-
Notifications
You must be signed in to change notification settings - Fork 0
/
main.py
201 lines (135 loc) · 4.17 KB
/
main.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
#matplotlib notebook
import numpy as np
import sigpy as sp
import sigpy.mri as mr
import sigpy.plot as pl
import cupy as cp
import time
import matplotlib.pyplot as plt
import numpy.ma as ma
import pathlib
import os
import sys
from sys import getsizeof
use_gpu=False
# load data after coil compression
#
ksp = np.load('/home/valery/DICOM/mp2rage_data.npy')
# remove
ksp=np.squeeze(ksp)
# tranpose from [RO E1 E2 CHA] to [CHA E2 E1 RO]
# [CHA E2 E1 RO] = (4, 240, 320, 320)
ksp=np.transpose(ksp, (3, 2 , 1, 0))
# remove coils to satisfy sigpy and my laptop
ksp=ksp[0:4,:,:,:]*100
print(sys.getsizeof(ksp))
print(np.shape(ksp))
# check kspace sampling
mask=ma.masked_greater(np.abs(ksp), 0)
#display sampling
#plt.figure(1)
#lala=ksp[0,:,:,160]
#print(np.shape(lala))
#plt.imshow(np.abs(mask[0,:,:,160]))
#plt.show()
# get image dimension
img_shape = ksp.shape[1:]
# host to device
if (use_gpu==True):
ksp_on_gpu0 = sp.to_device(ksp, 0)
# compute ifft on host
tstart = time.time()
F = sp.linop.FFT(ksp.shape, axes=(-1, -2, -3))
I=F.H * ksp
print("Ifft3D estimation duration numpy: {}s".format(time.time() - tstart))
if (use_gpu==True):
try:
# compute ifft on device
tstart = time.time()
F_gpu = sp.linop.FFT(ksp_on_gpu0.shape, axes=(-1, -2, -3))
I_gpu=F_gpu.H * ksp_on_gpu0
print("Ifft3D estimation duration cupy: {}s".format(time.time() - tstart))
# display result
#pl.ImagePlot(I_gpu, z=0, title=r'$F^H y$')
except:
print("Ifft3D could not be computed on device")
# compute csm on host
tstart = time.time()
#mps = mr.app.EspiritCalib(ksp).run()
mps=sp.mri.app.JsenseRecon(ksp).run()
print("EspiritCalib estimation duration numpy: {}s".format(time.time() - tstart))
# display result
#pl.ImagePlot(mps, z=0, title=r'$F^H y$')
if (use_gpu==True):
try:
# compute ifft on device
tstart = time.time()
mps_on_gpu = mr.app.EspiritCalib(ksp_on_gpu0).run()
print("EspiritCalib estimation duration cupy: {}s".format(time.time() - tstart))
# display result
#pl.ImagePlot(mps_on_gpu, z=0, title=r'$F^H y$')
except:
print("EspiritCalib could not be computed on device")
###########################################################################
print("Define S")
S = sp.linop.Multiply(img_shape, mps)
#pl.ImagePlot(S.H * F.H * ksp, title=r'$S^H F^H y$')
mask = np.sum(abs(ksp), axis=0) > 0
#pl.ImagePlot(mask, title='Sampling Mask')
P = sp.linop.Multiply(ksp.shape, mask)
## W Linop
print("Define W Linop")
W = sp.linop.Wavelet(img_shape)
wav = W * S.H * F.H * ksp
#pl.ImagePlot(wav**0.1, title=r'$W S^H F^H y$')
print(np.amax(np.abs(wav)))
print(np.amin(np.abs(wav)))
print(np.shape(wav))
plt.figure(1)
lala=ksp[0,:,:,160]
print(np.shape(lala))
plt.imshow(np.abs(wav[:,:,160]))
plt.clim(0.0001,0.001)
plt.show()
pl.ImagePlot(wav, title=r'$W S^H F^H y$')
A = P * F * S * W.H
## Prox
print("Define Prox")
lamda = 0.005
proxg = sp.prox.L1Reg(wav.shape, lamda)
alpha = 1
wav_thresh = proxg(alpha, wav)
pl.ImagePlot(wav_thresh**0.1)
## Alg
print("Define Alg")
max_iter = 30
alpha = 1
def gradf(x):
return A.H * (A * x - ksp)
wav_hat = np.zeros(wav.shape, np.complex)
alg = sp.alg.GradientMethod(gradf, wav_hat, alpha, proxg=proxg, max_iter=max_iter)
while not alg.done():
alg.update()
print('\rL1WaveletRecon, Iteration={}'.format(alg.iter), end='')
pl.ImagePlot(W.H(wav_hat))
## App
class L1WaveletRecon(sp.app.App):
def __init__(self, ksp, mask, mps, lamda, max_iter):
img_shape = mps.shape[1:]
S = sp.linop.Multiply(img_shape, mps)
F = sp.linop.FFT(ksp.shape, axes=(-1, -2))
P = sp.linop.Multiply(ksp.shape, mask)
self.W = sp.linop.Wavelet(img_shape)
A = P * F * S * self.W.H
proxg = sp.prox.L1Reg(A.ishape, lamda)
self.wav = np.zeros(A.ishape, np.complex)
alpha = 1
def gradf(x):
return A.H * (A * x - ksp)
alg = sp.alg.GradientMethod(gradf, self.wav, alpha, proxg=proxg,
max_iter=max_iter)
super().__init__(alg)
def _output(self):
return self.W.H(self.wav)
img = L1WaveletRecon(ksp, mask, mps, lamda, max_iter).run()
pl.ImagePlot(img)