Picket Fence using CR Karestream

Pedro Argolo Piedade

unread,

Mar 26, 2020, 11:15:31 AM3/26/20

to Pylinac

James it si possible uses CR Karestream to do this test ? I do not have EPID.

Thanks

James Kerns

unread,

Apr 3, 2020, 6:09:23 PM4/3/20

to Pylinac

If your CR outputs DICOM then it should be possible although I have not tried. CR is not a use case in Pylinac.

Pedro Argolo Piedade

unread,

Mar 15, 2023, 4:37:31 PM3/15/23

to Pylinac

Hi James, how are you?

I create a code to convert CR into EPID image.

I run the pylinac picket fence. But I think still getting the colimated part and not the irradiated gap.

Attached the files. I think we need to chage the way we irradiate de CR. It has a sensitivity very different from the EPID.

My best regards

Pedro P

rtplan.ipynb

James Kerns

unread,

Mar 20, 2023, 5:04:43 PM3/20/23

to Pylinac

Hello Pedro,

Thanks for sharing. It looks like your picket detection is inverted? The tongue and groove lines appear on the darker areas of the image, meaning the brighter areas are the pickets, no?

James Kerns

unread,

May 15, 2023, 3:20:10 PM5/15/23

to Pedro Argolo Piedade, Pylinac

Analyzing both of your images with a default analysis gives me the correct pickets as far as I can tell. What version of pylinac are you using? Also, you should consider spacing out the "pickets" further apart. See point 2: Picket Fence — pylinac 3.10.0 documentation

On Mon, May 15, 2023 at 11:28 AM Pedro Argolo Piedade <pedroar...@gmail.com> wrote:

Hi James, I have made a new test, but I believe is a Garden Fence, a static shoots.

As a CR image we have (0028, 0004) Photometric Interpretation = 'MONOCHROME2'

I have inverted the pixel but still getting in the wrong area

I have attached the dicom files for you.

Can you give me some ideas?

My best regards and sorry for the delay on answering

Pedro P

--
You received this message because you are subscribed to a topic in the Google Groups "Pylinac" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/pylinac/-LiX5Gj7JXM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to pylinac+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pylinac/37f89e57-fb97-416f-808e-4b1cc82824adn%40googlegroups.com.

--
Pedro Argôlo Piedade
Físico Médico - Radioterapia São Sebastião

www.radioterapiass.com.br
+55-048-91672105
pedroar...@gmail.com
skype: pedroargolopie

whatsapp: +554891672105

Pedro Argolo Piedade

unread,

May 15, 2023, 4:36:55 PM5/15/23

to James Kerns, Pylinac

Thanks for your feedback.

I am using Pylinac 3.4.0., python 3.9.7 and notebooks [GCC 9.4.0] 6.4.4

Pedro Argolo Piedade

unread,

Dec 4, 2023, 11:11:07 AM12/4/23

to James Kerns, Pylinac

Hi James how are you?

I finished the code for converting CR to EPID images.

I have a upload file issue when performing the picketed fence test. Is there a limitation on the file size to upload? I struggle to debug with Django framework. I do not find the error on the Logs files debug.log.

Do you have any insights?

My best regards

Pedro P

Em seg., 15 de mai. de 2023 16:20, James Kerns <jker...@gmail.com> escreveu:

James Kerns

unread,

Dec 4, 2023, 5:48:38 PM12/4/23

to Pylinac

Django? Pylinac does not use Django. It looked like you were using Google colab previously. Are you using QATrack+? If so, this is a question to ask on the correct forum: QATrack+ - Google Groups. There is a file upload setting you might be running into here, but this has to do w/ streaming vs writing to disk which may or may not be related to your issue: https://docs.djangoproject.com/en/4.2/ref/settings/#std-setting-FILE_UPLOAD_MAX_MEMORY_SIZE

Pedro Argolo Piedade

unread,

Dec 5, 2023, 10:03:40 AM12/5/23

to James Kerns, Pylinac, Thiago Fontana, Felipe Eisenhut

Sorry James, I am using QaTrack+ to analyze the garden fence (CR plates images saturate with picket fence).

I am using the test list:

The Macro function I use is Pylinac.

Usually the code look like this:

import io
import pylinac

# run the picket fence analysis using pylinac
# note the use of `BIN_FILE` rather than `FILE` since we are dealing with an image
pf = pylinac.PicketFence(BIN_FILE.path, filter=5)
pf.analyze(tolerance=0.5, action_tolerance=0.03, invert=False)

# our dictionary of results to return
pf_upload_analysis = {
"percent passing": pf.percent_passing,
"max error": pf.max_error,
"number of pickets": pf.num_pickets,
"orientation": pf.orientation,
}

# create a pylinac PDF and create an attachment with it
data = io.BytesIO()
pf.publish_pdf(data)
UTILS.write_file("testingpfpdf.pdf", data)

# convert the image to a png file so it can be displayed when performing the test list
UTILS.write_file("pf-image.png", pf.image)

I was going to adapt the function CRtoEpid in this code.

def CRtoEPID(fpath):
# authors : Guillaume Lemaitre <g.lema...@gmail.com>
# license : MIT

# FIXME: add a full-sized MR image in the testing data
filename = ('EPID_base/AS1200.dcm')
ds = pydicom.dcmread(filename)

filename2 = (fpath)
ds2 = pydicom.dcmread(filename2)

# get the pixel information into a numpy array
data = ds.pixel_array
data2 = ds2.pixel_array

#(0028, 0004) Photometric Interpretation CS: 'MONOCHROME2'
if ds2.PhotometricInterpretation == 'MONOCHROME2':
# Inverter os valores de pixel
max_value = np.amax(data2)
imagem_invertida = max_value - data2
else:
imagem_invertida = data2

print('The image has {} x {} voxels'.format(data.shape[0],data.shape[1]))
data_downsampling = imagem_invertida
print('The downsampled image has {} x {} voxels'.format(data_downsampling.shape[0], data_downsampling.shape[1]))

# copy the data back to the original data set
ds.PixelData = data_downsampling.tobytes()
# update the information regarding the shape of the data array
ds.Rows, ds.Columns = data_downsampling.shape

#Photometric Interpretation MONO1 para MONO2
ds.PhotometricInterpretation = ds2.PhotometricInterpretation

#CR (0018, 1164) Imager Pixel Spacing DS: [0.168, 0.168]
#EPID (3002, 0011) Image Plane Pixel Spacing DS: [0.336, 0.336]
ds.ImagePlanePixelSpacing = ds2.ImagerPixelSpacing

#RT Image SID
ds.RTImageSID = "1000.0"

#(0028, 0002) Samples per Pixel US: 1
#ds.SamplesperPixel = ds2.SamplesperPixel
#So tem no CR (0028, 0034) Pixel Aspect Ratio IS: [1, 1]
#(0028, 0100) Bits Allocated US: 16
ds.BitsAllocated = ds2.BitsAllocated
#(0028, 0101) Bits Stored US: 12
ds.BitsStored =ds2.BitsStored
#(0028, 0102) High Bit US: 11
ds.HighBit =ds2.HighBit
#(0028, 0103) Pixel Representation US: 0
ds.PixelRepresentation = ds2.PixelRepresentation
#(0028, 0106) Smallest Image Pixel Value US: 0
# Verificar e copiar SmallestImagePixelValue, se existir
if hasattr(ds2, 'SmallestImagePixelValue'):
ds.SmallestImagePixelValue = ds2.SmallestImagePixelValue
else:
ds.SmallestImagePixelValue = 0
#(0028, 0107) Largest Image Pixel Value US: 4095
if hasattr(ds2, 'LargestImagePixelValue'):
ds.LargestImagePixelValue = ds2.LargestImagePixelValue
else:
ds.LargestImagePixelValue = 0

# so tem no CR (0028, 0301) Burned In Annotation CS: 'NO'
#(0028, 1050) Window Center DS: '2048.0'
ds.WindowCenter = ds2.WindowCenter
#(0028, 1051) Window Width DS: '4096.0'
ds.WindowWidth = ds2.WindowWidth
#(0028, 1052) Rescale Intercept DS: '0.0'
ds.RescaleIntercept = ds2.RescaleIntercept
#(0028, 1053) Rescale Slope DS: '1.0'
ds.RescaleSlope = ds2.RescaleSlope
#(0028, 1054) Rescale Type LO: 'US'
ds.RescaleType = ds2.RescaleType
#so tem no CR (0028, 2110) Lossy Image Compression CS: '00'

#gerando novo UID
novouid = generate_uid()
#print(ds.SOPInstanceUID)
ds.SOPInstanceUID = novouid

pat_name = ds2.PatientName
# salvando a imagem
# Obter a data atual
data_atual = datetime.now()
nome_arquivo = "ImagensConvertidas/CRepid_"+ds2.StudyDate+"."+ds2.StudyTime+"."+str(data_atual)+".dcm"
cr_MOD = nome_arquivo.replace(':', '')
#cr_MOD = "ImagensConvertidas/CRepid_"+ds2.StudyDate+"."+ds2.StudyTime+"."+str(data_atual)+".dcm"
ds.save_as(cr_MOD)

return cr_MOD

Performing the weekly test

Performing the test:

Uploanding the converted dicom image using the Notebook to create this new Image we usually have 10 to 30MB, I have changed the CR configuration to decrease the size of the files.