class PatientDataLoader:
def __init__(self, directory=None, case=None):
self.directory = directory
self.case = os.path.join(self.directory, case)
self.acpc = os.path.join(self.case, 'acpc')
self.mni_aff = os.path.join(self.case, 'mniE')
self.mni_f3d = os.path.join(self.case, 'f3dE')
self.medicalImages = MedicalImages.MedicalImages(main=self.case,
acpc=self.acpc,
mni_aff=self.mni_aff,
mni_f3d=self.mni_f3d)
self.electrodes = Electrodes.Electrodes(main=self.case,
acpc=self.acpc,
mni_aff=self.mni_aff,
mni_f3d=self.mni_f3d)
self.templates = TemplateImageLoader.TemplateImageLoader(
datasets=self.directory)
class PatientDataGenerator:
def __init__(self, directory=None, case=None):
self.directory = directory
self.case = os.path.join(self.directory, case)
self.acpc = os.path.join(self.case, 'acpc')
self.mni_aff = os.path.join(self.case, 'mniE')
self.mni_f3d = os.path.join(self.case, 'f3dE')
self.medicalImages = MedicalImages.MedicalImages(main=self.case,
acpc=self.acpc,
mni_aff=self.mni_aff,
mni_f3d=self.mni_f3d)
self.electrodes = Electrodes.Electrodes(main=self.case,
acpc=self.acpc,
mni_aff=self.mni_aff,
mni_f3d=self.mni_f3d)
self.templates = TemplateImageLoader.TemplateImageLoader(
datasets=self.directory)
self.surfaces = Surfaces.Surfaces()
self.filesystem = FileSystem.FileSystem()
# load electrodes
plan = self.electrodes.load(type='plan', suffix='i', space='patient')
impl = self.electrodes.load(type='impl', suffix='i', space='patient')
from Framework.Preprocessing import TemplateImageLoader
from Framework.Visualisation import VTKRenderer
datasets = 'E:\\UCL\\Backup\\medic-biomechanics\\Datasets\\BendingJournal\\datasets'
space = 'mni_aff' # 'patient', 'mni_f3d'
''' Pre-processed patient data '''
pickle_file = open('../Data/data_'+space+'.npy', "rb")
predata = pickle.load(pickle_file)
pickle_file.close()
# cases_generator = CaseGenerator.CaseGenerator()
# cases = cases_generator.get_all()
''' Surfaces '''
gif_ep, gif_tp = [], []
templates = TemplateImageLoader.TemplateLoader(datasets=datasets)
templates.open_mni_images()
surfaces = Surfaces.Surfaces()
''' Visualisation'''
renderer = VTKRenderer.VTKRenderer()
renderer.create_vtk_renderer()
# iterate through cases
N = 0
N_da = 0
for i in range(len(predata['case'])):
''' Pre-computed data '''
plan = predata['plan'][i]
impl = predata['impl'][i]
local_delta = predata['local_delta'][i]
class SyntheticDataGenerator:
def __init__(self, directory=None, case=None):
self.directory = directory
self.case = os.path.join(self.directory, case)
self.templates = TemplateImageLoader.TemplateImageLoader(datasets=self.directory)
def main():
# select regions of interest
plangen = PlanGenerator.PlanGenerator()
# plangen.ep_superior_frontal_gyrus()
# plangen.ep_middle_frontal_gyrus()
# plangen.ep_inferior_frontal_orbital_gyrus()
# plangen.ep_temporal_gyrus()
# plangen.ep_inferior_medial_temporal_gyrus()
# plangen.ep_superior_temporal_gyrus()
# plangen.ep_anterior_posterior_central_gyrus()
plangen.ep_parietal_occipital()
# plangen.ep_parietal_lobule()
# plangen.ep_occipital()
print('filter: filename={} ep={} tp={}'.format(plangen.filename, plangen.ep, plangen.tp))
# open data
input_data = InputData.InputData(directory=data_dir, file=data_file)
selection = input_data.search_by_region(ep=plangen.ep, tp=plangen.tp)
print('selection', selection)
# save selection
pickle_file = open(os.path.join(data_dir, plangen.filename), "wb")
pickle.dump(selection, pickle_file)
pickle_file.close()
''' Visualisation '''
renderer = VTKRenderer.VTKRenderer()
renderer.create_vtk_renderer()
''' Surfaces '''
templates = TemplateImageLoader.TemplateImageLoader(datasets=datasets)
templates.open_mni_images()
surfaces = Surfaces.Surfaces()
for s in plangen.ep + plangen.tp:
surfaces.load(directory=templates.templates_dir, image=templates.mni['GIF'], gif=s)
# iterate through selection
print('Selection:')
N = 0
for case in selection.keys():
# index of case
c = input_data.data['case'].index(case)
N += len(selection[case])
print('case={} electrodes={}'.format(case, selection[case]))
# Pre-computed data
plan = input_data.data['plan'][c]
impl = input_data.data['impl'][c]
local_delta = input_data.data['local_delta'][c]
delta = input_data.data['delta'][c]
# Visualisation
if space is not 'mni_f3d':
renderer.render_trajectories(data=plan, filter=selection[case], colour=[1.0, 1.0, 1.0], radius=0.4, opacity=0.4)
renderer.render_trajectories(data=impl, filter=selection[case], colour=[], radius=0.4, opacity=0.4)
renderer.render_vectorfield(data=delta, source='from', vector='u', filter=selection[case], colour=[])
renderer.render_vectorfield(data=local_delta, source='from', vector='u', filter=selection[case], colour=[1.0, 0.0, 0.0])
renderer.render_vectorfield(data=local_delta, source='from_dir', vector='dir', filter=selection[case], colour=[0.0, 1.0, 0.0])
for s in list(np.unique(plangen.ep + list(np.asarray(plangen.tp)[np.isin(plangen.tp, plangen.all_eps)]))):
if s not in list(MedicalImages.wmGif):
renderer.render_surface(polydata=surfaces.polydata[s], colour=[1.,1.,1.], opacity=0.2)
for s in list(np.asarray(plangen.tp)[np.isin(plangen.tp, plangen.all_eps, invert=True)]):
if s not in list(MedicalImages.wmGif):
renderer.render_surface(polydata=surfaces.polydata[s], colour=[0.,0.95,0.95], opacity=0.2)
print('filter={} num_cases={} num_electrodes={}'.format(plangen.filename, len(selection.keys()), N))
# render
renderer.execute()