Feature added: Get WSI at mpp #7574

NikolasSchmitz · 2024-03-25T01:25:09Z

Description

In this pull request, the feature to retrieve a whole slide image at a given mpp (microns per pixel) resolution was implemented for every WSIReader class in the function get_wsi_at_mpp.

While the implementations in the OpenslideWSIReader and CuCIMWSIReader classes were tested thoroughly, I could not find a suitable TIFF file for testing with the TiffFileWSIReader class.

For resizing, I have used PIL.Image.resize for Openslide and TiffFile, and cucim.sklearn.transform.resize for CuCIM. Originally, I used cv2.resize, but since the package isn't listed in requirements-dev.txt, I explored alternative solutions."

Types of changes

Non-breaking change (fix or new feature that would not break existing functionality).
Breaking change (fix or new feature that would cause existing functionality to change).
New tests added to cover the changes.
Integration tests passed locally by running ./runtests.sh -f -u --net --coverage.
Quick tests passed locally by running ./runtests.sh --quick --unittests --disttests.
In-line docstrings updated.
Documentation updated, tested make html command in the docs/ folder.

Please let me know what you think and how I can improve this feature.

Best
Niko

(Last PR was closed, because I changed the branch name to include the ticket id)

…si_reader.py

…izing function to Image.resize, cucim.skimage.transform.resize

…roject-MONAI#7308) ### Description Based on the discussion topic [here](Project-MONAI#7161 (comment)), we implemented the Conjugate-Gradient algorithm for linear operator inversion, and Stein's Unbiased Risk Estimator (SURE) [1] loss for ground-truth-date free diffusion process guidance that is proposed in [2] and illustrated in the algorithm below: <img width="650" alt="Screenshot 2023-12-10 at 10 19 25 PM" src="https://github.com/Project-MONAI/MONAI/assets/8581162/97069466-cbaf-44e0-b7a7-ae9deb8fd7f2"> The Conjugate-Gradient (CG) algorithm is used to solve for the inversion of the linear operator in Line-4 in the algorithm above, where the linear operator is too large to store explicitly as a matrix (such as FFT/IFFT of an image) and invert directly. Instead, we can solve for the linear inversion iteratively as in CG. The SURE loss is applied for Line-6 above. This is a differentiable loss function that can be used to train/giude an operator (e.g. neural network), where the pseudo ground truth is available but the reference ground truth is not. For example, in the MRI reconstruction, the pseudo ground truth is the zero-filled reconstruction and the reference ground truth is the fully sampled reconstruction. The reference ground truth is not available due to the lack of fully sampled. **Reference** [1] Stein, C.M.: Estimation of the mean of a multivariate normal distribution. Annals of Statistics 1981 [[paper link](https://projecteuclid.org/journals/annals-of-statistics/volume-9/issue-6/Estimation-of-the-Mean-of-a-Multivariate-Normal-Distribution/10.1214/aos/1176345632.full)] [2] B. Ozturkler et al. SMRD: SURE-based Robust MRI Reconstruction with Diffusion Models. MICCAI 2023 [[paper link](https://arxiv.org/pdf/2310.01799.pdf)] ### Types of changes  - [x] Non-breaking change (fix or new feature that would not break existing functionality). - [ ] Breaking change (fix or new feature that would cause existing functionality to change). - [x] New tests added to cover the changes. - [x] Integration tests passed locally by running `./runtests.sh -f -u --net --coverage`. - [x] Quick tests passed locally by running `./runtests.sh --quick --unittests --disttests`. - [x] In-line docstrings updated. - [x] Documentation updated, tested `make html` command in the `docs/` folder. --------- Signed-off-by: chaoliu <[email protected]> Signed-off-by: cxlcl <[email protected]> Signed-off-by: chaoliu <[email protected]> Signed-off-by: YunLiu <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: YunLiu <[email protected]> Co-authored-by: Eric Kerfoot <[email protected]> Signed-off-by: Nikolas Schmitz <[email protected]>

Signed-off-by: Nikolas Schmitz <[email protected]>

for more information, see https://pre-commit.ci Signed-off-by: Nikolas Schmitz <[email protected]>

Signed-off-by: Nikolas Schmitz <[email protected]>

Signed-off-by: monai-bot <[email protected]> Signed-off-by: monai-bot <[email protected]> Signed-off-by: Nikolas Schmitz <[email protected]>

Signed-off-by: Nikolas Schmitz <[email protected]>

…ject-MONAI#7569) Fixes Project-MONAI#7451 ### Description Reduces the length of error messages and error messages being propagated twice. This helps debug better when long `ConfigComponent`s are being instantiated. Refer to issue Project-MONAI#7451 for more details ### Types of changes  - [x] Non-breaking change (fix or new feature that would not break existing functionality). - [ ] Breaking change (fix or new feature that would cause existing functionality to change). - [ ] New tests added to cover the changes. - [ ] Integration tests passed locally by running `./runtests.sh -f -u --net --coverage`. - [x] Quick tests passed locally by running `./runtests.sh --quick --unittests --disttests`. - [ ] In-line docstrings updated. - [ ] Documentation updated, tested `make html` command in the `docs/` folder. Signed-off-by: Suraj Pai <[email protected]> Co-authored-by: Eric Kerfoot <[email protected]>

Fixes Project-MONAI#2872 ### Description Implementation of mixup, cutmix and cutout as described in the original papers. Current implementation support both, the dictionary-based batches and tuples of tensors. ### Types of changes  - [x] Non-breaking change (fix or new feature that would not break existing functionality). - [ ] Breaking change (fix or new feature that would cause existing functionality to change). - [x] New tests added to cover the changes. - [ ] Integration tests passed locally by running `./runtests.sh -f -u --net --coverage`. - [x] Quick tests passed locally by running `./runtests.sh --quick --unittests --disttests`. - [x] In-line docstrings updated. - [x] Documentation updated, tested `make html` command in the `docs/` folder. --------- Signed-off-by: Juan Pablo de la Cruz Gutiérrez <[email protected]> Signed-off-by: monai-bot <[email protected]> Signed-off-by: elitap <[email protected]> Signed-off-by: Felix Schnabel <[email protected]> Signed-off-by: YanxuanLiu <[email protected]> Signed-off-by: ytl0623 <[email protected]> Signed-off-by: Dženan Zukić <[email protected]> Signed-off-by: KumoLiu <[email protected]> Signed-off-by: YunLiu <[email protected]> Signed-off-by: Ishan Dutta <[email protected]> Signed-off-by: dependabot[bot] <[email protected]> Signed-off-by: kaibo <[email protected]> Signed-off-by: heyufan1995 <[email protected]> Signed-off-by: binliu <[email protected]> Signed-off-by: axel.vlaminck <[email protected]> Signed-off-by: Ibrahim Hadzic <[email protected]> Signed-off-by: Behrooz <[email protected]> Signed-off-by: Timothy Baker <[email protected]> Signed-off-by: Mathijs de Boer <[email protected]> Signed-off-by: Fabian Klopfer <[email protected]> Signed-off-by: Lucas Robinet <[email protected]> Signed-off-by: Lucas Robinet <[email protected]> Signed-off-by: chaoliu <[email protected]> Signed-off-by: cxlcl <[email protected]> Signed-off-by: chaoliu <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: monai-bot <[email protected]> Co-authored-by: elitap <[email protected]> Co-authored-by: Felix Schnabel <[email protected]> Co-authored-by: YanxuanLiu <[email protected]> Co-authored-by: ytl0623 <[email protected]> Co-authored-by: Dženan Zukić <[email protected]> Co-authored-by: Eric Kerfoot <[email protected]> Co-authored-by: YunLiu <[email protected]> Co-authored-by: Ishan Dutta <[email protected]> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: Kaibo Tang <[email protected]> Co-authored-by: Yufan He <[email protected]> Co-authored-by: binliunls <[email protected]> Co-authored-by: Ben Murray <[email protected]> Co-authored-by: axel.vlaminck <[email protected]> Co-authored-by: Mingxin Zheng <[email protected]> Co-authored-by: Ibrahim Hadzic <[email protected]> Co-authored-by: Dr. Behrooz Hashemian <[email protected]> Co-authored-by: Timothy J. Baker <[email protected]> Co-authored-by: Mathijs de Boer <[email protected]> Co-authored-by: Mathijs de Boer <[email protected]> Co-authored-by: Fabian Klopfer <[email protected]> Co-authored-by: Yiheng Wang <[email protected]> Co-authored-by: Lucas Robinet <[email protected]> Co-authored-by: Lucas Robinet <[email protected]> Co-authored-by: cxlcl <[email protected]>

Signed-off-by: Dr. Behrooz Hashemian <[email protected]>

drbeh · 2024-04-09T15:47:58Z

@NikolasSchmitz, I resolved the conflicts and updated the PR. Let's focus on the functionality and make this PR ready for reviewing. We can take care of any other issue once the PR is ready and please feel free to reach out to me or the working group if you still have any questions. Thanks for taking on this feature.

NikolasSchmitz · 2024-04-09T22:56:30Z

Thank you @drbeh ! I now marked it as ready for review.

ericspod · 2024-04-22T14:55:13Z

monai/data/wsi_reader.py

+
+ if within_tolerance:
+ # Take closest_level and continue with returning img at level
+ print(f"User-provided MPP lies within tolerance of level {closest_lvl}, returning wsi at this level.")


Print statements generally shouldn't be in production code, if you want to keep track of things you can log this instead. This goes for all the other print statements below.

Hi Eric, thanks for feedback!

ericspod · 2024-04-22T14:56:30Z

monai/data/wsi_reader.py

@@ -940,6 +1054,91 @@ def get_mpp(self, wsi, level: int) -> tuple[float, float]:

 raise ValueError("`mpp` cannot be obtained for this file. Please use `level` instead.")

+ def get_wsi_at_mpp(self, wsi, mpp: tuple, atol: float = 0.00, rtol: float = 0.05) -> np.ndarray:


This code appears to duplicate a lot of that in the other get_wsi_at_mpp method. Can this be factored out into a separate function to eliminate the duplication?

ericspod · 2024-04-22T14:57:35Z

@drbeh would you be able to review this? I think you're best qualified here. I made some minor comments about print and code duplication but otherwise it seems fine to me without having tested it. Thanks!

JHancox

Thanks for the effort @NikolasSchmitz. A welcome addition! I agree with @ericspod about the refactoring to remove duplication - that would be great.

JHancox · 2024-05-14T17:21:05Z

monai/data/wsi_reader.py

+ mpp_closest_lvl_x, mpp_closest_lvl_y = mpp_closest_lvl
+
+ # Define tolerance intervals for x and y of closest level
+ lower_bound_x = mpp_closest_lvl_x * (1 - rtol) - atol


I wonder whether the annotation ought to clarify the interaction between the absolute and relative tolerances. For example, someone might understandably assume that if either of the two values are exceeded then the value is deemed out of bounds, as opposed to the proposed behaviour which evaluates their combined tolerance.

We could clarify the interaction of both tolerances in the function's docstrings. Additionally, to prevent the two values from exceeding a certain limit, we could implement a validation check that raises a ValueError if either value is out of bounds. I will implement it, and we can review and decide later if we want to adopt this approach.

JHancox · 2024-05-14T17:39:35Z

monai/data/wsi_reader.py

+ target_res_x = int(np.round(closest_lvl_dim[1] * ds_factor_x))
+ target_res_y = int(np.round(closest_lvl_dim[0] * ds_factor_y))
+
+ closest_lvl_wsi = cucim_resize(wsi_arr, (target_res_x, target_res_y), order=0)


I think order=0 equates to nearest neighbour interpolation. The other methods seem to be using bilinear interpolation (order=1). If so, this would seem to be preferable. An alternative would be to have an optional parameter - possibly overkill.

Ah yes, you're right. That should be order=1 here. Thanks!
If wanted, I could keep bilinear interpolation as default, and add an optional parameter to enable the user to overwrite the default. Potentially, the user might also want to decide on which resize method to use (currently cucim.skimage.transform.resize and PIL.Image.resize), which could be another optional parameter.

NikolasSchmitz · 2024-05-16T17:16:27Z

Thank you for the feedback, @JHancox . Unfortunately, I couldn't join the meeting because I was on my way home from ICLR.
I will review the code and provide my updates as soon as possible.

Nikolas Schmitz added 5 commits March 22, 2024 14:07

Added function get_img_at_mpp to class OpenSlideWSIReader of module w…

f3e7d03

…si_reader.py

Added get_img_at_mpp to class CuCIMWSIReader

88002e8

Added function get_img_at_mpp to class TifffileWSIReader; changed res…

a9fe772

…izing function to Image.resize, cucim.skimage.transform.resize

Small changes

feac0dc

Small changes

8194026

NikolasSchmitz mentioned this pull request Mar 25, 2024

WSIReader support for reading by mpp or magnification #4980

Open

NikolasSchmitz force-pushed the 4980-get-wsi-at-mpp branch from 0b7322e to 1c10112 Compare March 27, 2024 00:38

cxlcl and others added 6 commits March 27, 2024 01:47

Renamed function to get_wsi_at_mpp

d989c18

Signed-off-by: Nikolas Schmitz <[email protected]>

[pre-commit.ci] auto fixes from pre-commit.com hooks

105f00b

for more information, see https://pre-commit.ci Signed-off-by: Nikolas Schmitz <[email protected]>

Reformatted wsi_reader.py

5db27c1

Signed-off-by: Nikolas Schmitz <[email protected]>

auto updates (Project-MONAI#7577)

18e82bd

Signed-off-by: monai-bot <[email protected]> Signed-off-by: monai-bot <[email protected]> Signed-off-by: Nikolas Schmitz <[email protected]>

Fixed return type

5bb531e

Signed-off-by: Nikolas Schmitz <[email protected]>

NikolasSchmitz force-pushed the 4980-get-wsi-at-mpp branch from 1c10112 to ae704f3 Compare March 27, 2024 00:49

NikolasSchmitz and others added 3 commits March 27, 2024 02:04

Small fixes

5214c56

Signed-off-by: Nikolas Schmitz <[email protected]>

NikolasSchmitz force-pushed the 4980-get-wsi-at-mpp branch from a65ef4a to 3264079 Compare March 27, 2024 01:08

Merge branch 'dev' into 4980-get-wsi-at-mpp

22ecc8c

Signed-off-by: Dr. Behrooz Hashemian <[email protected]>

NikolasSchmitz marked this pull request as ready for review April 9, 2024 22:55

ericspod requested a review from drbeh April 22, 2024 14:53

ericspod reviewed Apr 22, 2024

View reviewed changes

JHancox self-assigned this May 14, 2024

JHancox reviewed May 14, 2024

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Feature added: Get WSI at mpp #7574

Feature added: Get WSI at mpp #7574

NikolasSchmitz commented Mar 25, 2024 •

edited

drbeh commented Apr 9, 2024

NikolasSchmitz commented Apr 9, 2024

ericspod Apr 22, 2024

NikolasSchmitz Apr 24, 2024

ericspod Apr 22, 2024

ericspod commented Apr 22, 2024

JHancox left a comment

JHancox May 14, 2024

NikolasSchmitz May 20, 2024

JHancox May 14, 2024

NikolasSchmitz May 20, 2024

NikolasSchmitz commented May 16, 2024

		@@ -940,6 +1054,91 @@ def get_mpp(self, wsi, level: int) -> tuple[float, float]:

		raise ValueError("`mpp` cannot be obtained for this file. Please use `level` instead.")

		def get_wsi_at_mpp(self, wsi, mpp: tuple, atol: float = 0.00, rtol: float = 0.05) -> np.ndarray:

Feature added: Get WSI at mpp #7574

Are you sure you want to change the base?

Feature added: Get WSI at mpp #7574

Conversation

NikolasSchmitz commented Mar 25, 2024 • edited

Description

Types of changes

drbeh commented Apr 9, 2024

NikolasSchmitz commented Apr 9, 2024

ericspod Apr 22, 2024

Choose a reason for hiding this comment

NikolasSchmitz Apr 24, 2024

Choose a reason for hiding this comment

ericspod Apr 22, 2024

Choose a reason for hiding this comment

ericspod commented Apr 22, 2024

JHancox left a comment

Choose a reason for hiding this comment

JHancox May 14, 2024

Choose a reason for hiding this comment

NikolasSchmitz May 20, 2024

Choose a reason for hiding this comment

JHancox May 14, 2024

Choose a reason for hiding this comment

NikolasSchmitz May 20, 2024

Choose a reason for hiding this comment

NikolasSchmitz commented May 16, 2024

NikolasSchmitz commented Mar 25, 2024 •

edited