dynamic-t-rex alpha
CHO cells nuclei virtual staining - brightfield - Pix2Pix
Estibaliz Gómez de Mariscal, Mario Del Rosario
On this page
Model Information
Description: A conditional cycleGAN trained to infer the cell membrane as seen in a widefield fluorescence image from phase contrast images of CHO cells imaged with phase contrast microscopy. The cells are imaged with a 20x objective.
Task: prediction of cell membrane localisation
Licence: MIT
Tags: 2D, cells, in silico labeling, transmission light microscopy, fluorescence, conditional gan, cyclegan, zerocostdl4mic, deepimagej, pytorch, actin, dapi, nuclei
Benchmarking
This model was not tested on datasets with reference annotations, so we cannot produce benchmarking metrics.
Model predictions
| Study | Dataset | Original image | Prediction | Analysis parameters |
|---|---|---|---|---|
| S-BIAD1269 | CHO cell mitotic rounding raw brightfield microscopy images (for annotations) | 
Download | 
Download | xy_image_crop_size: 1024,1024 |
| S-BIAD1466 | Nanoscale analysis of human G1 and metaphase chromatin in situ | 
Download | 
Download | z_slices_analysed: 12,13 input_channel_analysed: 1 t_slices_analysed: 0,1 |