| Both sides previous revision
Previous revision
Next revision
|
Previous revision
|
software:gpu_dispimfusion [2019/08/08 21:02]
Melissa Glidewell |
software:gpu_dispimfusion [2024/11/04 18:36] (current)
Jon Daniels |
| ====== diSPIM preprocessing & GPU diSPIMFusion ====== | ====== diSPIM preprocessing & GPU diSPIMFusion ====== |
| |
| For a 3D image with isotropic resolution, diSPIM images typically require preprocessing before processing. An ImageJ macro with a straight-forward user interface, is provided for each purpose. | To obtain a 3D image with isotropic resolution, diSPIM images typically require preprocessing before fusion. An ImageJ macro with a straight-forward user interface, is provided for each purpose. |
| |
| Preprocessing involves background subtraction, region of interest (ROI) cropping, and 3D orientation; additionally, stage-scanning acquisitions must undergo [[https://www.ncbi.nlm.nih.gov/pubmed/27638693|coordinate transformation]]. The diSPIM Preprocessing macro performs all 4 tasks as necessary. It requires ImageJ version 1.48c or later, with Windows 7 OS. | Preprocessing involves background subtraction, region of interest (ROI) cropping, and 3D orientation; additionally, stage-scanning acquisitions must undergo [[https://www.ncbi.nlm.nih.gov/pubmed/27638693|coordinate transformation]]. The diSPIM Preprocessing macro performs all 4 tasks as necessary. It requires ImageJ version 1.48c or later, with Windows 7 OS. |
| |
| Processing is constituted by image registration, joint deconvolution, and finally fusion. The GPU diSPIMFusion macro seamlessly performs all three functions. It capitalizes on the GPU for parallelized computation and therefore much faster results. It requires ImageJ, Windows 7 or 10 OS, a graphics card supported by CUDA 9, and an appropriate driver for the graphics card. | Fusion is constituted by image registration and joint deconvolution. The GPU diSPIMFusion macro seamlessly performs both functions. It capitalizes on the GPU for parallelized computation and therefore much faster results. It requires ImageJ, Windows 7 or 10 OS, a graphics card supported by CUDA 9, and an appropriate driver for the graphics card. |
| \\ | \\ |
| \\ | \\ |
| Manual and code are available at [[https://github.com/eguomin/diSPIMFusion|GitHub]].\\ | Manual and code are available at [[https://github.com/eguomin/diSPIMFusion|GitHub]].\\ |
| Preprint is found on [[https://www.biorxiv.org/content/10.1101/647370v1|bioRxiv]].\\ | Published at https://doi.org/10.1038/s41587-020-0560-x (preprint at [[https://www.biorxiv.org/content/10.1101/647370v1|bioRxiv]]) \\ |
| Contact Min Guo at [email protected] with questions.\\ | Contact Min Guo at [email protected] with questions.\\ |
| |
