'DICOM-医疗数位影像传输协定'

Posted on

最近有一个医疗图像的比赛，医疗图像以前没有接触过，对于这种格式的数据不甚了解，做了一些调研，记录一下。

定义

医疗数位影像传输协定（DICOM，Digital Imaging and Communications in Medicine）是一组通用的标准协定，在对于医学影像的处理、储存、打印、传输上。它包含了档案格式的定义及网络通信协定。DICOM是以TCP/IP为基础的应用协定，并以TCP/IP联系各个系统。两个能接受DICOM格式的医疗仪器间，可借由DICOM格式的档案，来接收与交换影像及病人资料。—— 维基百科

简言之，DICOM就是一种医疗行业使用的一种数据存储传输标准，广泛应用于各种医疗设备，如:CT、核磁共振、超声等。

内容

协议包含编码格式，图像格式、仪器厂商、仪器序列号、数据生成时间、数据采集方式、病人性别、病人年龄等等。下面是使用PYDICOM读取数据打印的结果。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
(0008, 0005) Specific Character Set              CS: 'ISO_IR 100'
(0008, 0008) Image Type                          CS: ['ORIGINAL', 'PRIMARY', 'AXIAL']
(0008, 0012) Instance Creation Date              DA: '20170728'
(0008, 0013) Instance Creation Time              TM: '160113.752'
(0008, 0016) SOP Class UID                       UI: CT Image Storage
(0008, 0018) SOP Instance UID                    UI: *
(0008, 0020) Study Date                          DA: '20170728'
(0008, 0021) Series Date                         DA: '20170728'
(0008, 0022) Acquisition Date                    DA: '20170728'
(0008, 0023) Content Date                        DA: '20170728'
(0008, 002a) Acquisition DateTime                DT: '20170728155419.080'
(0008, 0030) Study Time                          TM: '155226.560'
(0008, 0031) Series Time                         TM: '160052.584'
(0008, 0032) Acquisition Time                    TM: '155419'
(0008, 0033) Content Time                        TM: '155419.410'
(0008, 0050) Accession Number                    SH: '*'
(0008, 0060) Modality                            CS: 'CT'
(0008, 0070) Manufacturer                        LO: 'Philips'
(0008, 0080) Institution Name                    LO: '*'
(0008, 0081) Institution Address                 ST: '*'
(0008, 0090) Referring Physician's Name          PN: '*'
(0008, 1010) Station Name                        SH: '*'
(0008, 1030) Study Description                   LO: 'CT158424'
(0008, 103e) Series Description                  LO: 'iDose (2)'
(0008, 1040) Institutional Department Name       LO: '*'
(0008, 1080) Admitting Diagnoses Description     LO: ''
(0008, 1084)  Admitting Diagnoses Code Sequence   0 item(s) ---- 
(0008, 1090) Manufacturer's Model Name           LO: '*'
(0008, 1111)  Referenced Performed Procedure Step Sequence   1 item(s) ---- 
(0008, 1150) Referenced SOP Class UID            UI: Modality Performed Procedure Step SOP Class
(0008, 1155) Referenced SOP Instance UID         UI:1.3.46.670589.33.1.63636853943535791300001.5748714835418010494
   ---------
(0008, 1140)  Referenced Image Sequence   1 item(s) ---- 
(0008, 1150) Referenced SOP Class UID            UI: CT Image Storage
(0008, 1155) Referenced SOP Instance UID         UI: 1.3.46.670589.33.1.63636853996983848400001.4836402666508773100
   ---------
(0008, 3010) Irradiation Event UID               UI: 1.3.46.670589.33.1.63636854035098028400003.5509738905066048408
(0010, 0010) Patient's Name                      PN: '*'
(0010, 0020) Patient ID                          LO: 'a8d5f610-bd93-4446-9680-1eeb2d6e606d'
(0010, 0030) Patient's Birth Date                DA: '*'
(0010, 0040) Patient's Sex                       CS: 'M'
(0010, 1010) Patient's Age                       AS: '050Y'
(0018, 0015) Body Part Examined                  CS: 'ABDOMEN'
(0018, 0022) Scan Options                        CS: 'HELIX'
(0018, 0050) Slice Thickness                     DS: "5"
(0018, 0060) KVP                                 DS: "120"
(0018, 0088) Spacing Between Slices              DS: "5"
(0018, 0090) Data Collection Diameter            DS: "500"
(0018, 1000) Device Serial Number                LO: '*'
(0018, 1020) Software Version(s)                 LO: '4.1'
(0018, 1030) Protocol Name                       LO: 'Liver C-/CTA/C+ MC /Abdomen'
(0018, 1100) Reconstruction Diameter             DS: "338"
(0018, 1110) Distance Source to Detector         DS: "1040"
(0018, 1111) Distance Source to Patient          DS: "570"
(0018, 1120) Gantry/Detector Tilt                DS: "0"
(0018, 1130) Table Height                        DS: "119.1"
(0018, 1150) Exposure Time                       IS: "470"
(0018, 1151) X-Ray Tube Current                  IS: "534"
(0018, 1152) Exposure                            IS: "251"
(0018, 1160) Filter Type                         SH: 'B'
(0018, 1210) Convolution Kernel                  SH: 'B'
(0018, 5100) Patient Position                    CS: 'HFS'
(0018, 9302) Acquisition Type                    CS: 'SPIRAL'
(0018, 9305) Revolution Time                     FD: 0.4
(0018, 9306) Single Collimation Width            FD: 0.625
(0018, 9307) Total Collimation Width             FD: 80.0
(0018, 9309) Table Speed                         FD: 170.4
(0018, 9310) Table Feed per Rotation             FD: 109.056
(0018, 9311) Spiral Pitch Factor                 FD: 0.852
(0018, 9323) Exposure Modulation Type            CS: '3D MODULATION'
(0018, 9324) Estimated Dose Saving               FD: -3.0
(0018, 9345) CTDIvol                             FD: 17.04320987654321
(0020, 000d) Study Instance UID                  UI: adce2399-4540-41b7-8018-ee81ea48655e
(0020, 000e) Series Instance UID                 UI: 2d5c02e9-4a75-4e67-959d-3d705479639c
(0020, 0010) Study ID                            SH: '*'
(0020, 0011) Series Number                       IS: "201"
(0020, 0012) Acquisition Number                  IS: "1"
(0020, 0013) Instance Number                     IS: "1"
(0020, 0032) Image Position (Patient)            DS: ['-161.902', '-33.1', '106.18']
(0020, 0037) Image Orientation (Patient)         DS: ['1', '0', '0', '0', '1', '0']
(0020, 0052) Frame of Reference UID              UI: 1.3.46.670589.33.1.63636853961048793000002.5747101679789372786
(0020, 1040) Position Reference Indicator        LO: ''
(0020, 1041) Slice Location                      DS: "106.18"
(0020, 4000) Image Comments                      LT: ''
(0028, 0002) Samples per Pixel                   US: 1
(0028, 0004) Photometric Interpretation          CS: 'MONOCHROME2'
(0028, 0010) Rows                                US: 512
(0028, 0011) Columns                             US: 512
(0028, 0030) Pixel Spacing                       DS: ['0.66015625', '0.66015625']
(0028, 0100) Bits Allocated                      US: 16
(0028, 0101) Bits Stored                         US: 12
(0028, 0102) High Bit                            US: 11
(0028, 0103) Pixel Representation                US: 0
(0028, 1050) Window Center                       DS: ['50', '50']
(0028, 1051) Window Width                        DS: ['220', '220']
(0028, 1052) Rescale Intercept                   DS: "-1024"
(0028, 1053) Rescale Slope                       DS: "1"
(0040, 1001) Requested Procedure ID              SH: 'CT158424'
(00e1, 0010) Private Creator                     LO: 'ELSCINT1'
(00e1, 1002) Private tag data                    SH: 'Adult'
(00e1, 1036) Private tag data                    CS: 'YES'
(00e1, 1040) [Image Label]                       SH: ''
(00e1, 1046) Private tag data                    OB: Array of 512 bytes
(00e1, 1050) [Acquisition Duration]              DS: "2.053"
(00e1, 1063) [Patient Language]                  SH: 'CHINESE'
(00e1, 10c4) Private tag data                    DS: "1059.67504882813"
(01f1, 0010) Private Creator                     LO: 'ELSCINT1'
(01f1, 1001) [Acquisition Type]                  CS: 'SPIRAL'
(01f1, 1002) [Unknown]                           CS: 'STANDARD'
(01f1, 1004) [Angular Sampling Density]          CS: 'NORMAL'
(01f1, 1008) [Acquisition Length]                DS: "215"
(01f1, 100c) [Scanner Relative Center]           DS: ['0', '-0.021']
(01f1, 100e) [Unknown]                           FL: 0.0
(01f1, 100f) Private tag data                    CS: 'IN'
(01f1, 1026) [Pitch]                             DS: "0.852"
(01f1, 1027) [Rotation Time]                     DS: "0.4"
(01f1, 1032) [Image View Convention]             CS: 'RIGHT_ON_LEFT'
(01f1, 1044) [Unknown]                           OW: Array of 644 bytes
(01f1, 1046) [Unknown]                           FL: 0.625
(01f1, 1047) [Unknown]                           SH: '3D'
(01f1, 1049) [Unknown]                           DS: ''
(01f1, 104a) [Unknown]                           SH: 'DOM'
(01f1, 104b) [Unknown]                           SH: '128x0.625'
(01f1, 104d) [Unknown]                           SH: 'YES'
(01f1, 104e) [Unknown]                           SH: 'Abdomen/Pelvis'
(01f1, 1054) Private tag data                    IS: "22"
(01f1, 1056) Private tag data                    LO: '28.204652423757'
(01f7, 0010) Private Creator                     LO: 'ELSCINT1'
(01f7, 1014) [Unknown]                           OW: Array of 108 bytes
(01f7, 1019) [Unknown]                           OW: Array of 2160 bytes
(01f7, 1022) [Unknown]                           UI: 1.3.46.670589.33.1.63636854035098028400003.5509738905066048408
(01f7, 1029) [Unknown]                           OW: Array of 440 bytes
(01f7, 1031) [Unknown]                           OW: Array of 124 bytes
(01f7, 1092) Private tag data                    OW: Array of 890 bytes
(01f7, 1095) Private tag data                    OW: Array of 128 bytes
(01f7, 1097) Private tag data                    OW: Array of 256 bytes
(01f7, 1099) Private tag data                    OW: Array of 64 bytes
(01f7, 109b) Private tag data                    IS: "2"
(07a1, 0010) Private Creator                     LO: 'ELSCINT1'
(7fe0, 0010) Pixel Data                          OW: Array of 524288 bytes

具体内容并不关心，只看看那些可能会用到，比如病人性别、年龄。

读取显示

使用pydicom

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
#-*-coding:utf-8-*-
import cv2
import numpy
import pydicom as dicom
from matplotlib import pyplot as plt

data_dir = 'demo'

dcm = dicom.read_file("demo/1.dcm")
dcm.image = dcm.pixel_array * dcm.RescaleSlope + dcm.RescaleIntercept

slices = []
slices.append(dcm)
img = slices[ int(len(slices)/2) ].image.copy()
ret,img = cv2.threshold(img, 90,3071, cv2.THRESH_BINARY)
img = numpy.uint8(img)

im2, contours, _ = cv2.findContours(img,cv2.RETR_LIST,cv2.CHAIN_APPROX_SIMPLE)
mask = numpy.zeros(img.shape, numpy.uint8)
for contour in contours:
    cv2.fillPoly(mask, [contour], 255)
img[(mask > 0)] = 255

kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE,(2,2))
img = cv2.morphologyEx(img, cv2.MORPH_OPEN, kernel)

img2 = slices[ int(len(slices)/2) ].image.copy()
img2[(img == 0)] = -2000

plt.figure(figsize=(12, 12))
plt.subplot(131)
plt.imshow(slices[int(len(slices) / 2)].image, 'gray')
plt.title('Original')
plt.subplot(132)
plt.imshow(img, 'gray')
plt.title('Mask')
plt.subplot(133)
plt.imshow(img2, 'gray')
plt.title('Result')
plt.show()

kZ63bF.png

参考-1
参考-2