# Changed sensors types after SSS

**URL:** <https://mne.discourse.group/t/changed-sensors-types-after-sss/3603>\
**Category:** Support & Discussions\
**Tags:** meg\
**Created:** [August 27, 2021, 9:40pm UTC](https://mne.discourse.group/t/changed-sensors-types-after-sss/3603 "2021-08-27T21:40:35Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![eort](https://avatars.discourse-cdn.com/v4/letter/e/e56c9b/32.png) [@eort](https://mne.discourse.group/u/eort)\
**Post date:** [August 27, 2021, 9:40pm UTC](https://mne.discourse.group/t/changed-sensors-types-after-sss/3603/1 "2021-08-27T21:40:35Z")

</div>

Hi all,

I was planning to run a SSS on my MEG data. But before doing so, I’d like to understand the consequences for the subsequent (pre)-processing steps. In, particular I wonder about following passage in [Jas et al. (2018):](https://www.frontiersin.org/articles/10.3389/fnins.2018.00530/full)

> It is important to highlight that after SSS, the magnetometer and gradiometer data are projected from a common lower dimensional SSS coordinate system that typically spans between 64 and 80 dimensions. As a result, both sensor types contain highly similar information, which also modifies the inter-channel correlation structure. This is the reason why MNE will treat them as a single sensor type in many of the analyses that follow.

I was wondering how are magnetometers and gradiometers treated as the same channel type? Are they a new “generic” type, or are now both called gradiometers, or magnetometers, respectively? From looking at the `Raw.info['chs']` it seems like magnetometers have become a new type, but gradiometers haven’t.  
Or do I need to actively take this into account at all when doing ICA, going into source space, etc, or will MNE handle things the correct way under the hood anyway ?

Thanks,

Eduard

---

<div class="post-metadata">

**Author:** ![larsoner](https://yyz2.discourse-cdn.com/free1/user_avatar/mne.discourse.group/larsoner/32/3_2.png) [@larsoner](https://mne.discourse.group/u/larsoner)\
**Post date:** [August 31, 2021, 1:18pm UTC](https://mne.discourse.group/t/changed-sensors-types-after-sss/3603/2 "2021-08-31T13:18:09Z")

</div>

> [@eort](#):
>
> I was wondering how are magnetometers and gradiometers treated as the same channel type? Are they a new “generic” type, or are now both called gradiometers, or magnetometers, respectively?

They are still coded and called magnetometers and gradiometers. It’s just that they are no longer independent, and MNE will recognize that Maxwell filtering / SSS has been done and (should) treat them properly during SSP, ICA, covariance estimation, rank estimation, inverse imaging, etc.

---

<div class="post-metadata">

**Author:** ![eort](https://avatars.discourse-cdn.com/v4/letter/e/e56c9b/32.png) [@eort](https://mne.discourse.group/u/eort)\
**Post date:** [September 1, 2021, 4:20pm UTC](https://mne.discourse.group/t/changed-sensors-types-after-sss/3603/3 "2021-09-01T16:20:45Z")

</div>

Ah, great! So as long as I use MNE functions to run analyses, I don’t need to take special care of SSS’ed data?

Thank you!

---

<div class="post-metadata">

**Author:** ![richard](https://yyz2.discourse-cdn.com/free1/user_avatar/mne.discourse.group/richard/32/15_2.png) [@richard](https://mne.discourse.group/u/richard)\
**Post date:** [September 1, 2021, 5:12pm UTC](https://mne.discourse.group/t/changed-sensors-types-after-sss/3603/4 "2021-09-01T17:12:31Z")

</div>

Yes, this is correct, MNE should always do the right thing automatically.