refactor learn2reg_nlst_paired_lung_ct

[KumoLiu]

refactor learn2reg_nlst_paired_lung_ct using ApplyTransformToPointsd

Checks

• Avoid including large-size files in the PR.
• Clean up long text outputs from code cells in the notebook.
• For security purposes, please check the contents and remove any sensitive info such as user names and private key.
• Ensure (1) hyperlinks and markdown anchors are working (2) use relative paths for tutorial repo files (3) put figure and graphs in the ./figure folder
• Notebook runs automatically ./runner.sh -t <path to .ipynb file>

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[review-notebook-app]

View / edit / reply to this conversation on ReviewNB

ericspod commented on 2024-09-03T12:37:54Z
----------------------------------------------------------------

Line #22.                if self.ensure_channel_first:

This should check that there isn't already a channel dimension. I think the EnsureChannelFirstd transform should work with point data already anyway.

KumoLiu commented on 2024-09-03T12:51:07Z
----------------------------------------------------------------

Yes, use EnsureChannelFirstd instead. Thanks.

[review-notebook-app]

View / edit / reply to this conversation on ReviewNB

ericspod commented on 2024-09-03T12:37:55Z
----------------------------------------------------------------

Line #18.                refer_keys=["fixed_image", "moving_image"],

Why are we setting the affine for the points when they are being loaded when ApplyTransformToPointsd is going to be used later on? This affine values goes into the metadata of the points but isn't actually applied, and then ApplyTransformToPointsd doesn't use this anyway.

KumoLiu commented on 2024-09-03T12:43:38Z
----------------------------------------------------------------

The reason why add the affine for the points when loading is that the points in this case is not at the world space. So the affine here will be used in ApplyTransformToPointsd as a "already applied affine".

[ericspod]

I had some comments on why some things were done but it looks correct regardless of that.

[KumoLiu]

The reason why add the affine for the points when loading is that the points in this case is not at the world space. So the affine here will be used in ApplyTransformToPointsd as a "already applied affine".

---

View entire conversation on ReviewNB

[KumoLiu]

Yes, use EnsureChannelFirstd instead. Thanks.

---

View entire conversation on ReviewNB

[vikashg]

Hey @KumoLiu
I will be able to go through this tonight.
Give me couple of hours

Thanks
Vikash