for you effort but I found scan 51093 almost completely black\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F8229747%2Feabaabf79c507161a296f85cc3ecb6d3%2FScreenshot%202024-04-06%20211530.png?generation=1712431083296569&alt=media) " } ], "comments": [ { "id": 2444217, "author_name": "Gyula Maloveczky4", "author_url": "", "post_date": "2023-09-18T08:02:10.473000", "content": "

Hey,
\nI think some further description would be useful.
\n-What does scaled mean?
\n-Maybe you could share the code for creating the dataset
\nAlso I can only see several versions in the 'based'. In the other datasets I can only see one version

", "votes": 1, "replies": [ { "id": 2445637, "author_name": "Minh-Long Pham", "author_url": "", "post_date": "2023-09-19T02:14:09.657000", "content": "

Hi, I have updated versions for other datasets. Please have a look if it is okay

\n", "votes": 3, "replies": [ { "id": 2446228, "author_name": "Gyula Maloveczky4", "author_url": "", "post_date": "2023-09-19T10:28:19.413000", "content": "

Hey, thank you very much!
\nWhat do you mean by splitting the data into groups? Didn't you do this image by image? (both the scaling and the windowing)

", "votes": 0, "replies": [] }, { "id": 2447293, "author_name": "Minh-Long Pham", "author_url": "", "post_date": "2023-09-20T01:50:28.393000", "content": "

Hi, I mean the pixel values of each image (in both scaling and windowing). I have edited it. Thank you

", "votes": 1, "replies": [] }, { "id": 2448031, "author_name": "Gyula Maloveczky4", "author_url": "", "post_date": "2023-09-20T11:40:17.617000", "content": "

Thanks for the clarification. Can you tell me why you have split it into groups? Could have you just scaled the all the pixel values in the image from 0 to 1?

", "votes": 0, "replies": [] } ] } ] }, { "id": 2739027, "author_name": "Ahmed Kamal El-Senussi", "author_url": "", "post_date": "2024-04-06T19:18:07.017000", "content": "

thanks for you effort but I found scan 51093 almost completely black
\n

", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "2434399": "Hello everyone, I've converted the DICOM files to 3 different types of Image (Grayscale) \n\n- Based convert: [RSNA_2023_ATD_base](https://www.kaggle.com/datasets/hundred3421/rsna-2023-atd-base)\n- Scaled (High contrast): [RSNA_2023_ATD_scale](https://www.kaggle.com/datasets/hundred3421/rsna-2023-atd-scale)\n- Windowed base on [abdomen soft tissues](https://radiopaedia.org/articles/windowing-ct): [RSNA_2023_ATD_Window](https://www.kaggle.com/datasets/hundred3421/rsna-2023-atd-window)\n\nHere is an example of 3 different types:\n| Based | Scaled | Windowed |\n| --- | --- | --- |\n| ![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F5888135%2Ffaff803df8a3ee1cdf8219bd3784c468%2F1000_base.png?generation=1695001326442443&alt=media) | ![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F5888135%2Fa15b016033c415b1eee917da0c928e2c%2F1000_scale.png?generation=1695001368715283&alt=media) | ![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F5888135%2F815a5e42dbc90d0eb7e0865a39ebc12e%2F1000_window.png?generation=1695001386916434&alt=media) | \n\n\nDue to the Kaggle limit, I have divided the dataset into sub-parts (the whole dataset is divisible by 7). Each part is a version of the data. So, please look at the version notes to choose the right part.\nLet me know if there are any issues, and I'll fix them as soon as possible. \nPlease upvote if you find it useful! Thank you all!\n", "2444217": "Hey,\nI think some further description would be useful.\n-What does scaled mean?\n-Maybe you could share the code for creating the dataset\nAlso I can only see several versions in the 'based'. In the other datasets I can only see one version", "2739027": "thanks for you effort but I found scan 51093 almost completely black\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F8229747%2Feabaabf79c507161a296f85cc3ecb6d3%2FScreenshot%202024-04-06%20211530.png?generation=1712431083296569&alt=media) " } }