search

peng-lab / BaSiCPy

MIT License
66 stars 22 forks source link

Temporal Drift Setting #155

Open nicoleaudia opened 3 months ago

nicoleaudia commented 3 months ago

Hello! Really hoping I can make BaSiCPy work for my use case, but I'm having trouble reproducing the same results as I get in FIJI. I wonder if this is due to the temporal drift setting; I have it set to "Replace with zero" in my FIJI macro and I'm not sure how to recreate that setting with BaSiCPy. Does this setting exist?

Here is the line from my macro I'm trying to reproduce: run("BaSiC ", "processing_stack=[Combined Stacks] flat-field=None dark-field=None shading_estimation=[Estimate shading profiles] shading_model=[Estimate flat-field only (ignore dark-field)] setting_regularisationparametes=Manual temporal_drift=[Replace with zero] correction_options=[Compute shading and correct images] lambda_flat=5 lambda_dark=0.50");

Here is my python code that doesn't work:

basic = BaSiC(
        get_darkfield=False, 
        smoothness_flatfield=5.0, 
        smoothness_darkfield=0.5,
    )
basic.fit(images)
images_transformed = basic.transform(images)

To verify whether my code is doing the same thing my FIJI macro does, I've been looking at the mean base fluorescence measurements for my image stack. I've tried including the timelapse parameter for the transform function, and while that does get the measurements somewhat closer to the goal numbers, they're still quite off.

yfukai commented 3 months ago

Hi thanks for the question! Maybe @tying84 have some idea.