# Resolution of segmentations change after saving. 

**URL:** <https://discourse.slicer.org/t/resolution-of-segmentations-change-after-saving/31191>\
**Category:** Support\
**Created:** [August 17, 2023, 7:40am UTC](https://discourse.slicer.org/t/resolution-of-segmentations-change-after-saving/31191 "2023-08-17T07:40:54Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![austingrove](https://avatars.discourse-cdn.com/v4/letter/a/41988e/32.png) [@austingrove](https://discourse.slicer.org/u/austingrove)\
**Post date:** [August 17, 2023, 7:40am UTC](https://discourse.slicer.org/t/resolution-of-segmentations-change-after-saving/31191/1 "2023-08-17T07:40:54Z")

</div>

Hello,

After saving segmentations and labels, I have noticed that sometimes certain structures change to a lower resolution. Not all segmentations change to a lower resolution, just some, but it is always the same structures. I am wondering if anyone knows why this is or if anyone has tips or suggestions to prevent this.

Thanks!

---

<div class="post-metadata">

**Author:** ![lassoan](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/lassoan/32/13_2.png) [@lassoan](https://discourse.slicer.org/u/lassoan)\
**Post date:** [August 17, 2023, 7:55am UTC](https://discourse.slicer.org/t/resolution-of-segmentations-change-after-saving/31191/2 "2023-08-17T07:55:38Z")

</div>

In a segmentation with binary labelmap representation, each layer is allowed to have different geometry (origin, spacing, axis directions, extents) temporarily - to allow moving segments between segmentations without unnecessary quality loss (each resampling of a binary labelmap can lead to slight changes). All layers are forced to have the same geometry during certain editing operations and when the segmentation is saved to file.

---

<div class="post-metadata">

**Author:** ![mikebind](https://avatars.discourse-cdn.com/v4/letter/m/71e660/32.png) [@mikebind](https://discourse.slicer.org/u/mikebind)\
**Post date:** [August 17, 2023, 10:11pm UTC](https://discourse.slicer.org/t/resolution-of-segmentations-change-after-saving/31191/3 "2023-08-17T22:11:39Z")

</div>

I didn’t know this. Is it documented under what conditions this happens? Since uniformity is forced on saving, shouldn’t it be forced when a segment is transferred into a different segmentation? What is the use case where allowing the temporary altered resolution, but forcing it on saving, avoids unnecessary resampling? The only case I can think of is if you have a high resolution segment in segmentation A, copy it to lower resolution segmentation B, and then copy it again from segmentation B to segmentation C, which has a different low resolution, but I’m not sure when someone would routinely be doing that.

---

<div class="post-metadata">

**Author:** ![lassoan](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/lassoan/32/13_2.png) [@lassoan](https://discourse.slicer.org/u/lassoan)\
**Post date:** [August 18, 2023, 9:09pm UTC](https://discourse.slicer.org/t/resolution-of-segmentations-change-after-saving/31191/4 "2023-08-18T21:09:40Z")

</div>

> [@mikebind](#):
>
> Is it documented under what conditions this happens?

It is described [here](https://slicer.readthedocs.io/en/latest/user_guide/image_segmentation.html#binary-labelmap-representation), but not in more detail than in my above post. It should not matter, because when you need segments to have the same geometry then the resampling happens automatically.

> [@mikebind](#):
>
> The only case I can think of is if you have a high resolution segment in segmentation A, copy it to lower resolution segmentation B, and then copy it again from segmentation B to segmentation C, which has a different low resolution, but I’m not sure when someone would routinely be doing that.

This is a good example but there are many more. Resampling is performed using “lazy” strategy (i.e., as late as possible), to avoid any _unnecessary_ loss of detail.