def __init__(self):
# Load test DICOM files
desired_path = Path.cwd().joinpath('test', 'testdata')
selected_files = find_DICOM_files(
desired_path) # list of DICOM test files
file_path = os.path.dirname(
os.path.commonprefix(selected_files)) # file path of DICOM files
read_data_dict, file_names_dict = ImageLoading.get_datasets(
selected_files)
# Create patient dict container object
patient_dict_container = PatientDictContainer()
patient_dict_container.clear()
patient_dict_container.set_initial_values(file_path, read_data_dict,
file_names_dict)
# Set additional attributes in patient dict container (otherwise program will crash and test will fail)
if "rtss" in file_names_dict:
dataset_rtss = dcmread(file_names_dict['rtss'])
self.rois = ImageLoading.get_roi_info(dataset_rtss)
dict_raw_contour_data, dict_numpoints = ImageLoading.get_raw_contour_data(
dataset_rtss)
dict_pixluts = ImageLoading.get_pixluts(read_data_dict)
patient_dict_container.set("rois", self.rois)
patient_dict_container.set("raw_contour", dict_raw_contour_data)
patient_dict_container.set("num_points", dict_numpoints)
patient_dict_container.set("pixluts", dict_pixluts)
dict_thickness = ImageLoading.get_thickness_dict(
dataset_rtss, read_data_dict)
if 'rtdose' in file_names_dict:
dataset_rtdose = dcmread(file_names_dict['rtdose'])
# Open the main window
self.main_window = MainWindow()
self.dvh_tab = self.main_window.dvh_tab
self.new_polygons = {}
self.raw_dvh = ImageLoading.multi_calc_dvh(dataset_rtss,
dataset_rtdose, self.rois,
dict_thickness)
self.dvh_x_y = ImageLoading.converge_to_0_dvh(self.raw_dvh)
def load(self, interrupt_flag, progress_callback):
"""
:param interrupt_flag: A threading.Event() object that tells the
function to stop loading.
:param progress_callback: A signal that receives the current
progress of the loading.
:return: PatientDictContainer object containing all values related
to the loaded DICOM files.
"""
progress_callback.emit(("Creating datasets...", 0))
try:
# Gets the common root folder.
path = os.path.dirname(os.path.commonprefix(self.selected_files))
read_data_dict, file_names_dict = ImageLoading.get_datasets(
self.selected_files)
except ImageLoading.NotAllowedClassError:
raise ImageLoading.NotAllowedClassError
# Populate the initial values in the PatientDictContainer singleton.
moving_dict_container = MovingDictContainer()
moving_dict_container.clear()
moving_dict_container.set_initial_values(
path,
read_data_dict,
file_names_dict,
existing_rtss_files=self.existing_rtss)
if interrupt_flag.is_set():
print("stopped")
return False
if 'rtss' in file_names_dict and 'rtdose' in file_names_dict:
self.parent_window.signal_advise_calc_dvh.connect(
self.update_calc_dvh)
self.signal_request_calc_dvh.emit()
while not self.advised_calc_dvh:
pass
if 'rtss' in file_names_dict:
dataset_rtss = dcmread(file_names_dict['rtss'])
progress_callback.emit(("Getting ROI info...", 10))
rois = ImageLoading.get_roi_info(dataset_rtss)
if interrupt_flag.is_set(): # Stop loading.
print("stopped")
return False
progress_callback.emit(("Getting contour data...", 30))
dict_raw_contour_data, dict_numpoints = \
ImageLoading.get_raw_contour_data(dataset_rtss)
# Determine which ROIs are one slice thick
dict_thickness = ImageLoading.get_thickness_dict(
dataset_rtss, read_data_dict)
if interrupt_flag.is_set(): # Stop loading.
print("stopped")
return False
progress_callback.emit(("Getting pixel LUTs...", 50))
dict_pixluts = ImageLoading.get_pixluts(read_data_dict)
if interrupt_flag.is_set(): # Stop loading.
print("stopped")
return False
# Add RTSS values to MovingDictContainer
moving_dict_container.set("rois", rois)
moving_dict_container.set("raw_contour", dict_raw_contour_data)
moving_dict_container.set("num_points", dict_numpoints)
moving_dict_container.set("pixluts", dict_pixluts)
if 'rtdose' in file_names_dict and self.calc_dvh:
dataset_rtdose = dcmread(file_names_dict['rtdose'])
# Spawn-based platforms (i.e Windows and MacOS) have a large
# overhead when creating a new process, which ends up making
# multiprocessing on these platforms more expensive than linear
# calculation. As such, multiprocessing is only available on
# Linux until a better solution is found.
fork_safe_platforms = ['Linux']
if platform.system() in fork_safe_platforms:
progress_callback.emit(("Calculating DVHs...", 60))
raw_dvh = ImageLoading.multi_calc_dvh(
dataset_rtss, dataset_rtdose, rois, dict_thickness)
else:
progress_callback.emit(
("Calculating DVHs... (This may take a while)", 60))
raw_dvh = ImageLoading.calc_dvhs(dataset_rtss,
dataset_rtdose, rois,
dict_thickness,
interrupt_flag)
if interrupt_flag.is_set(): # Stop loading.
print("stopped")
return False
progress_callback.emit(("Converging to zero...", 80))
dvh_x_y = ImageLoading.converge_to_0_dvh(raw_dvh)
if interrupt_flag.is_set(): # Stop loading.
print("stopped")
return False
# Add DVH values to MovingDictContainer
moving_dict_container.set("raw_dvh", raw_dvh)
moving_dict_container.set("dvh_x_y", dvh_x_y)
moving_dict_container.set("dvh_outdated", False)
create_moving_model()
else:
create_moving_model()
self.load_temp_rtss(path, progress_callback, interrupt_flag)
progress_callback.emit(("Loading Moving Model", 85))
if interrupt_flag.is_set(): # Stop loading.
progress_callback.emit(("Stopping", 85))
return False
return True
def load(self, interrupt_flag, progress_callback):
"""
:param interrupt_flag: A threading.Event() object that tells the function to stop loading.
:param progress_callback: A signal that receives the current progress of the loading.
:return: PatientDictContainer object containing all values related to the loaded DICOM files.
"""
progress_callback.emit(("Creating datasets...", 0))
try:
path = os.path.dirname(os.path.commonprefix(self.selected_files)) # Gets the common root folder.
read_data_dict, file_names_dict = ImageLoading.get_datasets(self.selected_files)
except ImageLoading.NotAllowedClassError:
raise ImageLoading.NotAllowedClassError
# Populate the initial values in the PatientDictContainer singleton.
patient_dict_container = PatientDictContainer()
patient_dict_container.clear()
patient_dict_container.set_initial_values(path, read_data_dict, file_names_dict)
# As there is no way to interrupt a QRunnable, this method must check after every step whether or not the
# interrupt flag has been set, in which case it will interrupt this method after the currently processing
# function has finished running. It's not very pretty, and the thread will still run some functions for, in some
# cases, up to a couple seconds after the close button on the Progress Window has been clicked, however it's
# the best solution I could come up with. If you have a cleaner alternative, please make your contribution.
if interrupt_flag.is_set():
print("stopped")
return False
if 'rtss' in file_names_dict and 'rtdose' in file_names_dict:
self.parent_window.signal_advise_calc_dvh.connect(self.update_calc_dvh)
self.signal_request_calc_dvh.emit()
while not self.advised_calc_dvh:
pass
if 'rtss' in file_names_dict:
dataset_rtss = dcmread(file_names_dict['rtss'])
progress_callback.emit(("Getting ROI info...", 10))
rois = ImageLoading.get_roi_info(dataset_rtss)
if interrupt_flag.is_set(): # Stop loading.
print("stopped")
return False
progress_callback.emit(("Getting contour data...", 30))
dict_raw_contour_data, dict_numpoints = ImageLoading.get_raw_contour_data(dataset_rtss)
# Determine which ROIs are one slice thick
dict_thickness = ImageLoading.get_thickness_dict(dataset_rtss, read_data_dict)
if interrupt_flag.is_set(): # Stop loading.
print("stopped")
return False
progress_callback.emit(("Getting pixel LUTs...", 50))
dict_pixluts = ImageLoading.get_pixluts(read_data_dict)
if interrupt_flag.is_set(): # Stop loading.
print("stopped")
return False
# Add RTSS values to PatientDictContainer
patient_dict_container.set("rois", rois)
patient_dict_container.set("raw_contour", dict_raw_contour_data)
patient_dict_container.set("num_points", dict_numpoints)
patient_dict_container.set("pixluts", dict_pixluts)
if 'rtdose' in file_names_dict and self.calc_dvh:
dataset_rtdose = dcmread(file_names_dict['rtdose'])
# Spawn-based platforms (i.e Windows and MacOS) have a large overhead when creating a new process, which
# ends up making multiprocessing on these platforms more expensive than linear calculation. As such,
# multiprocessing is only available on Linux until a better solution is found.
fork_safe_platforms = ['Linux']
if platform.system() in fork_safe_platforms:
progress_callback.emit(("Calculating DVHs...", 60))
raw_dvh = ImageLoading.multi_calc_dvh(dataset_rtss, dataset_rtdose, rois, dict_thickness)
else:
progress_callback.emit(("Calculating DVHs... (This may take a while)", 60))
raw_dvh = ImageLoading.calc_dvhs(dataset_rtss, dataset_rtdose, rois, dict_thickness, interrupt_flag)
if interrupt_flag.is_set(): # Stop loading.
print("stopped")
return False
progress_callback.emit(("Converging to zero...", 80))
dvh_x_y = ImageLoading.converge_to_0_dvh(raw_dvh)
if interrupt_flag.is_set(): # Stop loading.
print("stopped")
return False
# Add DVH values to PatientDictContainer
patient_dict_container.set("raw_dvh", raw_dvh)
patient_dict_container.set("dvh_x_y", dvh_x_y)
patient_dict_container.set("dvh_outdated", False)
return True
else:
return True
return True
