Skip to content

Lazy resampling entry for whats_new_1_2 #6590

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Nic-Ma merged 2 commits into from
Jun 7, 2023
Merged

Lazy resampling entry for whats_new_1_2 #6590

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
369aed7
Lazy resampling entry for whats_new_1_2
atbenmurray Jun 7, 2023
ccc7a2a
Update docs/source/whatsnew_1_2.md
atbenmurray Jun 7, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 18 additions & 0 deletions docs/source/whatsnew_1_2.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@
- MetricsReloaded integration
- Bundle workflow APIs
- Modular patch inference
- Lazy resampling for preprocessing

## Auto3DSeg enhancements and benchmarks
Auto3DSeg is an innovative solution for 3D medical image segmentation, leveraging the advancements in MONAI and GPUs for algorithm development and deployment.
Expand Down Expand Up @@ -57,3 +58,20 @@ to construct the result image corresponding to the input image. Although dependi
the exact implementations of a patch inference may vary, the overall process of splitting, running inference, and merging the results remains the same.
In this release, we have created a modular design for patch inference, which defines the overall process while abstracting away the specific
behavior of how to split the image into patches, how to pre and post process each patch, and how to merge the output patches.

## Lazy Resampling for preprocessing
Lazy Resampling is a new, experimental feature for preprocessing. It works under
the hood along with MONAI transforms to combine adjacent spatial and
cropping transforms into a single operation. This allows MONAI to reduce the number of data resamples
a pipeline undergoes. Depending on the preprocessing pipeline, it can potentially:

* reduce processing time
* reduce processing memory
* reduce incidental artifacts added by resampling
* preserve data that would otherwise be cropped and replaced with padding

Lazy Resampling pipelines can use a mixture of MONAI and non-MONAI transforms, so
should work with almost all existing pipelines simply by setting `lazy=True`
on MONAI `Compose` instances. See the
[Lazy Resampling topic](https://docs.monai.io/en/stable/lazy_resampling.html)
in the documentation for more details.