While setting up my preprocessing for the multiframe DICOMs (all with MR modality), I found out that: 1- all of them have ImageOrientationPatient = [1, 0, 0, 0, 1, 0], indicating that they should all be axial plane series; 2- 25 of these series are in fact either coronal or sagittal plane series. Also, using ImagePositionPatient to determine the plane is not possible, as this label in all the multiframe DICOMs is shown as [0, 0, x], again indicating the series should viewed as an axial plane. Therefore, there seems to be currently no straightforward way to determine the correct orientation of these series.
\nHere is the list of the erroneous series and their correct orientation:
\n1.2.826.0.1.3680043.8.498.10862138275035843887055171875480735964 (Coronal)
\n1.2.826.0.1.3680043.8.498.11396958000946738156009956455739305762 (Coronal)
\n1.2.826.0.1.3680043.8.498.11887329867812275491160566603814454129 (Sagittal)
\n1.2.826.0.1.3680043.8.498.11915319973409844345177713085783065237 (Coronal)
\n1.2.826.0.1.3680043.8.498.12163038646729971461006564302880090481 (Coronal)
\n1.2.826.0.1.3680043.8.498.13001629435974764211403087597568806527 (Coronal)
\n1.2.826.0.1.3680043.8.498.13334658148703615392388818414999249292 (Coronal)
\n1.2.826.0.1.3680043.8.498.18831402822041226140887003611379903167 (Coronal)
\n1.2.826.0.1.3680043.8.498.21275250875812455389777450891502640750 (Sagittal)
\n1.2.826.0.1.3680043.8.498.27235396640484934153639773593945542938 (Sagittal)
\n1.2.826.0.1.3680043.8.498.29351212950805314631667854934458469754 (Coronal)
\n1.2.826.0.1.3680043.8.498.35123157147325830213906326339070528034 (Coronal)
\n1.2.826.0.1.3680043.8.498.35440393683691371542782507480292365786 (Sagittal)
\n1.2.826.0.1.3680043.8.498.36861937197087749960171145883205456895 (Coronal)
\n1.2.826.0.1.3680043.8.498.40006562159206402632477316663171307697 (Coronal)
\n1.2.826.0.1.3680043.8.498.46538678358294253983538640149161986964 (Coronal)
\n1.2.826.0.1.3680043.8.498.49672398100697832208944634471809461961 (Coronal)
\n1.2.826.0.1.3680043.8.498.50668879928342593291812487079769153076 (Coronal)
\n1.2.826.0.1.3680043.8.498.50916621085656781540278427064467759139 (Coronal)
\n1.2.826.0.1.3680043.8.498.56222999331067503423242588210365055932 (Sagittal)
\n1.2.826.0.1.3680043.8.498.67129993505475797984506180089478722899 (Coronal)
\n1.2.826.0.1.3680043.8.498.85431182782929944864196573042506906105 (Coronal)
\n1.2.826.0.1.3680043.8.498.91280907751913581577764343702856084945 (Coronal)
\n1.2.826.0.1.3680043.8.498.93156694293030030637766074579373694728 (Coronal)
\n1.2.826.0.1.3680043.8.498.99804081131933373817667779922320327920 (Coronal)
Organizers, could you please make the necessary corrections at your earliest convenience.
", "messages": [ { "id": 3268460, "postDate": "2025-08-13T01:51:12.863Z", "content": "While setting up my preprocessing for the multiframe DICOMs (all with MR modality), I found out that: 1- all of them have ImageOrientationPatient = [1, 0, 0, 0, 1, 0], indicating that they should all be axial plane series; 2- 25 of these series are in fact either coronal or sagittal plane series. Also, using ImagePositionPatient to determine the plane is not possible, as this label in all the multiframe DICOMs is shown as [0, 0, x], again indicating the series should viewed as an axial plane. Therefore, there seems to be currently no straightforward way to determine the correct orientation of these series.
\nHere is the list of the erroneous series and their correct orientation:
\n1.2.826.0.1.3680043.8.498.10862138275035843887055171875480735964 (Coronal)
\n1.2.826.0.1.3680043.8.498.11396958000946738156009956455739305762 (Coronal)
\n1.2.826.0.1.3680043.8.498.11887329867812275491160566603814454129 (Sagittal)
\n1.2.826.0.1.3680043.8.498.11915319973409844345177713085783065237 (Coronal)
\n1.2.826.0.1.3680043.8.498.12163038646729971461006564302880090481 (Coronal)
\n1.2.826.0.1.3680043.8.498.13001629435974764211403087597568806527 (Coronal)
\n1.2.826.0.1.3680043.8.498.13334658148703615392388818414999249292 (Coronal)
\n1.2.826.0.1.3680043.8.498.18831402822041226140887003611379903167 (Coronal)
\n1.2.826.0.1.3680043.8.498.21275250875812455389777450891502640750 (Sagittal)
\n1.2.826.0.1.3680043.8.498.27235396640484934153639773593945542938 (Sagittal)
\n1.2.826.0.1.3680043.8.498.29351212950805314631667854934458469754 (Coronal)
\n1.2.826.0.1.3680043.8.498.35123157147325830213906326339070528034 (Coronal)
\n1.2.826.0.1.3680043.8.498.35440393683691371542782507480292365786 (Sagittal)
\n1.2.826.0.1.3680043.8.498.36861937197087749960171145883205456895 (Coronal)
\n1.2.826.0.1.3680043.8.498.40006562159206402632477316663171307697 (Coronal)
\n1.2.826.0.1.3680043.8.498.46538678358294253983538640149161986964 (Coronal)
\n1.2.826.0.1.3680043.8.498.49672398100697832208944634471809461961 (Coronal)
\n1.2.826.0.1.3680043.8.498.50668879928342593291812487079769153076 (Coronal)
\n1.2.826.0.1.3680043.8.498.50916621085656781540278427064467759139 (Coronal)
\n1.2.826.0.1.3680043.8.498.56222999331067503423242588210365055932 (Sagittal)
\n1.2.826.0.1.3680043.8.498.67129993505475797984506180089478722899 (Coronal)
\n1.2.826.0.1.3680043.8.498.85431182782929944864196573042506906105 (Coronal)
\n1.2.826.0.1.3680043.8.498.91280907751913581577764343702856084945 (Coronal)
\n1.2.826.0.1.3680043.8.498.93156694293030030637766074579373694728 (Coronal)
\n1.2.826.0.1.3680043.8.498.99804081131933373817667779922320327920 (Coronal)
Organizers, could you please make the necessary corrections at your earliest convenience.
", "rawMarkdown": "While setting up my preprocessing for the multiframe DICOMs (all with MR modality), I found out that: 1- all of them have ImageOrientationPatient = [1, 0, 0, 0, 1, 0], indicating that they should all be axial plane series; 2- 25 of these series are in fact either coronal or sagittal plane series. Also, using ImagePositionPatient to determine the plane is not possible, as this label in all the multiframe DICOMs is shown as [0, 0, x], again indicating the series should viewed as an axial plane. Therefore, there seems to be currently no straightforward way to determine the correct orientation of these series.\n\nHere is the list of the erroneous series and their correct orientation:\n\n1.2.826.0.1.3680043.8.498.10862138275035843887055171875480735964 (Coronal)\n1.2.826.0.1.3680043.8.498.11396958000946738156009956455739305762 (Coronal)\n1.2.826.0.1.3680043.8.498.11887329867812275491160566603814454129 (Sagittal)\n1.2.826.0.1.3680043.8.498.11915319973409844345177713085783065237 (Coronal)\n1.2.826.0.1.3680043.8.498.12163038646729971461006564302880090481 (Coronal)\n1.2.826.0.1.3680043.8.498.13001629435974764211403087597568806527 (Coronal)\n1.2.826.0.1.3680043.8.498.13334658148703615392388818414999249292 (Coronal)\n1.2.826.0.1.3680043.8.498.18831402822041226140887003611379903167 (Coronal)\n1.2.826.0.1.3680043.8.498.21275250875812455389777450891502640750 (Sagittal)\n1.2.826.0.1.3680043.8.498.27235396640484934153639773593945542938 (Sagittal)\n1.2.826.0.1.3680043.8.498.29351212950805314631667854934458469754 (Coronal)\n1.2.826.0.1.3680043.8.498.35123157147325830213906326339070528034 (Coronal)\n1.2.826.0.1.3680043.8.498.35440393683691371542782507480292365786 (Sagittal)\n1.2.826.0.1.3680043.8.498.36861937197087749960171145883205456895 (Coronal)\n1.2.826.0.1.3680043.8.498.40006562159206402632477316663171307697 (Coronal)\n1.2.826.0.1.3680043.8.498.46538678358294253983538640149161986964 (Coronal)\n1.2.826.0.1.3680043.8.498.49672398100697832208944634471809461961 (Coronal)\n1.2.826.0.1.3680043.8.498.50668879928342593291812487079769153076 (Coronal)\n1.2.826.0.1.3680043.8.498.50916621085656781540278427064467759139 (Coronal)\n1.2.826.0.1.3680043.8.498.56222999331067503423242588210365055932 (Sagittal)\n1.2.826.0.1.3680043.8.498.67129993505475797984506180089478722899 (Coronal)\n1.2.826.0.1.3680043.8.498.85431182782929944864196573042506906105 (Coronal)\n1.2.826.0.1.3680043.8.498.91280907751913581577764343702856084945 (Coronal)\n1.2.826.0.1.3680043.8.498.93156694293030030637766074579373694728 (Coronal)\n1.2.826.0.1.3680043.8.498.99804081131933373817667779922320327920 (Coronal)\n\nOrganizers, could you please make the necessary corrections at your earliest convenience.", "votes": 19 }, { "id": 3292425, "postDate": "2025-09-21T20:16:32.433Z", "content": "Would like to hear an update on this issue
", "rawMarkdown": "Would like to hear an update on this issue", "votes": 3 }, { "id": 3268965, "postDate": "2025-08-13T16:59:52.963Z", "content": "Did you check:
\n() SharedFunctionalGroupsSequence\n () PlaneOrientationSequence\n () ImageOrientationPatient\nThat's exactly where I got the data. I got the image orientation in a similar manner via SharedFunctionalGroupsSequence[0].PlaneOrientationSequence[0].ImageOrientationPatient.
", "rawMarkdown": "That's exactly where I got the data. I got the image orientation in a similar manner via SharedFunctionalGroupsSequence[0].PlaneOrientationSequence[0].ImageOrientationPatient.", "replies": [ { "id": 3269429, "postDate": "2025-08-14T13:21:53.763Z", "content": "OK thanks for update. We will look into these and see if we can find a fix.
", "rawMarkdown": "OK thanks for update. We will look into these and see if we can find a fix." }, { "id": 3287270, "postDate": "2025-09-11T08:29:02.687Z", "content": "Hi! is there any update on this issue?
\nI’m concerned that there may be data with this issue in the test set as well.
\n\nOK thanks for update. We will look into these and see if we can find a fix.
\n
Hi @evancalabrese, do you have any feedback or updates on this issue? Or, should we expect this to occur in the test data?
", "rawMarkdown": "> OK thanks for update. We will look into these and see if we can find a fix.\n\nHi @evancalabrese, do you have any feedback or updates on this issue? Or, should we expect this to occur in the test data?\n\n" } ] } ] }, { "id": 3298178, "postDate": "2025-10-04T18:43:30.867Z", "content": "Is it possible to detect and correct such dicom files.
", "rawMarkdown": "Is it possible to detect and correct such dicom files." }, { "id": 3296120, "postDate": "2025-09-30T08:18:52.167Z", "content": "Hi, I referenced this post in a compilation I wrote on this topic -> https://www.kaggle.com/competitions/rsna-intracranial-aneurysm-detection/discussion/609876
", "rawMarkdown": "Hi, I referenced this post in a compilation I wrote on this topic -> https://www.kaggle.com/competitions/rsna-intracranial-aneurysm-detection/discussion/609876" } ], "comments": [ { "id": 3292425, "author_name": "Ravon", "author_url": "", "post_date": "2025-09-21T20:16:32.433000", "content": "Would like to hear an update on this issue
", "votes": 3, "replies": [] }, { "id": 3268965, "author_name": "Evan Calabrese", "author_url": "", "post_date": "2025-08-13T16:59:52.963000", "content": "Did you check:
\n() SharedFunctionalGroupsSequence\n () PlaneOrientationSequence\n () ImageOrientationPatient\nThat's exactly where I got the data. I got the image orientation in a similar manner via SharedFunctionalGroupsSequence[0].PlaneOrientationSequence[0].ImageOrientationPatient.
", "votes": 0, "replies": [ { "id": 3269429, "author_name": "Evan Calabrese", "author_url": "", "post_date": "2025-08-14T13:21:53.763000", "content": "OK thanks for update. We will look into these and see if we can find a fix.
", "votes": 0, "replies": [] }, { "id": 3287270, "author_name": "David Kim", "author_url": "", "post_date": "2025-09-11T08:29:02.687000", "content": "Hi! is there any update on this issue?
\nI’m concerned that there may be data with this issue in the test set as well.
\n\nOK thanks for update. We will look into these and see if we can find a fix.
\n
Hi @evancalabrese, do you have any feedback or updates on this issue? Or, should we expect this to occur in the test data?
", "votes": 0, "replies": [] } ] } ] }, { "id": 3298178, "author_name": "Gowri Shankar Penugonda", "author_url": "", "post_date": "2025-10-04T18:43:30.867000", "content": "Is it possible to detect and correct such dicom files.
", "votes": 0, "replies": [] }, { "id": 3296120, "author_name": "MakeLoveAndPeace", "author_url": "", "post_date": "2025-09-30T08:18:52.167000", "content": "Hi, I referenced this post in a compilation I wrote on this topic -> https://www.kaggle.com/competitions/rsna-intracranial-aneurysm-detection/discussion/609876
", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "3268460": "While setting up my preprocessing for the multiframe DICOMs (all with MR modality), I found out that: 1- all of them have ImageOrientationPatient = [1, 0, 0, 0, 1, 0], indicating that they should all be axial plane series; 2- 25 of these series are in fact either coronal or sagittal plane series. Also, using ImagePositionPatient to determine the plane is not possible, as this label in all the multiframe DICOMs is shown as [0, 0, x], again indicating the series should viewed as an axial plane. Therefore, there seems to be currently no straightforward way to determine the correct orientation of these series.\n\nHere is the list of the erroneous series and their correct orientation:\n\n1.2.826.0.1.3680043.8.498.10862138275035843887055171875480735964 (Coronal)\n1.2.826.0.1.3680043.8.498.11396958000946738156009956455739305762 (Coronal)\n1.2.826.0.1.3680043.8.498.11887329867812275491160566603814454129 (Sagittal)\n1.2.826.0.1.3680043.8.498.11915319973409844345177713085783065237 (Coronal)\n1.2.826.0.1.3680043.8.498.12163038646729971461006564302880090481 (Coronal)\n1.2.826.0.1.3680043.8.498.13001629435974764211403087597568806527 (Coronal)\n1.2.826.0.1.3680043.8.498.13334658148703615392388818414999249292 (Coronal)\n1.2.826.0.1.3680043.8.498.18831402822041226140887003611379903167 (Coronal)\n1.2.826.0.1.3680043.8.498.21275250875812455389777450891502640750 (Sagittal)\n1.2.826.0.1.3680043.8.498.27235396640484934153639773593945542938 (Sagittal)\n1.2.826.0.1.3680043.8.498.29351212950805314631667854934458469754 (Coronal)\n1.2.826.0.1.3680043.8.498.35123157147325830213906326339070528034 (Coronal)\n1.2.826.0.1.3680043.8.498.35440393683691371542782507480292365786 (Sagittal)\n1.2.826.0.1.3680043.8.498.36861937197087749960171145883205456895 (Coronal)\n1.2.826.0.1.3680043.8.498.40006562159206402632477316663171307697 (Coronal)\n1.2.826.0.1.3680043.8.498.46538678358294253983538640149161986964 (Coronal)\n1.2.826.0.1.3680043.8.498.49672398100697832208944634471809461961 (Coronal)\n1.2.826.0.1.3680043.8.498.50668879928342593291812487079769153076 (Coronal)\n1.2.826.0.1.3680043.8.498.50916621085656781540278427064467759139 (Coronal)\n1.2.826.0.1.3680043.8.498.56222999331067503423242588210365055932 (Sagittal)\n1.2.826.0.1.3680043.8.498.67129993505475797984506180089478722899 (Coronal)\n1.2.826.0.1.3680043.8.498.85431182782929944864196573042506906105 (Coronal)\n1.2.826.0.1.3680043.8.498.91280907751913581577764343702856084945 (Coronal)\n1.2.826.0.1.3680043.8.498.93156694293030030637766074579373694728 (Coronal)\n1.2.826.0.1.3680043.8.498.99804081131933373817667779922320327920 (Coronal)\n\nOrganizers, could you please make the necessary corrections at your earliest convenience.", "3292425": "Would like to hear an update on this issue", "3268965": "Did you check:\n```\n(5200,9229) SharedFunctionalGroupsSequence\n (0020,9116) PlaneOrientationSequence\n (0020,0037) ImageOrientationPatient\n```", "3298178": "Is it possible to detect and correct such dicom files.", "3296120": "Hi, I referenced this post in a compilation I wrote on this topic -> https://www.kaggle.com/competitions/rsna-intracranial-aneurysm-detection/discussion/609876" } }