Datasets for baseline recovery

yinan-wan commented 7 years ago

Datasets for baseline recovery:

20160328 (best dataset so far, let's focus on this one first)
20170111
20170925 - this is a newly added dataset that you may not have processed yet, I just copied the data to the shared folder

yinan-wan commented 6 years ago

Found another good dataset that could be used for tracing: 170201. Baseline really need to be corrected

yinan-wan commented 6 years ago

To summarize what we have right now:

Datasets we have for lineaging analysis: DirNames{6} = '../Data/20160328'; DirNames{9} = '../Data/20161027'; DirNames{11} = '../Data/20170111'; DirNames{14} = '../Data/20170201'; DirNames{22} = '../Data/20170731'; DirNames{24} = '../Data/20170925';
Problems still existing in the current version of activation test 1) missed neurons with weak but correlated activity with others 2) some of the big initial peaks still missing (minor problem) 3) some frames have too many active neurons (only dataset 20170201) Videos and screenshots of problematic time windows are in dropbox (ActivationTest)
When we move to estimation of halfActTime and halfEVTime, please provide some confidence interval so we can better rank the activation order of neurons

zqwei commented 6 years ago

for nFile = [6, 9, 11, 14, 22, 24]
    Neuron_selection_v3(nFile)
end

in activation_test branch

zqwei / Zebrafish_spinal_cord_development