def polar_plot(polar_plot_name, cart_patch, cart_label, polarbd, cart_seg,
polar_patch, polar_label, cart_unet, cart_mask):
fz = 20
fig = plt.figure(figsize=(20, 8))
plt.subplot(2, 5, 1)
plt.title('Cartesian Patch', fontsize=fz)
plt.imshow(cart_patch, cmap='gray')
plt.subplot(2, 5, 2)
plt.title('Cartesian Label', fontsize=fz)
plt.imshow(cart_label, cmap='gray')
plt.subplot(2, 5, 3)
plt.title('Prediction \nDSC:%.3f' % (DSC(cart_seg, cart_label)),
fontsize=fz)
plt.imshow(diffmap(cart_seg, cart_label), cmap='gray')
if cart_unet is not None:
plt.subplot(2, 5, 4)
plt.title('U-Net Prediction \nDSC:%.3f' % (DSC(cart_unet, cart_label)),
fontsize=fz)
plt.imshow(diffmap(cart_unet, cart_label), cmap='gray')
if cart_mask is not None:
plt.subplot(2, 5, 5)
plt.title('Mask-RCNN Prediction \nDSC:%.3f' %
(DSC(cart_mask, cart_label)),
fontsize=fz)
plt.imshow(diffmap(cart_mask, cart_label), cmap='gray')
plt.subplot(2, 5, 6)
plt.title('Polar Patch', fontsize=fz)
plt.imshow(polar_patch, cmap='gray')
plt.subplot(2, 5, 7)
plt.title('Polar Label', fontsize=fz)
plt.imshow(polar_label, cmap='gray')
plt.subplot(2, 5, 8)
plt.title('Polar Prediction', fontsize=fz)
plt.xlim([0, 256])
plt.ylim([polarbd.shape[0], 0])
plt.plot(polarbd[::4, 0],
np.arange(0, polarbd.shape[0], 4),
'o',
markersize=2,
label='Lumen')
plt.plot(polarbd[::4, 1],
np.arange(0, polarbd.shape[0], 4),
'o',
markersize=2,
label='Wall')
plt.legend()
# fig.tight_layout()
if polar_plot_name is not None:
plt.savefig(polar_plot_name)
else:
plt.show()
plt.close()
def get_val_dice(cnn, config):
val_dsc = []
taskdir = cfg['taskdir']
# load array from selected val to test dsc and compare with cart vw label
if os.path.exists(taskdir + '/valarray_cart.npy'):
xarray_val = np.load(taskdir + '/valarray_cart.npy')
xlabel_val = np.load(taskdir + '/vallabel_cart.npy')
else:
val_exams = dbloader.list_exams('val', shuffle=False)
xarray_val = []
xlabel_val = []
bi = 0
for ei in val_exams:
caseloader = CaseLoader(ei)
print(caseloader)
for slicei in caseloader.slices:
if caseloader.valid_slicei(slicei) == False:
continue
if bi % 20 == 0:
polarstack = caseloader.loadstack(slicei,
'cart_patch',
nei=config['depth'] // 2)
xarray_val.append(polarstack[..., None])
cartabel = caseloader.loadstack(slicei,
'cart_label',
nei=config['depth'] //
2)[...,
config['depth'] // 2]
xlabel_val.append(cartabel[:, :, 1] - cartabel[:, :, 0])
bi += 1
xarray_val = np.array(xarray_val)
xlabel_val = np.array(xlabel_val)
print(xarray_val.shape, xlabel_val.shape)
np.save(taskdir + '/valarray_cart.npy', xarray_val)
np.save(taskdir + '/vallabel_cart.npy', xlabel_val)
polarbd_all = cnn.predict(xarray_val,
batch_size=config['G'] * config['batchsize'])
for ti in range(xarray_val.shape[0]):
carlabel = xlabel_val[ti]
polarimg = xarray_val[ti]
polarbd = polarbd_all[ti] * config['width'] // 2
# polarprob = polarprobr[:,:,0,1]-polarprobr[:,:,0,0]
contourin, contourout = toctbd(polarbd, config['patchheight'],
config['patchheight'])
carseg = plotct(polarimg.shape[1], contourin, contourout)
cdsc = DSC(carlabel, carseg)
val_dsc.append(cdsc)
# pxconf = (np.sum(polarprob[polarseg>0])-np.sum(polarprob[polarseg==0]))/np.sum(polarseg>0)
# sconf.append(pxconf)
print('\rPred',
ti,
'/',
xarray_val.shape[0],
'val dice',
'%.5f' % np.mean(val_dsc),
end="")
return np.mean(val_dsc)