Not finding Median values in CIP

Saffana · October 6, 2020, 1:22pm

Hello

I am Saffana a PhD researcher in the UK working on COPD CT scans. I am currently using CIP extension to perform lung segmentation using parenchymal analysis

I am wondering if there is any way I can have median values, not just the mean values

Please can you help me with this?

Thank you

raul · October 6, 2020, 1:42pm

Hi Saffana,

The module does not provide median values but it could be easily modify to do so.
This are the lines in the SlicerCIP where the computation of the different phenotypes are performed:

github.com

acil-bwh/SlicerCIP/blob/develop/Scripted/CIP_ParenchymaAnalysis/CIP_ParenchymaAnalysis.py#L650-L670


self.labelStats['LAA%-950', tag] = 100.0 * (data < -950).sum() / float(data.size)
self.labelStats['LAA%-925', tag] = 100.0 * (data < -925).sum() / float(data.size)
self.labelStats['LAA%-910', tag] = 100.0 * (data < -910).sum() / float(data.size)
self.labelStats['LAA%-856', tag] = 100.0 * (data < -856).sum() / float(data.size)
self.labelStats['HAA%-700', tag] = 100.0 * (data > -700).sum() / float(data.size)
self.labelStats['HAA%-600', tag] = 100.0 * (data > -600).sum() / float(data.size)
self.labelStats['HAA%-500', tag] = 100.0 * (data > -500).sum() / float(data.size)
self.labelStats['HAA%-250', tag] = 100.0 * (data > -250).sum() / float(data.size)
self.labelStats['HAA%-600-250', tag] = 100.0 * (numpy.logical_and(data > -600,
                                                                  data < -250)).sum() / float(data.size)
# self.labelStats[cycle,'Perc10',tag]=self.percentile(data,.1)
# self.labelStats[cycle,'Perc15',tag]=self.percentile(data,.15)
self.labelStats['Perc10',tag]=numpy.percentile(data,10)
self.labelStats['Perc15',tag]=numpy.percentile(data,15)
self.labelStats['Mean', tag] = mean_data
self.labelStats['Std', tag] = std_data
self.labelStats['Kurtosis', tag] = self.kurt(data, mean_data, std_data)
self.labelStats['Skewness', tag] = self.skew(data, mean_data, std_data)
self.labelStats['Ventilation Heterogeneity', tag] =  self.vh(data)
self.labelStats['Mass',tag] = self.mass(data,cubicMMPerVoxel)
self.labelStats['Volume', tag] = data.size * cubicMMPerVoxel * litersPerCubicMM

You could extend this list of phenotypes with your own operation.

An alternative approach, it is directly used the mask to build your own module that computes the quantities that you want to explore.

I hope this helps

Raul

Saffana · October 6, 2020, 2:10pm

Hi Raul

Thank you for your reply

Unfortunately, I don’t have any programming background I am from the medical field only to perform CT scan analysis

can you please tell me if there is any way in the software without coding or programming?

Thank you

Saffana

Topic		Replies	Views
Exporting CSV with parenchyma analysis module Support cip	58	2003	April 13, 2022
Parenchyma analysis Support lungctanalyzer , lung	4	651	October 25, 2022
Lung airway segmentation SlicerCIP cip	43	5693	July 10, 2018
How to go to cumulative histogram from lung density histogram in parenchyma analysis module? SlicerCIP cip	4	1688	April 24, 2017
Problem with the Interactive lobe segmentation module in the CIP SlicerCIP	1	743	August 20, 2020

Not finding Median values in CIP

Related topics