Neste tutorial veremos exemplos de como manipular imagens em arquivos DICOM com as bibliotecas pydicom e ImageIO.
Documentação básica sobre leitura, acesso, modificação e escrita de arquivos DICOM
pydicom
imageio
Bibliotecas e pacotes
!pip install pydicom
Collecting pydicom
[?25l Downloading https://files.pythonhosted.org/packages/f4/15/df16546bc59bfca390cf072d473fb2c8acd4231636f64356593a63137e55/pydicom-2.1.2-py3-none-any.whl (1.9MB)
[K |████████████████████████████████| 1.9MB 5.4MB/s
[?25hInstalling collected packages: pydicom
Successfully installed pydicom-2.1.2
import numpy as np
import imageio
from pydicom import dcmread
import matplotlib.pyplot as plt
%matplotlib inline
Tutorial
Fontes das imagens
FAÇA VOCÊ
Baixe e faça o upload do arquivo DICOM
pydicom
Carregar arquivos
dicom_ds = dcmread("0015.DCM")
Visualizar a lista de elementos de dados
print(dicom_ds)
Dataset.file_meta -------------------------------
(0002, 0000) File Meta Information Group Length UL: 202
(0002, 0001) File Meta Information Version OB: b'\x00\x01'
(0002, 0002) Media Storage SOP Class UID UI: X-Ray Radiofluoroscopic Image Storage
(0002, 0003) Media Storage SOP Instance UID UI: 1.2.840.113619.2.15.1008000062035011254.825190719.0.31.2.1
(0002, 0010) Transfer Syntax UID UI: Explicit VR Little Endian
(0002, 0012) Implementation Class UID UI: 1.2.840.113619.6.36
(0002, 0013) Implementation Version Name SH: '1_2_5'
(0002, 0016) Source Application Entity Title AE: 'ard-demo'
-------------------------------------------------
(0008, 0005) Specific Character Set CS: 'ISO_IR 100'
(0008, 0008) Image Type CS: ['ORIGINAL', 'PRIMARY', 'SINGLE PLANE']
(0008, 0016) SOP Class UID UI: X-Ray Radiofluoroscopic Image Storage
(0008, 0018) SOP Instance UID UI: 1.2.840.113619.2.15.1008000062035011254.825190719.0.31.2.1
(0008, 0020) Study Date DA: '19960308'
(0008, 0021) Series Date DA: '19960308'
(0008, 0022) Acquisition Date DA: '19960308'
(0008, 0023) Content Date DA: '19960308'
(0008, 0030) Study Time TM: ''
(0008, 0032) Acquisition Time TM: '105650'
(0008, 0033) Content Time TM: '105650'
(0008, 0050) Accession Number SH: ''
(0008, 0060) Modality CS: 'RF'
(0008, 0070) Manufacturer LO: 'GE MEDICAL SYSTEMS'
(0008, 0090) Referring Physician's Name PN: ''
(0008, 1010) Station Name SH: ''
(0008, 1030) Study Description LO: '5'
(0008, 103e) Series Description LO: ''
(0008, 1050) Performing Physician's Name PN: '00558747^'
(0008, 1090) Manufacturer's Model Name LO: 'DRS'
(0010, 0010) Patient's Name PN: 'Rubo DEMO'
(0010, 0020) Patient ID LO: '10-55-87'
(0010, 0030) Patient's Birth Date DA: ''
(0010, 0040) Patient's Sex CS: 'F'
(0010, 1010) Patient's Age AS: ''
(0018, 0060) KVP DS: None
(0018, 1020) Software Versions LO: '4.00'
(0018, 1150) Exposure Time IS: None
(0018, 1151) X-Ray Tube Current IS: None
(0018, 1155) Radiation Setting CS: 'GR'
(0018, 1400) Acquisition Device Processing Descr LO: '10 OUT OF 20. 0=LOW, 20=HIGH CONVOLUTION KERNEL'
(0018, 1600) Shutter Shape CS: ['CIRCULAR', 'RECTANGULAR']
(0018, 1602) Shutter Left Vertical Edge IS: "10"
(0018, 1604) Shutter Right Vertical Edge IS: "950"
(0018, 1606) Shutter Upper Horizontal Edge IS: "10"
(0018, 1608) Shutter Lower Horizontal Edge IS: "950"
(0018, 1610) Center of Circular Shutter IS: [0480, 0480]
(0018, 1612) Radius of Circular Shutter IS: "470"
(0020, 000d) Study Instance UID UI: 1.2.840.113619.2.15.1008000062035011254.825190719.2.31
(0020, 000e) Series Instance UID UI: 1.2.840.113619.2.15.1008000062035011254.825190719.1.31
(0020, 0010) Study ID SH: ''
(0020, 0011) Series Number IS: "1"
(0020, 0013) Instance Number IS: "2"
(0020, 0020) Patient Orientation CS: ''
(0020, 4000) Image Comments LT: 'L1'
(0028, 0002) Samples per Pixel US: 1
(0028, 0004) Photometric Interpretation CS: 'MONOCHROME2'
(0028, 0010) Rows US: 1024
(0028, 0011) Columns US: 1024
(0028, 0100) Bits Allocated US: 8
(0028, 0101) Bits Stored US: 8
(0028, 0102) High Bit US: 7
(0028, 0103) Pixel Representation US: 0
(0028, 1040) Pixel Intensity Relationship CS: 'LIN'
(0028, 1050) Window Center DS: "127.0"
(0028, 1051) Window Width DS: "255.0"
(0028, 2110) Lossy Image Compression CS: '00'
(0028, 6010) Representative Frame Number US: 1
(0028, 6020) Frame Numbers of Interest (FOI) US: 2
(0028, 6022) Frame of Interest Description LO: 'L1'
(0037, 0010) Private Creator LO: 'GEMS_DRS_1'
(0037, 1010) [Refering Department] LO: 'EX 101^'
(0037, 1020) [Screen Number] US: 2
(0037, 1040) [Left Orientation] CS: ''
(0037, 1042) [Right Orientation] CS: ''
(0037, 1050) [Inversion] CS: 'INV_NORMAL'
(0037, 1060) [DSA] CS: ''
(7fe0, 0010) Pixel Data OB: Array of 1048576 elements
Metadados
dicom_ds.file_meta
(0002, 0000) File Meta Information Group Length UL: 202
(0002, 0001) File Meta Information Version OB: b'\x00\x01'
(0002, 0002) Media Storage SOP Class UID UI: X-Ray Radiofluoroscopic Image Storage
(0002, 0003) Media Storage SOP Instance UID UI: 1.2.840.113619.2.15.1008000062035011254.825190719.0.31.2.1
(0002, 0010) Transfer Syntax UID UI: Explicit VR Little Endian
(0002, 0012) Implementation Class UID UI: 1.2.840.113619.6.36
(0002, 0013) Implementation Version Name SH: '1_2_5'
(0002, 0016) Source Application Entity Title AE: 'ard-demo'
Acessando elementos pelas tags
dicom_ds[0x0008,0x0016]
(0008, 0016) SOP Class UID UI: X-Ray Radiofluoroscopic Image Storage
Acessando o array da imagem
img_xray = dicom_ds.pixel_array
img_xray.shape
(1024, 1024)
plt.imshow(img_xray, cmap="gray", vmin=0, vmax=255)
plt.colorbar()
<matplotlib.colorbar.Colorbar at 0x7f617d2a9da0>
Outro exemplo
Exemplo disponível na documentação.
# authors : Guillaume Lemaitre <g.lemaitre58@gmail.com>
# license : MIT
import matplotlib.pyplot as plt
from pydicom import dcmread
from pydicom.data import get_testdata_file
fpath = get_testdata_file('CT_small.dcm')
ds = dcmread(fpath)
# Normal mode:
print()
print(f"File path........: {fpath}")
print(f"SOP Class........: {ds.SOPClassUID} ({ds.SOPClassUID.name})")
print()
pat_name = ds.PatientName
display_name = pat_name.family_name + ", " + pat_name.given_name
print(f"Patient's Name...: {display_name}")
print(f"Patient ID.......: {ds.PatientID}")
print(f"Modality.........: {ds.Modality}")
print(f"Study Date.......: {ds.StudyDate}")
print(f"Image size.......: {ds.Rows} x {ds.Columns}")
print(f"Pixel Spacing....: {ds.PixelSpacing}")
# use .get() if not sure the item exists, and want a default value if missing
print(f"Slice location...: {ds.get('SliceLocation', '(missing)')}")
# plot the image using matplotlib
plt.imshow(ds.pixel_array, cmap=plt.cm.gray)
plt.show()
File path........: /usr/local/lib/python3.6/dist-packages/pydicom/data/test_files/CT_small.dcm
SOP Class........: 1.2.840.10008.5.1.4.1.1.2 (CT Image Storage)
Patient's Name...: CompressedSamples, CT1
Patient ID.......: 1CT1
Modality.........: CT
Study Date.......: 20040119
Image size.......: 128 x 128
Pixel Spacing....: [0.661468, 0.661468]
Slice location...: -77.2040634155
ImageIO
Carregar arquivos
im = imageio.imread("0015.DCM")
im.shape
(1024, 1024)
Metadados
im.meta
Dict([('TransferSyntaxUID', '1.2.840.10008.1.2.1'),
('SOPClassUID', '1.2.840.10008.5.1.4.1.1.12.2'),
('SOPInstanceUID',
'1.2.840.113619.2.15.1008000062035011254.825190719.0.31.2.1'),
('StudyDate', '19960308'),
('SeriesDate', '19960308'),
('AcquisitionDate', '19960308'),
('ContentDate', '19960308'),
('StudyTime', ''),
('AcquisitionTime', '105650'),
('ContentTime', '105650'),
('Modality', 'RF'),
('Manufacturer', 'GE MEDICAL SYSTEMS'),
('StudyDescription', '5'),
('SeriesDescription', ''),
('PatientName', 'Rubo DEMO'),
('PatientID', '10-55-87'),
('PatientBirthDate', ''),
('PatientSex', 'F '),
('PatientAge', ''),
('StudyInstanceUID',
'1.2.840.113619.2.15.1008000062035011254.825190719.2.31'),
('SeriesInstanceUID',
'1.2.840.113619.2.15.1008000062035011254.825190719.1.31'),
('SeriesNumber', 1),
('InstanceNumber', 2),
('PatientOrientation', ''),
('SamplesPerPixel', 1),
('Rows', 1024),
('Columns', 1024),
('BitsAllocated', 8),
('BitsStored', 8),
('HighBit', 7),
('PixelRepresentation', 0),
('PixelData',
b'Data converted to numpy array, raw data removed to preserve memory'),
('shape', (1024, 1024)),
('sampling', (1.0, 1.0))])
im.meta["BitsAllocated"]
8
Acessando o array da imagem
im
Array([[46, 49, 46, ..., 0, 0, 0],
[ 0, 0, 0, ..., 0, 0, 0],
[ 0, 0, 0, ..., 0, 0, 0],
...,
[ 0, 0, 0, ..., 0, 0, 0],
[ 0, 0, 0, ..., 0, 0, 0],
[ 0, 0, 0, ..., 0, 0, 0]], dtype=uint8)
plt.imshow(im, cmap="gray", vmin=0, vmax=255)
<matplotlib.image.AxesImage at 0x7f617cd37940>
