{ "id": 603197, "title": "Code + dataset (all train series metadata) to help starters", "url": "/competitions/rsna-intracranial-aneurysm-detection/discussion/603197", "author_name": "Suman Kumar Gangopadhyay", "post_date": "2025-09-01T04:33:36.471000", "votes": 0, "comment_count": 3, "views": 0, "content": "

Hello everyone,
\nI have collected the DICOM metadata for all the training series images along with the other training metadata in this dataset.
\nIt is a single parquet file generated using polars and duckdb.
\nThe code for generating this data is in this notebook
\nHope this helps those who are joining now or facing initial bottlenecks.
\nFeedbacks are welcome.

", "messages": [ { "id": 3280323, "postDate": "2025-09-02T13:07:57.487Z", "content": "

Thanks for sharing this resource — it’s really helpful for people just starting out with the dataset. 🙌

\n

By the way, if anyone here is interested in exploring more medical imaging challenges, I recently launched a Kaggle competition: Grand X-ray Slam. It focuses on chest X-ray classification with a curated multi-institution dataset. Would be great to see more people join in, test approaches, and share ideas.

\n

Here’s the link: Grand X-ray Slam Competition

", "rawMarkdown": "Thanks for sharing this resource — it’s really helpful for people just starting out with the dataset. 🙌\n\nBy the way, if anyone here is interested in exploring more medical imaging challenges, I recently launched a Kaggle competition: Grand X-ray Slam. It focuses on chest X-ray classification with a curated multi-institution dataset. Would be great to see more people join in, test approaches, and share ideas.\n\nHere’s the link: [Grand X-ray Slam Competition](https://www.kaggle.com/competitions/grand-xray-slam-division-a)" }, { "id": 3280274, "postDate": "2025-09-02T12:27:14.860Z", "content": "

Thanks so much for sharing your work!

", "rawMarkdown": "Thanks so much for sharing your work!", "replies": [ { "id": 3281380, "postDate": "2025-09-04T07:34:50.097Z", "content": "

thank you @evancalabrese

", "rawMarkdown": "thank you @evancalabrese " } ] }, { "id": 3279569, "postDate": "2025-09-01T04:33:36.470Z", "content": "

Hello everyone,
\nI have collected the DICOM metadata for all the training series images along with the other training metadata in this dataset.
\nIt is a single parquet file generated using polars and duckdb.
\nThe code for generating this data is in this notebook
\nHope this helps those who are joining now or facing initial bottlenecks.
\nFeedbacks are welcome.

", "rawMarkdown": "Hello everyone, \nI have collected the DICOM metadata for all the training series images along with the other training metadata in this [dataset](https://www.kaggle.com/datasets/sumaniitm/rsna-aneurysm-training-metadata-polars-duckdb/data).\nIt is a single parquet file generated using `polars` and `duckdb`.\nThe code for generating this data is in this [notebook](https://www.kaggle.com/code/sumaniitm/suman-notebook-1)\nHope this helps those who are joining now or facing initial bottlenecks.\nFeedbacks are welcome." } ], "comments": [ { "id": 3280323, "author_name": "Guntas Dhanjal", "author_url": "", "post_date": "2025-09-02T13:07:57.487000", "content": "

Thanks for sharing this resource — it’s really helpful for people just starting out with the dataset. 🙌

\n

By the way, if anyone here is interested in exploring more medical imaging challenges, I recently launched a Kaggle competition: Grand X-ray Slam. It focuses on chest X-ray classification with a curated multi-institution dataset. Would be great to see more people join in, test approaches, and share ideas.

\n

Here’s the link: Grand X-ray Slam Competition

", "votes": 0, "replies": [] }, { "id": 3280274, "author_name": "Evan Calabrese", "author_url": "", "post_date": "2025-09-02T12:27:14.860000", "content": "

Thanks so much for sharing your work!

", "votes": 0, "replies": [ { "id": 3281380, "author_name": "Suman Kumar Gangopadhyay", "author_url": "", "post_date": "2025-09-04T07:34:50.097000", "content": "

thank you @evancalabrese

", "votes": 0, "replies": [] } ] } ], "raw_markdown_by_id": { "3280323": "Thanks for sharing this resource — it’s really helpful for people just starting out with the dataset. 🙌\n\nBy the way, if anyone here is interested in exploring more medical imaging challenges, I recently launched a Kaggle competition: Grand X-ray Slam. It focuses on chest X-ray classification with a curated multi-institution dataset. Would be great to see more people join in, test approaches, and share ideas.\n\nHere’s the link: [Grand X-ray Slam Competition](https://www.kaggle.com/competitions/grand-xray-slam-division-a)", "3280274": "Thanks so much for sharing your work!", "3279569": "Hello everyone, \nI have collected the DICOM metadata for all the training series images along with the other training metadata in this [dataset](https://www.kaggle.com/datasets/sumaniitm/rsna-aneurysm-training-metadata-polars-duckdb/data).\nIt is a single parquet file generated using `polars` and `duckdb`.\nThe code for generating this data is in this [notebook](https://www.kaggle.com/code/sumaniitm/suman-notebook-1)\nHope this helps those who are joining now or facing initial bottlenecks.\nFeedbacks are welcome." } }