def create_voxel_plots(voxel_name, ppmst):
print '...Working on %s %s ' %(analysis_type, ppmst)
#create output QC directory
mkdir_path(os.path.join(subject_dir, 'quality_control', analysis_type, voxel_name, 'ppm_%s' %ppmst, 'tmp'))
qc_dir = os.path.join(subject_dir, 'quality_control', analysis_type, voxel_name, 'ppm_%s' %ppmst)
tmp_dir = os.path.join(qc_dir, 'tmp')
#grab lcmodel plots
svs_lcmodel = os.path.join(subject_dir, 'lcmodel_%s'%analysis_type, voxel_name, 'ppm_%s'%ppmst, 'ps.pdf')
# create localization pngs
make_png = ['convert', '-density', '300', '-trim', '%s'%svs_lcmodel, '-quality', '300', '-sharpen', '0x1.0', '%s/%s_lcmodel.png'%(tmp_dir, voxel_name)]
subprocess.call(make_png)
lcm_plot = os.path.join(tmp_dir, '%s_lcmodel-0.png'%voxel_name)
#grab snr/fwhm data
svs_snr = np.genfromtxt(os.path.join(subject_dir, 'lcmodel_%s'%analysis_type, voxel_name, 'ppm_%s'%ppmst, 'snr.txt'), delimiter = ',')
#grab voxel mask
svs = os.path.join(subject_dir, 'svs_voxel_mask','%s%s_%s_RDA_MASK.nii'%(subject,workspace_dir[-10:-9], voxel_name))
#get data into matrix
anat_load = nb.load(anatomical)
svs_load = nb.load(svs)
anat_data = anat_load.get_data()
svs_data = svs_load.get_data()
# get svs cut coords
coords = find_cut_coords(svs_load)
# convert zeros to nans for visualization purposes
svs_data[svs_data==0]=np.nan
# plot voxel on anat
fig =plt.figure()
fig.set_size_inches(6.5, 6.5)
fig.subplots_adjust(wspace=0.005)
#1
ax1 = plt.subplot2grid((1,3), (0,0), colspan = 1, rowspan =1)
ax1.imshow(anat_data[coords[0],:,:], matplotlib.cm.bone_r)
ax1.imshow(svs_data[coords[0],:,:] , matplotlib.cm.rainbow_r, alpha = 0.7)
ax1.set_xlim(23, 157)
ax1.set_ylim(101, 230)
ax1.axes.get_yaxis().set_visible(False)
ax1.axes.get_xaxis().set_visible(False)
#2
ax2 = plt.subplot2grid((1,3), (0,1), colspan = 1, rowspan =1)
ax2.imshow(np.rot90(anat_data[:,:,coords[2]]), matplotlib.cm.bone_r )
ax2.imshow(np.rot90(svs_data[:,:,coords[2]]) , matplotlib.cm.rainbow_r, alpha = 0.7 )
ax2.set_xlim(230, 20)
ax2.set_ylim(207, 4)
ax2.axes.get_yaxis().set_visible(False)
ax2.axes.get_xaxis().set_visible(False)
#3
ax3 = plt.subplot2grid((1,3), (0,2), colspan = 1, rowspan =1)
ax3.imshow(anat_data[:,coords[1],:], matplotlib.cm.bone_r, origin='lower')
ax3.imshow(svs_data[:,coords[1],:] , matplotlib.cm.rainbow_r, alpha = 0.7, origin='lower')
ax3.set_xlim(38, 140)
ax3.set_ylim(160, 60)
ax3.axes.get_yaxis().set_visible(False)
ax3.axes.get_xaxis().set_visible(False)
fig.tight_layout()
fig.savefig('%s/localization_%s.png'%(qc_dir, voxel_name), dpi=200, bbox_inches='tight')
# create qc report
report = canvas.Canvas(os.path.join(qc_dir,'QC_REPORT_%s.pdf'%voxel_name), pagesize=(1280, 1556))
report.setFont("Helvetica", 40)
report.drawImage(os.path.join(qc_dir,'localization_%s.png'%voxel_name), 1, inch*13.5)
report.drawImage(lcm_plot, 30, inch*1, width = 1200, height = 800)
report.drawString(230, inch*20, ' %s%s, %s_%s , SNR=%s FWHM=%s ' %(subject,workspace_dir[-10:-9], voxel_name, analysis_type, svs_snr[2],svs_snr[1]) )
report.showPage()
if analysis_type is 'twix':
fig_f6 = os.path.join(subject_dir, 'svs_twix', voxel_name, 'f6_frequency_drift_correction.png')
fig_f7 = os.path.join(subject_dir, 'svs_twix', voxel_name, 'f7_estimated_phase_phase_drift.png')
fig_f8 = os.path.join(subject_dir, 'svs_twix', voxel_name, 'f8_estimated_frequency_drift.png')
reader = open(os.path.join(subject_dir, 'svs_twix', voxel_name, voxel_name, 'readme.txt'), 'r')
for line in reader:
if 'bad' in line:
badavn = line[34:38]
report.drawImage(fig_f6, 1, inch*7.2)
report.drawImage(fig_f7, 90, inch*1, width = 540, height = 450)
report.drawImage(fig_f8, 590, inch*1, width = 540, height = 450)
report.setFont("Helvetica", 40)
report.drawString(350, inch*20, 'Number of Bad Averages =%s' %(badavn))
report.save()
else:
report.save()
def plot_svs(svs_a, svs_b, anatomical, fname):
import os
import numpy as np
import nibabel as nb
import matplotlib
import matplotlib.pyplot as plt
import seaborn as sns
from matplotlib import colors
#get data into matrix
anat_load = nb.load(anatomical)
anat_data = anat_load.get_data()
svs_load_a = nb.load(svs_a)
svs_data_a = svs_load_a.get_data().astype(float)
svs_load_b = nb.load(svs_b)
svs_data_b = svs_load_b.get_data().astype(float)
# get svs cut coords
coords = find_cut_coords(svs_load_b)
# convert zeros to nans for visualization purposes
svs_data_a[svs_data_a==0]=np.nan
svs_data_b[svs_data_b==0]=np.nan
svs_x = svs_data_a * svs_data_b
# plot voxel on anat
fig =plt.figure()
fig.set_size_inches(15, 15)
fig.subplots_adjust(wspace=0.005)
red = colors.ListedColormap(['red'])
blue = colors.ListedColormap(['blue'])
mix = colors.ListedColormap(['purple'])
#1
ax1 = plt.subplot2grid((1,3), (0,0), colspan = 1, rowspan =1)
ax1.imshow(anat_data[coords[0],:,:], matplotlib.cm.bone_r)
ax1.imshow(svs_data_b[coords[0],:,:] , red, alpha = 0.7)
ax1.imshow(svs_data_a[coords[0],:,:] , blue, alpha = 0.7)
ax1.imshow(svs_x[coords[0],:,:] , mix , alpha = 1, interpolation='nearest')
ax1.set_xlim(23, 157)
ax1.set_ylim(101, 230)
ax1.axes.get_yaxis().set_visible(False)
ax1.axes.get_xaxis().set_visible(False)
ax1.axes.get_xaxis().set_visible(False)
#2
ax2 = plt.subplot2grid((1,3), (0,1), colspan = 1, rowspan =1)
ax2.imshow(np.rot90(anat_data[:,:,coords[2]]), matplotlib.cm.bone_r )
ax2.imshow(np.rot90(svs_data_b[:,:,coords[2]]) , red, alpha = 0.7 )
ax2.imshow(np.rot90(svs_data_a[:,:,coords[2]]) , blue, alpha = 0.7 )
ax2.imshow(np.rot90(svs_x[:,:,coords[2]]) , mix, alpha = 0.7 )
ax2.set_xlim(230, 20)
ax2.set_ylim(207, 4)
ax2.axes.get_yaxis().set_visible(False)
ax2.axes.get_xaxis().set_visible(False)
#3
ax3 = plt.subplot2grid((1,3), (0,2), colspan = 1, rowspan =1)
ax3.imshow(anat_data[:,coords[1],:], matplotlib.cm.bone_r, origin='lower')
ax3.imshow(svs_data_b[:,coords[1],:] , red, alpha = 0.7, origin='lower')
ax3.imshow(svs_data_a[:,coords[1],:] , blue, alpha = 0.7, origin='lower')
ax3.imshow(svs_x[:,coords[1],:] , mix, alpha = 0.7, origin='lower')
ax3.set_xlim(37, 140)
ax3.set_ylim(160, 61)
ax3.axes.get_yaxis().set_visible(False)
ax3.axes.get_xaxis().set_visible(False)
for ax in [ax1, ax2, ax3]:
for axis in ['top','bottom','left','right']:
ax.spines[axis].set_color('black')
ax.spines[axis].set_linewidth(3)
fig.tight_layout()
fig.savefig('plot_overlap_%s.png'%fname, dpi=200, bbox_inches='tight')
def create_voxel_plots(voxel_name, ppmst):
print '...Working on %s %s ' % (analysis_type, ppmst)
#create output QC directory
mkdir_path(
os.path.join(subject_dir, 'quality_control', analysis_type,
voxel_name, 'ppm_%s' % ppmst, 'tmp'))
qc_dir = os.path.join(subject_dir, 'quality_control',
analysis_type, voxel_name, 'ppm_%s' % ppmst)
tmp_dir = os.path.join(qc_dir, 'tmp')
#grab lcmodel plots
svs_lcmodel = os.path.join(subject_dir,
'lcmodel_%s' % analysis_type,
voxel_name, 'ppm_%s' % ppmst, 'ps.pdf')
# create localization pngs
make_png = [
'convert', '-density', '300', '-trim',
'%s' % svs_lcmodel, '-quality', '300', '-sharpen', '0x1.0',
'%s/%s_lcmodel.png' % (tmp_dir, voxel_name)
]
subprocess.call(make_png)
lcm_plot = os.path.join(tmp_dir, '%s_lcmodel-0.png' % voxel_name)
#grab snr/fwhm data
svs_snr = np.genfromtxt(os.path.join(subject_dir,
'lcmodel_%s' % analysis_type,
voxel_name, 'ppm_%s' % ppmst,
'snr.txt'),
delimiter=',')
#grab voxel mask
svs = os.path.join(
subject_dir, 'svs_voxel_mask', '%s%s_%s_RDA_MASK.nii' %
(subject, workspace_dir[-10:-9], voxel_name))
#get data into matrix
anat_load = nb.load(anatomical)
svs_load = nb.load(svs)
anat_data = anat_load.get_data()
svs_data = svs_load.get_data()
# get svs cut coords
coords = find_cut_coords(svs_load)
# convert zeros to nans for visualization purposes
svs_data[svs_data == 0] = np.nan
# plot voxel on anat
fig = plt.figure()
fig.set_size_inches(6.5, 6.5)
fig.subplots_adjust(wspace=0.005)
#1
ax1 = plt.subplot2grid((1, 3), (0, 0), colspan=1, rowspan=1)
ax1.imshow(anat_data[coords[0], :, :], matplotlib.cm.bone_r)
ax1.imshow(svs_data[coords[0], :, :],
matplotlib.cm.rainbow_r,
alpha=0.7)
ax1.set_xlim(23, 157)
ax1.set_ylim(101, 230)
ax1.axes.get_yaxis().set_visible(False)
ax1.axes.get_xaxis().set_visible(False)
#2
ax2 = plt.subplot2grid((1, 3), (0, 1), colspan=1, rowspan=1)
ax2.imshow(np.rot90(anat_data[:, :, coords[2]]),
matplotlib.cm.bone_r)
ax2.imshow(np.rot90(svs_data[:, :, coords[2]]),
matplotlib.cm.rainbow_r,
alpha=0.7)
ax2.set_xlim(230, 20)
ax2.set_ylim(207, 4)
ax2.axes.get_yaxis().set_visible(False)
ax2.axes.get_xaxis().set_visible(False)
#3
ax3 = plt.subplot2grid((1, 3), (0, 2), colspan=1, rowspan=1)
ax3.imshow(anat_data[:, coords[1], :],
matplotlib.cm.bone_r,
origin='lower')
ax3.imshow(svs_data[:, coords[1], :],
matplotlib.cm.rainbow_r,
alpha=0.7,
origin='lower')
ax3.set_xlim(38, 140)
ax3.set_ylim(160, 60)
ax3.axes.get_yaxis().set_visible(False)
ax3.axes.get_xaxis().set_visible(False)
fig.tight_layout()
fig.savefig('%s/localization_%s.png' % (qc_dir, voxel_name),
dpi=200,
bbox_inches='tight')
# create qc report
report = canvas.Canvas(os.path.join(
qc_dir, 'QC_REPORT_%s.pdf' % voxel_name),
pagesize=(1280, 1556))
report.setFont("Helvetica", 40)
report.drawImage(
os.path.join(qc_dir, 'localization_%s.png' % voxel_name), 1,
inch * 13.5)
report.drawImage(lcm_plot, 30, inch * 1, width=1200, height=800)
report.drawString(
230, inch * 20, ' %s%s, %s_%s , SNR=%s FWHM=%s ' %
(subject, workspace_dir[-10:-9], voxel_name, analysis_type,
svs_snr[2], svs_snr[1]))
report.showPage()
if analysis_type is 'twix':
fig_f6 = os.path.join(subject_dir, 'svs_twix', voxel_name,
'f6_frequency_drift_correction.png')
fig_f7 = os.path.join(subject_dir, 'svs_twix', voxel_name,
'f7_estimated_phase_phase_drift.png')
fig_f8 = os.path.join(subject_dir, 'svs_twix', voxel_name,
'f8_estimated_frequency_drift.png')
reader = open(
os.path.join(subject_dir, 'svs_twix', voxel_name,
voxel_name, 'readme.txt'), 'r')
for line in reader:
if 'bad' in line:
badavn = line[34:38]
report.drawImage(fig_f6, 1, inch * 7.2)
report.drawImage(fig_f7, 90, inch * 1, width=540, height=450)
report.drawImage(fig_f8, 590, inch * 1, width=540, height=450)
report.setFont("Helvetica", 40)
report.drawString(350, inch * 20,
'Number of Bad Averages =%s' % (badavn))
report.save()
else:
report.save()
def plot_svs(svs_a, svs_b, anatomical, fname):
import os
import numpy as np
import nibabel as nb
import matplotlib
import matplotlib.pyplot as plt
import seaborn as sns
from matplotlib import colors
#get data into matrix
anat_load = nb.load(anatomical)
anat_data = anat_load.get_data()
svs_load_a = nb.load(svs_a)
svs_data_a = svs_load_a.get_data().astype(float)
svs_load_b = nb.load(svs_b)
svs_data_b = svs_load_b.get_data().astype(float)
# get svs cut coords
coords = find_cut_coords(svs_load_b)
# convert zeros to nans for visualization purposes
svs_data_a[svs_data_a == 0] = np.nan
svs_data_b[svs_data_b == 0] = np.nan
svs_x = svs_data_a * svs_data_b
# plot voxel on anat
fig = plt.figure()
fig.set_size_inches(15, 15)
fig.subplots_adjust(wspace=0.005)
red = colors.ListedColormap(['red'])
blue = colors.ListedColormap(['blue'])
mix = colors.ListedColormap(['purple'])
#1
ax1 = plt.subplot2grid((1, 3), (0, 0), colspan=1, rowspan=1)
ax1.imshow(anat_data[coords[0], :, :], matplotlib.cm.bone_r)
ax1.imshow(svs_data_b[coords[0], :, :], red, alpha=0.7)
ax1.imshow(svs_data_a[coords[0], :, :], blue, alpha=0.7)
ax1.imshow(svs_x[coords[0], :, :],
mix,
alpha=1,
interpolation='nearest')
ax1.set_xlim(23, 157)
ax1.set_ylim(101, 230)
ax1.axes.get_yaxis().set_visible(False)
ax1.axes.get_xaxis().set_visible(False)
ax1.axes.get_xaxis().set_visible(False)
#2
ax2 = plt.subplot2grid((1, 3), (0, 1), colspan=1, rowspan=1)
ax2.imshow(np.rot90(anat_data[:, :, coords[2]]),
matplotlib.cm.bone_r)
ax2.imshow(np.rot90(svs_data_b[:, :, coords[2]]), red, alpha=0.7)
ax2.imshow(np.rot90(svs_data_a[:, :, coords[2]]), blue, alpha=0.7)
ax2.imshow(np.rot90(svs_x[:, :, coords[2]]), mix, alpha=0.7)
ax2.set_xlim(230, 20)
ax2.set_ylim(207, 4)
ax2.axes.get_yaxis().set_visible(False)
ax2.axes.get_xaxis().set_visible(False)
#3
ax3 = plt.subplot2grid((1, 3), (0, 2), colspan=1, rowspan=1)
ax3.imshow(anat_data[:, coords[1], :],
matplotlib.cm.bone_r,
origin='lower')
ax3.imshow(svs_data_b[:, coords[1], :],
red,
alpha=0.7,
origin='lower')
ax3.imshow(svs_data_a[:, coords[1], :],
blue,
alpha=0.7,
origin='lower')
ax3.imshow(svs_x[:, coords[1], :], mix, alpha=0.7, origin='lower')
ax3.set_xlim(37, 140)
ax3.set_ylim(160, 61)
ax3.axes.get_yaxis().set_visible(False)
ax3.axes.get_xaxis().set_visible(False)
for ax in [ax1, ax2, ax3]:
for axis in ['top', 'bottom', 'left', 'right']:
ax.spines[axis].set_color('black')
ax.spines[axis].set_linewidth(3)
fig.tight_layout()
fig.savefig('plot_overlap_%s.png' % fname,
dpi=200,
bbox_inches='tight')
