# Behind-the-scene Calculation for ICE

**URL:** <https://discourse.slicer.org/t/behind-the-scene-calculation-for-ice/21233>\
**Category:** Support\
**Created:** [December 27, 2021, 11:46am UTC](https://discourse.slicer.org/t/behind-the-scene-calculation-for-ice/21233 "2021-12-27T11:46:57Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Wong\_Yun\_Ming](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/wong_yun_ming/32/13645_2.png) [@Wong\_Yun\_Ming](https://discourse.slicer.org/u/Wong_Yun_Ming)\
**Post date:** [December 27, 2021, 11:46am UTC](https://discourse.slicer.org/t/behind-the-scene-calculation-for-ice/21233/1 "2021-12-27T11:46:57Z")

</div>

Hi, I am trying to compute ICE on my own as I did not use Slicer to perform deformable registration on my images. May I know the way ICE is computed in Slicer, so that I can reproduce the results using my own code?

---

<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:** [December 27, 2021, 11:59am UTC](https://discourse.slicer.org/t/behind-the-scene-calculation-for-ice/21233/2 "2021-12-27T11:59:38Z")

</div>

ICE acronym has several very different meanings in medical imaging. Do you mean inverse consistency error?

Are you already using a the Transform module or a Slicer extension (e.g., RegistrationQA) to compute it and interested n how they work?

What is your overall goal? Comparison of registration algorithms? What is the target organ? What registration software are you comparing?

---

<div class="post-metadata">

**Author:** ![Wong\_Yun\_Ming](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/wong_yun_ming/32/13645_2.png) [@Wong\_Yun\_Ming](https://discourse.slicer.org/u/Wong_Yun_Ming)\
**Post date:** [December 27, 2021, 1:01pm UTC](https://discourse.slicer.org/t/behind-the-scene-calculation-for-ice/21233/3 "2021-12-27T13:01:12Z")

</div>

Hi, yes I was referring to inverse consistency error.

I did the registration on Velocity software and I could not find a way to evaluate the ICE for certain ROIs using the deformation vector field obtained from Velocity.

I am not comparing registration algorithms, instead I would like to check whether ICE correlates with some other metrics for deformation accuracy evaluation (eg. DICE) for prostate cancer patients.