I2K 2022 “developing open source image analysis platforms/tools”
- Abstract til 1st March
- Event 6-10 May (virtual)
- https://forum.image.sc/t/i2k-2022-conference-talk-workshop-submissions-are-open/62833
- Zarr and in-house data [experimental]
- Overview table
.md
for maintaining.html
(use table gen service, which I don’t like it... but whatever)- put sample image links
- embed in docs and github readme
- Docs, notebook examples
- Module docs
- Quickstart
- clean README.md
- Hard-code numbers of datasets so that developers don't need to download all
datasets to plot histograms.
- Modify the plotting notebook accordingly
- indicate datasets that need pre-parse somehow?
- Data vis
- Models
- bioimage.io
- Run and eval models
- Summary table which model excels in which dataset
- CommonDataset, CommonMaskDataset
- re-ordering channels np.ascontiguousarray, 'C' order, arr[..., (2, 1, 0)]
- random sampling, shuffle in BatchDataLoader
-
DatasetList
- [ ] Where to put? types.py is not a good place... - Way to set
num_samples
withtransforms
argument inConfig
Two arguments are closely related - Metrics for benchmarking (StarDist has done a great job, their license is BSD-3)
- (maybe nope) Download scripts
- Take out those that do not have mask anno and put them in
Dataset
- Implement
__getitem__
forDataset
- Change base for them
- Implement
- Fix data[’mask’] # (b, h, w) → (b, 1, h, w)? (necessary?) Not really necessary? Just have them (b, h, w) for now
- Fix data['mask'].dtype == bool
When mask has a single channel, make them have the same dtype.
Albumentations supports only UINT8 and FLOAT32.
- bool -> uint8
- or int16 (because why not... mask.dtype does not matter)
- fix plt.rcParams['image.interpolation'] does not work in
./notebooks/_sample_images.ipynb
cv2.resize was an issue, not matplotlib - update overview table
- Reordered
- Add missing ones
- BBBC041
- consistent
root_dir
. Now it is a mess. We want something described in thedocs/user_guides/basic0_prepare_datasets.rst
. - substitute
num_calls
withnum_samples
? (I think samples sound right) - Differentiate ● (O) and ● (OI: instance outlines)
- FRUNet
normalize
, better way? correct way? - number of images, including test sets
- Load all anno types, if there are more than one (e.g. BBBC007)
-
image_ch
if possible (in any case where channels are separable)- Done: BBBC006, BBBC020
- Not done: need to find them
- BBBC020 anno dir name is not
BBBC020*
butBBC020*
. Need to check if it's the case for both types. FollowBBC020*
?
- Migrate repo to LOB account and open to public
- DatasetList: List[Dataset]
- [x] to do batch processing, such as
foreach_sample_by_indices
- [ ] Where to put?types.py
is not a good place - Normalization
normalize
andnormalize_mode
? -
grayscale_mode
: list[float] Important one. Images with 2 channels usinggrayscale_mode=eqaul
should be fixed by list[float]. - More data
- TissueNet
- https://twitter.com/cmci_/status/1498156970700046337?t=9n2gDTpeM1Oq3JcCOV9QOg
All 3 are related. Rooted from one and modified somehow?
- Romain Guiet. (2022). HeLa "Kyoto" cells under the scope (Version v0) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.6139958
- Romain Guiet, & Olivier Burri. (2022). Cellpose models for Label Prediction from Brightfield and Digital Phase Contrast images (Version v0) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.6140111
- Romain Guiet. (2022). Automatic labelling of HeLa "Kyoto" cells using Deep Learning tools (Version v0) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.6140064
- CellImageLibrary http://www.cellimagelibrary.org/images/CCDB_6843 Included in Cellpose dataset, cytoplasmic (phalloidin) and nuclear stains (DAPI)
- CRCHisto
- CEM500K
- OpenCell https://opencell.czbiohub.org/
- LIVECellt
- StarDist (subset of DSB2018)
- Cellpose
- hydra for Config https://hydra.cc/docs/intro/
- Multi-threading LIVECell
- Config.expand(): concatenate Configs. Q) What if there are overlaps?
- Refactor
__getitem__()
for "image" inDataset
andMaskDataset
, so that they can share the same code for getting "image". Also it is repeated inoutput=image
andoutput=both
. - AnnoDatasetInterface
Interface for Dataset with annotation, base for all Dataset with annotation.
Currently attr like
anno_dict
is not an abstract property, though it should be once we have other types, BboxDataset for instance, - BboxDataset
- BBBC041
- and more
- 3D [experimental]
- need 3D augmentation lib
- Custom augmentation ex) 5 channels
- time-series
All datasets are very unique. Read each docstring for details.
- BBBC006 The same field of view with varying focus. zind=16 is the most in-focus plane. Thus it loads only zind=16 by default.