Example

To classify a single image:

from tbpcxr.model import Model
from tbpcxr.utilities import read_dcm

outlier_model = Model.load_outlier_pcamodel()

img = read_dcm(path_to_file)

arr = outlier_model.to_observations([img])

if outlier_model.outlier_predictor(arr)[0] == -1:
   print("{} is an outlier".format(path_to_file))

Multiple images can efficiently be processed by using Python’s map function, which

from tbpcxr.model import Model
from tbpcxr.utilities import read_dcm

outlier_model = Model.load_outlier_pcamodel()

arr = outlier_model.to_observations(map(read_dcm, image_file_list))

results = outlier_model.outlier_predictor(arr)

for fn in [fn for fn, o in zip(image_file_list, results) if o == -1]:
   print("{} is an outlier".format(fn))