-
Notifications
You must be signed in to change notification settings - Fork 12
/
acr_geometric_accuracy.py
241 lines (190 loc) · 9.38 KB
/
acr_geometric_accuracy.py
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
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
"""
ACR Geometric Accuracy
https://www.acraccreditation.org/-/media/acraccreditation/documents/mri/largephantomguidance.pdf
Calculates geometric accuracy for slices 1 and 5 of the ACR phantom.
This script calculates the horizontal and vertical lengths of the ACR phantom in Slice 1 in accordance with the ACR Guidance.
This script calculates the horizontal, vertical and diagonal lengths of the ACR phantom in Slice 5 in accordance with the ACR Guidance.
The average distance measurement error, maximum distance measurement error and coefficient of variation of all distance
measurements is reported as recommended by IPEM Report 112, "Quality Control and Artefacts in Magnetic Resonance Imaging".
This is done by first producing a binary mask for each respective slice. Line profiles are drawn with aid of rotation
matrices around the centre of the test object to determine each respective length. The results are also visualised.
Created by Yassine Azma
18/11/2022
"""
import sys
import traceback
import os
import numpy as np
import skimage.morphology
import skimage.measure
import skimage.transform
from hazenlib.HazenTask import HazenTask
class ACRGeometricAccuracy(HazenTask):
def __init__(self, **kwargs):
super().__init__(**kwargs)
def run(self) -> dict:
results = {}
z = []
for dcm in self.data:
z.append(dcm.ImagePositionPatient[2])
idx_sort = np.argsort(z)
for dcm in self.data:
if dcm.ImagePositionPatient[2] == z[idx_sort[0]]:
try:
result1 = self.get_geometric_accuracy_slice1(dcm)
except Exception as e:
print(f"Could not calculate the geometric accuracy for {self.key(dcm)} because of : {e}")
traceback.print_exc(file=sys.stdout)
continue
results[self.key(dcm)] = result1
elif dcm.ImagePositionPatient[2] == z[idx_sort[4]]:
try:
result5 = self.get_geometric_accuracy_slice5(dcm)
except Exception as e:
print(f"Could not calculate the geometric accuracy for {self.key(dcm)} because of : {e}")
traceback.print_exc(file=sys.stdout)
continue
results[self.key(dcm)] = result5
results['reports'] = {'images': self.report_files}
L = result1 + result5
mean_err, max_err, cov_l = self.distortion_metric(L)
print(f"Mean relative measurement error is equal to {np.round(mean_err, 2)}mm")
print(f"Maximum absolute measurement error is equal to {np.round(max_err, 2)}mm")
print(f"Coefficient of variation of measurements is equal to {np.round(cov_l, 2)}%")
return results
def centroid_com(self, dcm):
# Calculate centroid of object using a centre-of-mass calculation
thresh_img = dcm > 0.25 * np.max(dcm)
open_img = skimage.morphology.area_opening(thresh_img, area_threshold=500)
bhull = skimage.morphology.convex_hull_image(open_img)
coords = np.nonzero(bhull) # row major - first array is columns
sum_x = np.sum(coords[1])
sum_y = np.sum(coords[0])
cx, cy = sum_x / coords[0].shape[0], sum_y / coords[1].shape[0]
cxy = (round(cx), round(cy))
return bhull, cxy
def horizontal_length(self, res, mask, cxy):
dims = mask.shape
start_h = (cxy[1], 0)
end_h = (cxy[1], dims[0] - 1)
line_profile_h = skimage.measure.profile_line(mask, start_h, end_h, mode='reflect')
extent_h = np.nonzero(line_profile_h)[0]
dist_h = (extent_h[-1] - extent_h[0]) * res[0]
h_dict = {
'Start': start_h,
'End': end_h,
'Extent': extent_h,
'Distance': dist_h
}
return h_dict
def vertical_length(self, res, mask, cxy):
dims = mask.shape
start_v = (0, cxy[0])
end_v = (dims[1] - 1, cxy[0])
line_profile_v = skimage.measure.profile_line(mask, start_v, end_v, mode='reflect')
extent_v = np.nonzero(line_profile_v)[0]
dist_v = (extent_v[-1] - extent_v[0]) * res[1]
v_dict = {
'Start': start_v,
'End': end_v,
'Extent': extent_v,
'Distance': dist_v
}
return v_dict
def rotate_point(self, origin, point, angle):
theta = np.radians(angle)
c, s = np.cos(theta), np.sin(theta)
x_prime = origin[0] + c * (point[0] - origin[0]) - s * (point[1] - origin[1])
y_prime = origin[1] + s * (point[0] - origin[0]) + c * (point[1] - origin[1])
return x_prime, y_prime
def diagonal_lengths(self, res, mask, cxy):
eff_res = np.sqrt(np.mean(np.square(res)))
mask_rotate = skimage.transform.rotate(mask, 45, center=(cxy[0], cxy[1]))
h_dict = self.horizontal_length(res, mask_rotate, cxy)
extent_h = h_dict['Extent']
origin = (cxy[0], cxy[1])
start = (extent_h[0], cxy[1])
end = (extent_h[-1], cxy[1])
se_x_start, se_y_start = self.rotate_point(origin, start, 45)
se_x_end, se_y_end = self.rotate_point(origin, end, 45)
dist_se = np.sqrt(np.sum(np.square([se_x_end - se_x_start, se_y_end - se_y_start]))) * eff_res
se_dict = {
'Start': (se_x_start, se_y_start),
'End': (se_x_end, se_y_end),
'Extent': (se_x_end - se_x_start, se_y_end - se_y_start),
'Distance': dist_se
}
v_dict = self.vertical_length(res, mask_rotate, cxy)
extent_v = v_dict['Extent']
start = (cxy[0], extent_v[0])
end = (cxy[0], extent_v[-1])
sw_x_start, sw_y_start = self.rotate_point(origin, start, 45)
sw_x_end, sw_y_end = self.rotate_point(origin, end, 45)
dist_sw = np.sqrt(np.sum(np.square([sw_x_end - sw_x_start, sw_y_end - sw_y_start]))) * eff_res
sw_dict = {
'Start': (sw_x_start, sw_y_start),
'End': (sw_x_end, sw_y_end),
'Extent': (sw_x_end - sw_x_start, sw_y_end - sw_y_start),
'Distance': dist_sw
}
return sw_dict, se_dict
def get_geometric_accuracy_slice1(self, dcm):
img = dcm.pixel_array
res = dcm.PixelSpacing
mask, cxy = self.centroid_com(img)
h_dict = self.horizontal_length(res, mask, cxy)
v_dict = self.vertical_length(res, mask, cxy)
if self.report:
import matplotlib.pyplot as plt
fig = plt.figure()
fig.set_size_inches(8, 8)
plt.imshow(img)
plt.arrow(h_dict['Extent'][0], cxy[1], h_dict['Extent'][-1] - h_dict['Extent'][0], 1, color='blue',
length_includes_head=True, head_width=5)
plt.arrow(cxy[0], v_dict['Extent'][0], 1, v_dict['Extent'][-1] - v_dict['Extent'][0], color='orange',
length_includes_head=True, head_width=5)
plt.legend([str(np.round(h_dict['Distance'], 2)) + 'mm',
str(np.round(v_dict['Distance'], 2)) + 'mm'])
plt.axis('off')
plt.title('Geometric Accuracy for Slice 1')
img_path = os.path.realpath(os.path.join(self.report_path, f'{self.key(dcm)}.png'))
fig.savefig(img_path)
self.report_files.append(img_path)
return h_dict['Distance'], v_dict['Distance']
def get_geometric_accuracy_slice5(self, dcm):
img = dcm.pixel_array
res = dcm.PixelSpacing
mask, cxy = self.centroid_com(img)
h_dict = self.horizontal_length(res, mask, cxy)
v_dict = self.vertical_length(res, mask, cxy)
sw_dict, se_dict = self.diagonal_lengths(res, mask, cxy)
if self.report:
import matplotlib.pyplot as plt
fig = plt.figure()
fig.set_size_inches(8, 8)
plt.imshow(img)
plt.arrow(h_dict['Extent'][0], cxy[1], h_dict['Extent'][-1] - h_dict['Extent'][0], 1, color='blue',
length_includes_head=True, head_width=5)
plt.arrow(cxy[0], v_dict['Extent'][0], 1, v_dict['Extent'][-1] - v_dict['Extent'][0], color='orange',
length_includes_head=True, head_width=5)
plt.arrow(se_dict['Start'][0], se_dict['Start'][1], se_dict['Extent'][0], se_dict['Extent'][1],
color='purple', length_includes_head=True, head_width=5)
plt.arrow(sw_dict['Start'][0], sw_dict['Start'][1], sw_dict['Extent'][0], sw_dict['Extent'][1],
color='yellow', length_includes_head=True, head_width=5)
plt.legend([str(np.round(h_dict['Distance'], 2)) + 'mm',
str(np.round(v_dict['Distance'], 2)) + 'mm',
str(np.round(sw_dict['Distance'], 2)) + 'mm',
str(np.round(se_dict['Distance'], 2)) + 'mm'])
plt.axis('off')
plt.title('Geometric Accuracy for Slice 5')
img_path = os.path.realpath(os.path.join(self.report_path, f'{self.key(dcm)}.png'))
fig.savefig(img_path)
self.report_files.append(img_path)
return h_dict['Distance'], v_dict['Distance'], sw_dict['Distance'], se_dict['Distance']
def distortion_metric(self, L):
err = [x - 190 for x in L]
mean_err = np.mean(err)
max_err = np.max(np.absolute(err))
cov_l = 100 * np.std(L) / np.mean(L)
return mean_err, max_err, cov_l