ImagesPrepocessor

A lot of biomedical image processing python scripts, like pre-process the tiff images of membrane channel to NTFI {embryo_name}_{tp}.nii.gz (medical 3D MRT image).

The following Papers/Projects Used this Repository, and Please Cite Them if You Use Our Tools and Data

• Paper Tile: Cell lineage-resolved embryonic morphological map reveals novel signaling regulating cell fate and size asymmetry; Website: https://bcc.ee.cityu.edu.hk/cmos/index.html ; Code Link: https://github.com/cao13jf/CMap and https://github.com/chiellini/GUIData ; Related Data Link: https://figshare.com/s/fc9b67e91a38eea86bee .
• Paper Title: Deep Learning-based Enhancement of Fluorescence Labeling for Accurate Cell Lineage Tracing During Embryogenesis; Code Link: https://github.com/plcx/NucApp-develop ; Related Data Link: https://doi.org/10.6084/m9.figshare.26778475.v1 .
• Paper Title: EmbSAM: Cell boundary localization and Segment Anything Model for 3D fast-growing embryos; Website: https://bcc.ee.cityu.edu.hk/cmos/embsam/ ; Code Link: https://github.com/CunminZhao/EmbSAM; Related Data Link: https://portland-my.sharepoint.com/:f:/g/personal/zelinli6-c_my_cityu_edu_hk/Epj5LhqViNZCmNtmRZrE2D8BvvGTr09Jg9u9aFBstL-3cg .

Compact the 2D tiff image to 3D NTFT MRT file *.nii.gz.

• Function stack_memb_slices in file compose_slice.py
  • Using skimage.transform.resize: spline interpolation (z axis)
• According to their ratio of xy_resolution and z_resolution

Conduct some traditional methods to enhance the images

• Like 3DMNS density map

Transform 3D NTfTI MRT file *.nii.gz to 3D TIFF FILES

• Open file "image2d3d_format_transformation.py"

• Add all embryos in list of embryo_names.

• Run function "nifti2tiff_seperated(seg_cell_root, tiff_root, segmented=True)"

• Seg_cell_root is input path and tiff_root is output path

• Input path contains floders(samples) of nii.gz files

• Output path contains .tiff floders and tiffmaptxt floders.

  

Transform 3D TIFF FILES to .obj FILES by ImageJ

• Modify draw3DObject.ijm

• Modify tiff_input path which is output path contains .tiff floders and tiffmaptxt floders by 3D_format_transformation.py. and obj_output path in .ijm. Add embryos in embryonames_list.

• Open ImageJ click Plugions, Macros, Run and add draw3DObject.ijm.

Solve imageJ crash

• Run get_index.ijm and save log as txt.

• Find last output obj files and find its index.

• Because there are serval obj files in a index. Log has information to remapping it.

• Modify start index "i" in second for loop in draw3DObject.ijm.

• ImageJ can continue generating obj after last time.

Combine the Seperated Objs into one and assign the cell Name

• run file "generate_obj_from_seperated_objs.py"

Obj view in ImageJ

File structure

CITATION

Guan, G., Li, Z., Ma, Y. et al. Cell lineage-resolved embryonic morphological map reveals signaling associated with cell fate and size asymmetry. Nat Commun 16, 3700 (2025). https://doi.org/10.1038/s41467-025-58878-0