def samples_replace(
source_path="/data/fyl/data_samples/tianchild_cubes_overbound/npy_non_fine",
target_path="/data/fyl/data_samples/tianchild_cubes_overbound/npy_non"
):
'''
filelist = glob.glob(source_path+'/*.npy')
for file in filelist:
filename = os.path.basename(file)
target_file = target_path + '/' + filename
shutil.copyfile(file, target_file)
print(target_file)
'''
bt.directory_arrange(source_path, target_path, mode='move')
def __init__(self,
filelist,
data_size=64,
mode=None,
file_retrieval=False):
if type(filelist) == str:
self.filelist = bt.filelist_load(filelist)
elif type(filelist) == list:
self.filelist = filelist
else:
print('invalid filelist type:{}'.format(type(filelist)))
exit()
if 'lidc' in filelist[0]:
self.dataset_name = 'LIDC'
elif 'sph' in filelist[0]:
self.dataset_name = 'SPH'
elif 'luna' in filelist[0]:
self.dataset_name = 'LUNA'
elif 'tianchild' in filelist[0]:
self.dataset_name = 'TLD'
self.data_size = data_size
self.mode = mode
self.file_retrieval = file_retrieval
os.makedirs(net_store_path)
if os.access(tensorboard_path, os.F_OK):
shutil.rmtree(tensorboard_path)
#data arrangement
luna_dir = "../data_samples/luna_cubes_56_overbound"
#data_dir2 = "../data_samples/tianchi_cubes_56_overbound"
slh_dir = "../data_samples/slh_cubes_56_overbound"
luna_trainsets = [
"subset0", "subset1", "subset2", "subset3", "subset4", "subset5",
"subset6", "subset7", "subset8"
]
luna_valsets = ["subset9"]
tianchi_trainsets = ["train", "val"]
#train_sets = ["subset9"]
'''
if "pfilelist_path" in dir() and os.path.exists(pfilelist_path):
print("read pfilelist from: %s" %(pfilelist_path))
pfiles = bt.filelist_load(pfilelist_path)
#pfilelist_file = open(pfilelist_path, "r")
#pfiles = pfilelist_file.readlines()
#for pfi in range(len(pfiles)):
# pfiles[pfi] = pfiles[pfi][:-1]
#pfilelist_file.close()
else:
pfiles = []
for set in train_sets:
train_dir = os.path.join(data_dir1, set)
pdir = os.path.join(train_dir,"npy","*.npy")
pfiles.extend(glob.glob(pdir))
#ndir = os.path.join(train_dir,"npy_non","*.npy")
MAX_BOUND = float(config["MAX_BOUND"])
MIN_BOUND = float(config["MIN_BOUND"])
PIXEL_MEAN = float(config["PIXEL_MEAN"])
NORM_CROP = config["NORM_CROP"]
REGION_SIZE = 64
BATCH_SIZE = 10
net_init_path = "models_tensorflow/lidc_3D_multi_crop_net_64_aug2"
net_init_files = [
net_init_path + "/epoch7/epoch7", net_init_path + "/epoch9/epoch9",
net_init_path + "/epoch10/epoch10"
]
test_filelist_path = net_init_path + "/filelist_val.log"
test_files = bt.filelist_load(test_filelist_path)
test_num = len(test_files)
#net construct
volume_input = tf.placeholder(tf.float32,
[None, REGION_SIZE, REGION_SIZE, REGION_SIZE])
volume_reshape = tf.reshape(volume_input,
[-1, REGION_SIZE, REGION_SIZE, REGION_SIZE, 1])
real_label = tf.placeholder(tf.float32, [None, 2])
net_outs1, variables1 = tmc.multi_crop_net(volume_reshape, poolings=[1, 1, 2])
net_outs2, variables2 = tmc.multi_crop_net(volume_reshape, poolings=[1, 1, 2])
net_outs3, variables3 = tmc.multi_crop_net(volume_reshape, poolings=[1, 1, 2])
prediction_fusion = net_outs1['sm_out'] + net_outs2['sm_out'] + net_outs3[
'sm_out']
correct_prediction = tf.equal(tf.argmax(prediction_fusion, 1),
tf.argmax(real_label, 1))
from tensorflow.python import debug as tf_debug
import tensorflow as tf
import numpy as np
import os
import sys
import shutil
import glob
import math
import random
sys.path.append("/home/fyl/programs/lung_project")
from toolbox import BasicTools as bt
from toolbox import MITools as mt
from toolbox import Data_Augmentation as da
from toolbox import TensorflowTools as tft
constants = bt.read_constants("./constants2.txt")
os.environ['CUDA_VISIBLE_DEVICES'] = constants["GPU_DEVICE"]
REGION_SIZE = constants["REGION_SIZE"]
MAX_BOUND = float(constants["MAX_BOUND"])
MIN_BOUND = float(constants["MIN_BOUND"])
PIXEL_MEAN = float(constants["PIXEL_MEAN"])
#NUM_EPOCH = 200
SNAPSHOT_EPOCH = 20
#DECAY_EPOCH = 0
#INITIAL_LEARNING_RATE = 0.001
#DECAY_LEARNING_RATE = 1.0
BATCH_SIZE = 30
VALIDATION_RATE = 0.2
label = label_dict[opt.label_mode]
use_gpu = opt.use_gpu
net_file = opt.load_model_path
if "test_paths" in dir():
all_patients = []
for path in test_paths:
all_patients += glob(path + "/*.mhd")
if len(all_patients) <= 0:
print("No patient found")
exit()
else:
print("No test data")
exit()
if hasattr(opt, 'filelists') and 'test' in opt.filelists.keys():
test_samples = bt.filelist_load(opt.filelists['test'])
test_uids = []
for test_sample in test_samples:
sample_uid = os.path.basename(test_sample).split('_')[0]
if sample_uid not in test_uids:
test_uids.append(sample_uid)
#else:
# for path in opt.filelists['test']:
# test_samples = glob(path + '/*.mhd')
if 'vision_path' in dir() and vision_path is not None and not os.access(
vision_path, os.F_OK):
os.makedirs(vision_path)
if os.access(evaluation_path, os.F_OK): shutil.rmtree(evaluation_path)
if not os.access(evaluation_path, os.F_OK): os.makedirs(evaluation_path)
#model = models.DensecropNet(input_size=region_size, drop_rate=0, growth_rate=64, num_blocks=4, num_fin_growth=3).eval()
def get_pathology_filelists(datasetidx=''):
root_name = 'root_dir'
if datasetidx is not None: root_name += str(datasetidx)
if hasattr(opt, root_name) and getattr(opt, root_name) is not None:
#filelists, folddict = lt.filelist_training(opt.root_dir, remove_uncertain=opt.remove_uncertain, shuffle = opt.filelist_shuffle, cross_fold = opt.num_cross_folds, val_fold = opt.val_fold, test_fold = opt.test_fold)
file_root = getattr(opt, root_name)
if os.path.isfile(file_root):
filelist = bt.filelist_load(file_root)
else:
filelist = glob.glob(
os.path.join(getattr(opt, root_name), "npy", "*.npy"))
#mt.sample_statistics(filelist, True)
if 'lidc' in getattr(opt, root_name) and hasattr(
opt, 'remove_uncertain') and opt.remove_uncertain:
filelist = lt.filelist_remove_uncertain(filelist)
if opt.filelist_shuffle:
random.shuffle(filelist)
filelists, folddict = bt.foldlist(
filelist, opt.num_cross_folds, {
'val': getattr(opt, 'val_fold' + str(datasetidx)),
'test': getattr(opt, 'test_fold' + str(datasetidx))
})
if 'train' in filelists.keys():
bt.filelist_store(
filelists['train'], save_path + '/filelist' + str(datasetidx) +
'_train_fold' + str(folddict['train']) + '.log')
if 'val' in filelists.keys():
bt.filelist_store(
filelists['val'], save_path + '/filelist' + str(datasetidx) +
'_val_fold' + str(folddict['val']) + '.log')
if 'test' in filelists.keys():
bt.filelist_store(
filelists['test'], save_path + '/filelist' + str(datasetidx) +
'_test_fold' + str(folddict['test']) + '.log')
#vis.log("trainfold:{} valfold:{} testfold:{}" .format(folddict['train'], folddict['val'], folddict['test']))
print("filelist generated")
else:
filelists = {}
filelists['train'] = bt.filelist_load(
getattr(opt, 'filelists' + str(datasetidx))['train'])
filelists['val'] = bt.filelist_load(
getattr(opt, 'filelists' + str(datasetidx))['val'])
bt.filelist_store(filelists['train'],
save_path + '/filelist_train.log')
bt.filelist_store(filelists['val'], save_path + '/filelist_val.log')
if 'test' in getattr(opt, 'filelists' + str(datasetidx)).keys():
filelists['test'] = bt.filelist_load(
getattr(opt, 'filelists' + str(datasetidx))['test'])
bt.filelist_store(filelists['test'],
save_path + '/filelist_test.log')
print("filelist loaded")
return filelists
batch_normalization_statistic=False,
bn_params=bn_params)
prediction_out = outputs['sm_out']
saver = tf.train.Saver()
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
saver.restore(sess, net_file)
start_time = time.time()
#patient_evaluations = open(evaluation_path + "/patient_evaluations.log", "w")
results = []
CPMs = []
CPMs2 = []
test_patients = all_patients[4:5]
bt.filelist_store(all_patients, evaluation_path + "/patientfilelist.log")
#random.shuffle(test_patients)
for p in range(len(test_patients)):
patient = test_patients[p]
#patient = "./LUNA16/subset9/1.3.6.1.4.1.14519.5.2.1.6279.6001.212608679077007918190529579976.mhd"
patient = "./LUNA16/subset9/1.3.6.1.4.1.14519.5.2.1.6279.6001.102681962408431413578140925249.mhd"
#patient = "./TIANCHI_examples/LKDS-00005.mhd"
uid = mt.get_serie_uid(patient)
annotations = mt.get_annotations(uid, annotation_file)
if len(annotations) == 0:
print('%d/%d patient %s has no annotations, ignore it.' %
(p + 1, len(test_patients), uid))
#patient_evaluations.write('%d/%d patient %s has no annotations, ignore it\n' %(p+1, len(test_patients), uid))
continue
print('%d/%d processing patient:%s' % (p + 1, len(test_patients), uid))
net_init_paths[1] + "/batch_normalization_statistic.npy",
net_init_paths[2] + "/batch_normalization_statistic_25.npy"
]
#pfilelist_path = net_init_paths[0] + "/pfilelist.log"
#nfilelist_path = net_init_paths[0] + "/nfilelist.log"
if not 'pfilelist_path' in dir():
pfilelist_path = None
if not 'nfilelist_path' in dir():
nfilelist_path = None
if os.access(net_store_path, os.F_OK):
shutil.rmtree(net_store_path)
os.makedirs(net_store_path)
if os.access(tensorboard_path, os.F_OK):
shutil.rmtree(tensorboard_path)
'''
#data arrangement
luna_dir = "luna_cubes_56_overbound"
#data_dir2 = "tianchi_cubes_56_overbound"
slh_dir = "slh_cubes_56_overbound"
luna_trainsets = ["subset0", "subset1", "subset2", "subset3", "subset4", "subset5", "subset6", "subset7", "subset8"]
luna_valsets = ["subset9"]
tianchi_trainsets = ["train", "val"]
#luna_trainsets = ["subset9"]
if "pfilelist_path" in dir() and os.path.exists(pfilelist_path):
print("read pfilelist from: %s" %(pfilelist_path))
pfiles = bt.filelist_load(pfilelist_path)
#pfilelist_file = open(pfilelist_path, "r")
#pfiles = pfilelist_file.readlines()
#for pfi in range(len(pfiles)):
if "test_paths" in dir():
all_patients = []
for path in test_paths:
all_patients += glob(path + "/*.mhd")
if len(all_patients)<=0:
print("No patient found")
exit()
else:
print("No test data")
exit()
if "test_sample_filelist" not in dir():
for path in test_sample_paths:
test_samples = glob(path + '/*.npy')
else:
test_samples = bt.filelist_load(test_sample_filelist)
test_uids = []
for test_sample in test_samples:
sample_uid = os.path.basename(test_sample).split('_')[0]
if sample_uid not in test_uids:
test_uids.append(sample_uid)
if 'vision_path' in dir() and vision_path is not None and not os.access(vision_path, os.F_OK):
os.makedirs(vision_path)
#if os.access(evaluation_path, os.F_OK): shutil.rmtree(evaluation_path)
if not os.access(evaluation_path, os.F_OK): os.makedirs(evaluation_path)
shutil.copyfile(net_file, evaluation_path+'/'+net_file.split('/')[-1])
model = models.DensecropNet(input_size=REGION_SIZE, drop_rate=0, growth_rate=64, num_blocks=4, num_fin_growth=3)
model.load(net_file)
model.eval()
model.cuda()
shutil.rmtree(tensorboard_path)
#data arrangement
data_dir1 = "luna_cubes_56_overbound"
#data_dir2 = "tianchi_cubes_56_overbound"
data_dir3 = "slh_cubes_56_overbound"
train_sets = [
"subset0", "subset1", "subset2", "subset3", "subset4", "subset5",
"subset6", "subset7", "subset8"
]
train_sets2 = ["train", "val"]
#train_sets = ["subset9"]
if "pfilelist_path" in dir() and os.path.exists(pfilelist_path):
print("read pfilelist from: %s" % (pfilelist_path))
pfiles = bt.filelist_load(pfilelist_path)
#pfilelist_file = open(pfilelist_path, "r")
#pfiles = pfilelist_file.readlines()
#for pfi in range(len(pfiles)):
# pfiles[pfi] = pfiles[pfi][:-1]
#pfilelist_file.close()
else:
pfiles = []
for set in train_sets:
train_dir = os.path.join(data_dir1, set)
pdir = os.path.join(train_dir, "npy", "*.npy")
pfiles.extend(glob.glob(pdir))
#ndir = os.path.join(train_dir,"npy_non","*.npy")
#nfiles.extend(glob.glob(ndir))
if "data_dir2" in dir():
for set in train_sets2:
fusion_file = "models_tensorflow/luna_slh_3D_fusion2/epoch6/epoch6"
bn_files = [
"models_tensorflow/luna_slh_3D_bndo_flbias_l5_20_aug_stage3/batch_normalization_statistic.npy",
"models_tensorflow/luna_slh_3D_bndo_flbias_l5_30_aug2_stage2/batch_normalization_statistic.npy",
"models_tensorflow/luna_slh_3D_bndo_flbias_l6_40_aug_stage2/batch_normalization_statistic_25.npy"
]
#annotation_file = "LUNA16/csvfiles/annotations_corrected.csv"
annotation_file = "../datasets/SPH_data/annotations.xlsx"
#exclude_file = "LUNA16/csvfiles/annotations_excluded.csv"
vision_path = "./detection_vision/sph"
result_path = "./results"
evaluation_path = result_path + "/evaluation_" + FUSION_MODE + "fusion_segmentation_sph"
result_file = evaluation_path + "/result.csv"
if "test_filelist" in dir() and os.access(test_filelist, os.F_OK):
all_patients = bt.filelist_load(test_filelist)
elif "test_paths" in dir():
all_patients = []
#for path in test_paths:
# all_patients += glob(path + "/*.mhd")
for path in test_paths:
all_patients.extend(bt.get_dirs(path))
if len(all_patients) <= 0:
print("No patient found")
exit()
#random.shuffle(all_patients)
bt.filelist_store(all_patients, evaluation_path + "_filelist.log")
else:
print("No test data")
exit()
if 'vision_path' in dir() and 'vision_path' is not None and not os.access(
predlist[0].append(predsplit[0])
predlist[1].append(float(predsplit[1]))
return predlist
def prediction_filter(predlist, lower, upper):
filteredlist = ([], [])
for pi in range(len(predlist[0])):
if predlist[1][pi] >= lower and predlist[1][pi] <= upper:
filteredlist[0].append(predlist[0][pi])
filteredlist[1].append(predlist[1][pi])
return filteredlist
strlist1 = bt.filelist_load(
'predictions/train/DensecropNet_Iterative_detection_2_stage2_epoch4_corrpreds.log'
)
predlist1 = extract_predictions(strlist1)
filteredlist1 = prediction_filter(predlist1, 0, 0.1)
predset1 = set(filteredlist1[0])
strlist2 = bt.filelist_load(
'predictions/train/DensecropNet_Iterative_detection_3_epoch6_corrpreds.log'
)
predlist2 = extract_predictions(strlist2)
filteredlist2 = prediction_filter(predlist2, 0, 0.1)
predset2 = set(filteredlist2[0])
strlist3 = bt.filelist_load(
'predictions/train/DensecropNet_Iterative_detection_4_epoch4_corrpreds.log'
)
predlist3 = extract_predictions(strlist3)
filteredlist3 = prediction_filter(predlist3, 0, 0.1)
def get_detection_filelists(patient_uids=None,
filepaths=None,
easy_eliminate_filelist=None,
config={}):
save_path = config['model_root'] + '/' + config['env']
if easy_eliminate_filelist is not None:
easy_eliminate_filelist = bt.filelist_load(easy_eliminate_filelist)
if filepaths is None:
filelists = {}
filelists['train'] = bt.filelist_load(config['filelists']['train'])
filelists['val'] = bt.filelist_load(config['filelists']['val'])
if easy_eliminate_filelist is not None:
filelists['train'] = bt.filelist_eliminate(
filelists['train'], easy_eliminate_filelist)
filelists['val'] = bt.filelist_eliminate(filelists['val'],
easy_eliminate_filelist)
bt.filelist_store(filelists['train'],
save_path + '/filelist_train.log')
bt.filelist_store(filelists['val'], save_path + '/filelist_val.log')
print("filelist loaded")
else:
#filepaths=["/home/fyl/datasets/luna_64/train", "/home/fyl/datasets/luna_64/test", "/home/fyl/datasets/npy_non_set"]
#filelist = []
if patient_uids is None:
patient_uids = []
for filepath in filepaths:
files = glob.glob(filepath + '/*.npy')
#filelist.extend(files)
for file in files:
filename = os.path.basename(file)
filenamenoext = os.path.splitext(filename)[0]
fileinfos = filenamenoext.split('_')
annolabel = fileinfos[-1]
patient_uid = fileinfos[0]
if patient_uid not in patient_uids:
patient_uids.append(patient_uid)
elif type(patient_uids) == str:
patient_uids = bt.filelist_load(patient_uids)
#patient_temp = patient_uids[int(len(patient_uids)/10.0+0.5)*3:]
#random.shuffle(patient_temp)
#patient_uids[int(len(patient_uids)/10.0+0.5)*3:] = patient_temp
if config['filelist_shuffle']: random.shuffle(patient_uids)
bt.filelist_store(patient_uids, save_path + '/patientlist.log')
patient_folds, folddict = bt.foldlist(patient_uids,
config['num_cross_folds'], {
'val': config['val_fold'],
'test': config['test_fold']
})
filelist_overall = []
filelists = {}
for setname in patient_folds.keys():
filelists[setname] = []
for filelist in filepaths:
if os.path.isfile(filelist):
files = bt.filelist_load(filelist)
else:
files = os.listdir(filelist)
for f in range(len(files)):
files[f] = filelist + '/' + files[f]
#files = glob.glob(filelist+'/*.%s' %(fileext))
filelist_overall.extend(files)
if 'lidc' in filelist and 'remove_uncertain' in config.keys(
) and config['remove_uncertain']:
filelist_overall = lt.filelist_remove_uncertain(
filelist_overall)
if easy_eliminate_filelist is not None:
filelist_overall = bt.filelist_eliminate(filelist_overall,
easy_eliminate_filelist)
if config['filelist_shuffle']: random.shuffle(filelist_overall)
for file in filelist_overall:
filename_split = os.path.splitext(
os.path.basename(file))[0].split('_')
#if 'label_choice' in config.keys() and filename_split[-1]=='annotation' and config['label_choice']!=filename_split[2]: continue
patient_uid = filename_split[0]
for setname in patient_folds.keys():
if patient_uid in patient_folds[setname]:
filelists[setname].append(file)
for setname in patient_folds.keys():
bt.filelist_store(
filelists[setname], save_path + '/filelist_' + setname +
'_fold' + str(folddict[setname]) + '.log')
bt.filelist_store(filelist_overall, save_path + '/filelist.log')
print("filelist generated")
'''
filelists = get_filelists_patientwise(patient_uids, filepaths, fileext='npy', config=config)
for setname in filelists.keys():
if easy_eliminate_filelist is not None:
filelists[setname] = bt.filelist_eliminate(filelists[setname], easy_eliminate_filelist)
'''
return filelists
shutil.rmtree(tensorboard_path)
#data arrangement
data_dir1 = "luna_cubes_56_overbound"
#data_dir2 = "tianchi_cubes_56_overbound"
data_dir3 = "slh_cubes_56_overbound"
train_sets = [
"subset0", "subset1", "subset2", "subset3", "subset4", "subset5",
"subset6", "subset7", "subset8"
]
train_sets2 = ["train", "val"]
#train_sets = ["subset9"]
if "pfilelist_path" in dir() and os.path.exists(pfilelist_path):
print("read pfilelist from: %s" % (pfilelist_path))
pfiles = bt.filelist_load(pfilelist_path)
#pfilelist_file = open(pfilelist_path, "r")
#pfiles = pfilelist_file.readlines()
#for pfi in range(len(pfiles)):
# pfiles[pfi] = pfiles[pfi][:-1]
#pfilelist_file.close()
else:
pfiles = []
for set in train_sets:
train_dir = os.path.join(data_dir1, set)
pdir = os.path.join(train_dir, "npy", "*.npy")
pfiles.extend(glob.glob(pdir))
#ndir = os.path.join(train_dir,"npy_non","*.npy")
#nfiles.extend(glob.glob(ndir))
if "data_dir2" in dir():
for set in train_sets2:
def detection_fusion(
test_path=None,
result_path="./experiments_dt/evaluations_tianchild_densecropnet_31,32",
**kwargs):
opt.parse(kwargs)
if test_path is None:
test_paths = ["/data/fyl/datasets/Tianchi_Lung_Disease/train"]
else:
test_paths = [test_path]
#test_sample_filelist = "/data/fyl/models_pytorch/DensecropNet_detection_test_rfold1/filelist_val_fold0.log"
net_files = [
"/data/fyl/models_pytorch/DensecropNet_arterio_detection_rfold1/DensecropNet_arterio_detection_rfold1_epoch2",
"/data/fyl/models_pytorch/DensecropNet_lymphnodecal_detection_rfold1/DensecropNet_lymphnodecal_detection_rfold1_epoch2"
]
annotation_file = "/data/fyl/datasets/Tianchi_Lung_Disease/chestCT_round1_annotation.csv"
#candidate_file = "/data/fyl/datasets/Tianchi_Lung_Disease/candidate.csv"
labels = [31, 32]
#result_path = "./experiments_dt/evaluations_tianchild_densecropnet_fusion"
#vision_path = result_path
#result_file = result_path + "/result.csv"
hard_negatives_file = result_path + "/hard_negatives.csv"
region_size = opt.input_size
batch_size = opt.batch_size
use_gpu = opt.use_gpu
if 'vision_path' in dir() and vision_path is not None and not os.access(
vision_path, os.F_OK):
os.makedirs(vision_path)
#if os.access(result_path, os.F_OK): shutil.rmtree(result_path)
if not os.access(result_path, os.F_OK): os.makedirs(result_path)
if "test_paths" in dir():
all_patients = []
for path in test_paths:
all_patients += glob(path + "/*.mhd")
if len(all_patients) <= 0:
print("No patient found")
exit()
else:
print("No test data")
exit()
if hasattr(opt, 'filelists') and 'test' in opt.filelists.keys():
test_samples = bt.filelist_load(opt.filelists['test'])
test_uids = []
for test_sample in test_samples:
sample_uid = os.path.basename(test_sample).split('_')[0]
if sample_uid not in test_uids:
test_uids.append(sample_uid)
pd.DataFrame(data=test_uids,
columns=['series_uid'
]).to_csv(result_path + '/patients_uid.csv',
index=False)
#else:
# for path in opt.filelists['test']:
# test_samples = glob(path + '/*.mhd')
#model = models.DensecropNet(input_size=region_size, drop_rate=0, growth_rate=64, num_blocks=4, num_fin_growth=3).eval()
networks = [
getattr(models, opt.model)(input_size=region_size,
**opt.model_setup).eval()
for m in range(len(net_files))
]
for n in range(len(net_files)):
networks[n].load(net_files[n])
print('model loaded from %s' % (net_files[n]))
shutil.copyfile(net_files[n],
result_path + '/' + net_files[n].split('/')[-1])
if use_gpu: networks[n].cuda()
start_time = time.time()
#patient_evaluations = open(result_path + "/patient_evaluations.log", "w")
results = []
labeled_results = [[] for l in range(len(labels))]
CPMs = [[] for l in range(len(labels))]
#hard_negatives = []
test_patients = all_patients
#random.shuffle(test_patients)
bt.filelist_store(test_patients, result_path + "/patientfilelist.log")
for p in range(len(test_patients)):
patient = test_patients[p]
uid = mt.get_mhd_uid(patient)
if 'test_uids' in dir() and uid not in test_uids:
print("%d/%d patient %s not belong to test set" %
(p + 1, len(test_patients), uid))
continue
print('%d/%d processing patient:%s' % (p + 1, len(test_patients), uid))
full_image_info = sitk.ReadImage(patient)
full_scan = sitk.GetArrayFromImage(full_image_info)
origin = np.array(full_image_info.GetOrigin(
))[::-1] #the order of origin and old_spacing is initially [z,y,x]
old_spacing = np.array(full_image_info.GetSpacing())[::-1]
image, new_spacing = mt.resample(full_scan, old_spacing,
np.array([1, 1, 1]))
#image = np.load(patient)
#new_spacing = np.array([1, 1, 1])
#origin = np.array([0, 0, 0])
print('Resample Done. time:{}s'.format(time.time() - start_time))
candidate_results = nd.slic_candidate(image, 20, focus_area='body')
if candidate_results is None:
continue
candidate_coords, candidate_labels, cluster_labels = candidate_results
if 'vision_path' in dir() and vision_path is not None:
np.save(vision_path + "/" + uid + "_segmask.npy", cluster_labels)
#segresult = lc.segment_vision(image, cluster_labels)
#np.save(vision_path + "/" + uid + "_segresult.npy", segresult)
print('Candidate Done. time:{}s'.format(time.time() - start_time))
print('candidate number:%d' % (len(candidate_coords)))
candidate_predictions = nd.precise_detection_pt(
image,
region_size,
candidate_coords,
networks,
None,
batch_size,
use_gpu=use_gpu,
prediction_threshold=0.4)
labeled_predictions = []
for l in range(len(labels)):
label = labels[l]
print('label: %d' % (label))
evaluation_path = result_path + '/' + str(label)
if not os.access(evaluation_path, os.F_OK):
os.makedirs(evaluation_path)
if 'annotation_file' in dir():
annotations = mt.get_challenge_annotations(uid,
annotation_file,
label=label)
if len(annotations) == 0:
print("%d/%d patient %s has no annotations, ignore it." %
(p + 1, len(test_patients), uid))
#patient_evaluations.write('%d/%d patient %s has no annotations, ignore it\n' %(p+1, len(test_patients), uid))
continue
#make a real lesion visualization
if 'vision_path' in dir() and vision_path is not None:
real_lesions = []
for annotation in annotations:
#real_lesion = np.int_([abs(annotation[2]-origin[0])/new_spacing[0], abs(annotation[1]-origin[1])/new_spacing[1], abs(annotation[0]-origin[2])/new_spacing[2]])
real_lesion = mt.coord_conversion(annotation[:3][::-1],
origin,
old_spacing,
full_scan.shape,
image.shape,
dir_array=True)
real_lesions.append(real_lesion)
annotation_vision = cvm.view_coordinates(image,
real_lesions,
window_size=10,
reverse=False,
slicewise=False,
show=False)
np.save(evaluation_path + "/" + uid + "_annotations.npy",
annotation_vision)
positive_predictions = candidate_predictions[l] > 0
result_predictions, result_labels = nd.predictions_map_fast(
cluster_labels, candidate_predictions[l][positive_predictions],
candidate_labels[positive_predictions])
labeled_predictions.append(result_predictions)
if 'vision_path' in dir() and vision_path is not None:
np.save(evaluation_path + "/" + uid + "_detlabels.npy",
result_labels)
np.save(evaluation_path + "/" + uid + "_detpredictions.npy",
result_predictions)
#detresult = lc.segment_vision(image, result_labels)
#np.save(evaluation_path+"/"+uid+"_detresult.npy", detresult)
lesion_center_predictions = nd.prediction_centering_fast(
result_predictions)
#lesion_center_predictions, prediction_labels = nd.prediction_cluster(result_predictions)
if 'vision_path' in dir() and vision_path is not None:
lesions = []
for nc in range(len(lesion_center_predictions)):
lesions.append(np.int_(lesion_center_predictions[nc][0:3]))
volume_predicted = cvm.view_coordinates(result_predictions *
1000,
lesions,
window_size=10,
reverse=False,
slicewise=False,
show=False)
np.save(evaluation_path + "/" + uid + "_prediction.npy",
volume_predicted)
if 'prediction_labels' in dir():
prediction_cluster_vision = lc.segment_color_vision(
prediction_labels)
np.save(
evaluation_path + "/" + uid +
"_prediction_clusters.npy", prediction_cluster_vision)
print('Detection Done. time:{}s'.format(time.time() - start_time))
'''
#randomly create a result for testing
lesion_center_predictions = []
for nc in range(10):
lesion_center_predictions.append([random.randint(0,image.shape[0]-1), random.randint(0,image.shape[1]-1), random.randint(0,image.shape[2]-1), random.random()])
'''
for nc in range(len(lesion_center_predictions)):
#the output coordination order is [x,y,z], while the order for volume image should be [z,y,x]
result = [uid]
result.extend(
mt.coord_conversion(lesion_center_predictions[nc][:3],
origin,
old_spacing,
full_scan.shape,
image.shape,
dir_array=False)[::-1])
if label is not None: result.append(label)
result.append(lesion_center_predictions[nc][3])
#results.append(result)
labeled_results[l].append(result)
columns = [
'seriesuid', 'coordX', 'coordY', 'coordZ', 'probability'
]
if label is not None:
columns.insert(4, 'class')
result_frame = pd.DataFrame(data=labeled_results[l],
columns=columns)
result_frame.to_csv("{}/result_{}.csv".format(
evaluation_path, label),
index=False,
float_format='%f')
#np.save("{}/result_{}.npy"%(evaluation_path, label), np.array(results))
if 'annotation_file' in dir():
assessment = eva.detection_assessment(labeled_results[l],
annotation_file,
label=label)
if assessment is None:
print('assessment failed')
#patient_evaluations.write('%d/%d patient %s assessment failed\n' %(p+1, len(test_patients), uid))
continue
#num_scans, FPsperscan, sensitivities, CPMscore, FPsperscan2, sensitivities2, CPMscore2, lesions_detected = assessment
num_scans = assessment['num_scans']
FPsperscan, sensitivities = assessment['FROC']
CPMscore = assessment['CPM']
prediction_order = assessment['prediction_order']
lesions_detected = assessment['detection_cites']
if len(FPsperscan) <= 0 or len(sensitivities) <= 0:
print("No results to evaluate, continue")
else:
eva.evaluation_vision(CPMs[l],
num_scans,
FPsperscan,
sensitivities,
CPMscore,
lesions_detected,
output_path=evaluation_path)
#patient_evaluations.write('%d/%d patient %s CPM score:%f\n' %(p+1, len(test_patients), uid, single_assessment[6]))
print('Evaluation Done. time:{}s'.format(time.time() -
start_time))
labeled_predictions = np.array(labeled_predictions)
prediction_labels = np.argmax(labeled_predictions, axis=0)
predictions_fusion = labeled_predictions.sum(axis=0) / 4.0
fused_center_predictions = nd.prediction_centering_fast(
predictions_fusion)
if 'vision_path' in dir() and vision_path is not None:
np.save(vision_path + "/" + uid + "_classlabels.npy",
prediction_labels)
for lcp in range(len(fused_center_predictions)):
#the output coordination order is [x,y,z], while the order for volume image should be [z,y,x]
center = fused_center_predictions[lcp]
result = [uid]
result.extend(
mt.coord_conversion(center[:3],
origin,
old_spacing,
full_scan.shape,
image.shape,
dir_array=False)[::-1])
result.append(labels[prediction_labels[center[0], center[1],
center[2]]])
result.append(center[3])
results.append(result)
columns = ['seriesuid', 'coordX', 'coordY', 'coordZ', 'probability']
if label is not None:
columns.insert(4, 'class')
result_frame = pd.DataFrame(data=results, columns=columns)
result_frame.to_csv(result_path + '/result.csv',
index=False,
float_format='%f')
np.save(result_path + '/result.npy', np.array(results))
print('Overall Detection Done')
from toolbox import Nodule_Detection as nd
from toolbox import Evaluations as eva
try:
from tqdm import tqdm # long waits are not fun
except:
print('tqdm not installed')
tqdm = lambda x: x
'''
ENVIRONMENT_FILE = "./constants.txt"
IMG_WIDTH, IMG_HEIGHT, NUM_VIEW, MAX_BOUND, MIN_BOUND, PIXEL_MEAN = mt.read_environment(ENVIRONMENT_FILE)
WINDOW_SIZE = min(IMG_WIDTH, IMG_HEIGHT)
NUM_CHANNELS = 3
'''
constants = bt.read_constants("./constants.txt")
# REGION_SIZE = constants["REGION_SIZE"]
MAX_BOUND = float(constants["MAX_BOUND"])
MIN_BOUND = float(constants["MIN_BOUND"])
PIXEL_MEAN = constants["PIXEL_MEAN"]
REGION_SIZE = 40
CANDIDATE_BATCH = 10
RESULT_VISION = False
def precise_detection(volume, candidate_coords, candidate_labels, sess, input_tensor, output_tensor):
region_size = np.array([REGION_SIZE, REGION_SIZE, REGION_SIZE], dtype=int)
region_prehalf = np.int_(region_size / 2)
volume_padded = MAX_BOUND * np.ones(
(volume.shape[0] + region_size[0], volume.shape[1] + region_size[1], volume.shape[2] + region_size[2]),
dtype=int)
def detection(**kwargs):
opt.parse(kwargs)
test_paths = ["/data/fyl/datasets/Tianchi_Lung_Disease/train"]
#test_sample_filelist = "/data/fyl/models_pytorch/DensecropNet_detection_test_rfold1/filelist_val_fold0.log"
#net_file = "/data/fyl/models_pytorch/DensecropNet_stripe_detection_rfold1/DensecropNet_stripe_detection_rfold1_epoch27"
annotation_file = "/data/fyl/datasets/Tianchi_Lung_Disease/chestCT_round1_annotation.csv"
#candidate_file = "/data/fyl/datasets/Tianchi_Lung_Disease/candidate.csv"
evaluation_path = "./experiments_dt/evaluations_tianchild_densecropnet_nodule_rfold1"
#evaluation_path = "experiments_dt/evaluations_test"
#vision_path = evaluation_path
result_file = evaluation_path + "/result.csv"
hard_negatives_file = evaluation_path + "/hard_negatives.csv"
region_size = opt.input_size
batch_size = opt.batch_size
label_dict = {
'noduleclass': 1,
'stripeclass': 5,
'arterioclass': 31,
'lymphnodecalclass': 32
}
label = label_dict[opt.label_mode]
use_gpu = opt.use_gpu
net_file = opt.load_model_path
if 'vision_path' in dir() and vision_path is not None and not os.access(
vision_path, os.F_OK):
os.makedirs(vision_path)
#if os.access(evaluation_path, os.F_OK): shutil.rmtree(evaluation_path)
if not os.access(evaluation_path, os.F_OK): os.makedirs(evaluation_path)
if "test_paths" in dir():
all_patients = []
for path in test_paths:
all_patients += glob(path + "/*.mhd")
if len(all_patients) <= 0:
print("No patient found")
exit()
else:
print("No test data")
exit()
if hasattr(opt, 'filelists') and 'test' in opt.filelists.keys():
test_samples = bt.filelist_load(opt.filelists['test'])
test_uids = []
for test_sample in test_samples:
sample_uid = os.path.basename(test_sample).split('_')[0]
if sample_uid not in test_uids:
test_uids.append(sample_uid)
pd.DataFrame(data=test_uids,
columns=['series_uid'
]).to_csv(result_path + '/patients_uid.csv',
index=False)
#else:
# for path in opt.filelists['test']:
# test_samples = glob(path + '/*.mhd')
#model = models.DensecropNet(input_size=region_size, drop_rate=0, growth_rate=64, num_blocks=4, num_fin_growth=3).eval()
model = getattr(models, opt.model)(input_size=region_size,
**opt.model_setup).eval()
if net_file is not None:
model.load(net_file)
print('model loaded from %s' % (net_file))
shutil.copyfile(net_file,
evaluation_path + '/' + net_file.split('/')[-1])
#model.eval()
if use_gpu: model.cuda()
start_time = time.time()
#patient_evaluations = open(evaluation_path + "/patient_evaluations.log", "w")
results = []
CPMs = []
CPMs2 = []
hard_negatives = []
test_patients = all_patients
#random.shuffle(test_patients)
bt.filelist_store(test_patients, evaluation_path + "/patientfilelist.log")
for p in range(len(test_patients)):
patient = test_patients[p]
#patient = "./LUNA16/subset9/1.3.6.1.4.1.14519.5.2.1.6279.6001.212608679077007918190529579976.mhd"
#patient = "./LUNA16/subset9/1.3.6.1.4.1.14519.5.2.1.6279.6001.102681962408431413578140925249.mhd"
#patient = "./TIANCHI_examples/LKDS-00005.mhd"
uid = mt.get_mhd_uid(patient)
if 'test_uids' in dir() and uid not in test_uids:
print("%d/%d patient %s not belong to test set" %
(p + 1, len(test_patients), uid))
continue
#if uid!='656867':
# continue
print('%d/%d processing patient:%s' % (p + 1, len(test_patients), uid))
full_image_info = sitk.ReadImage(patient)
full_scan = sitk.GetArrayFromImage(full_image_info)
origin = np.array(full_image_info.GetOrigin(
))[::-1] #the order of origin and old_spacing is initially [z,y,x]
old_spacing = np.array(full_image_info.GetSpacing())[::-1]
image, new_spacing = mt.resample(full_scan, old_spacing,
np.array([1, 1, 1]))
#image = np.load(patient)
#new_spacing = np.array([1, 1, 1])
#origin = np.array([0, 0, 0])
print('Resample Done. time:{}s'.format(time.time() - start_time))
if 'annotation_file' in dir():
annotations = mt.get_challenge_annotations(uid, annotation_file,
label)
if len(annotations) == 0:
print("%d/%d patient %s has no annotations, ignore it." %
(p + 1, len(test_patients), uid))
#patient_evaluations.write('%d/%d patient %s has no annotations, ignore it\n' %(p+1, len(test_patients), uid))
continue
#make a real lesion visualization
if 'vision_path' in dir() and vision_path is not None:
real_lesions = []
for annotation in annotations:
#real_lesion = np.int_([abs(annotation[2]-origin[0])/new_spacing[0], abs(annotation[1]-origin[1])/new_spacing[1], abs(annotation[0]-origin[2])/new_spacing[2]])
real_lesion = mt.coord_conversion(annotation[:3][::-1],
origin,
old_spacing,
full_scan.shape,
image.shape,
dir_array=True)
real_lesions.append(real_lesion)
annotation_vision = cvm.view_coordinates(image,
real_lesions,
window_size=10,
reverse=False,
slicewise=False,
show=False)
np.save(vision_path + "/" + uid + "_annotations.npy",
annotation_vision)
if 'candidate_file' in dir():
print('Detection with given candidates:{}'.format(candidate_file))
candidate_coords = nd.luna_candidate(image,
uid,
origin,
new_spacing,
candidate_file,
lung_segment=True,
vision_path=vision_path)
if 'vision_path' in dir() and vision_path is not None:
volume_candidate = cvm.view_coordinates(image,
candidate_coords,
window_size=10,
reverse=False,
slicewise=False,
show=False)
np.save(vision_path + "/" + uid + "_candidate.npy",
volume_candidate)
print('Candidate Done. time:{}s'.format(time.time() - start_time))
print('candidate number:%d' % (len(candidate_coords)))
candidate_predictions = nd.precise_detection_pt(
image,
region_size,
candidate_coords,
model,
None,
batch_size,
use_gpu=use_gpu,
prediction_threshold=0.4)
positive_predictions = candidate_predictions > 0
predicted_coords = np.delete(
candidate_coords,
np.logical_not(positive_predictions).nonzero()[0],
axis=0)
predictions = candidate_predictions[positive_predictions]
lesion_center_predictions = nd.prediction_combine(
predicted_coords, predictions)
if 'vision_path' in dir() and vision_path is not None:
volume_predicted = cvm.view_coordinates(image,
predicted_coords,
window_size=10,
reverse=False,
slicewise=False,
show=False)
np.save(vision_path + "/" + uid + "_predicted.npy",
volume_predicted)
lesions = []
for nc in range(len(lesion_center_predictions)):
lesions.append(np.int_(lesion_center_predictions[nc][0:3]))
volume_prediction = cvm.view_coordinates(image,
lesions,
window_size=10,
reverse=False,
slicewise=False,
show=False)
np.save(vision_path + "/" + uid + "_prediction.npy",
volume_prediction)
else:
print('Detection with slic candidates')
candidate_results = nd.slic_candidate(image, 30, focus_area='lung')
if candidate_results is None:
continue
candidate_coords, candidate_labels, cluster_labels = candidate_results
if 'vision_path' in dir() and vision_path is not None:
np.save(vision_path + "/" + uid + "_segmask.npy",
cluster_labels)
#segresult = lc.segment_vision(image, cluster_labels)
#np.save(vision_path + "/" + uid + "_segresult.npy", segresult)
print('Candidate Done. time:{}s'.format(time.time() - start_time))
print('candidate number:%d' % (len(candidate_coords)))
candidate_predictions = nd.precise_detection_pt(
image,
region_size,
candidate_coords,
model,
None,
batch_size,
use_gpu=use_gpu,
prediction_threshold=0.4)
positive_predictions = candidate_predictions > 0
result_predictions, result_labels = nd.predictions_map_fast(
cluster_labels, candidate_predictions[positive_predictions],
candidate_labels[positive_predictions])
if 'vision_path' in dir() and vision_path is not None:
np.save(vision_path + "/" + uid + "_detlabels.npy",
result_labels)
np.save(vision_path + "/" + uid + "_detpredictions.npy",
result_predictions)
#detresult = lc.segment_vision(image, result_labels)
#np.save(vision_path+"/"+uid+"_detresult.npy", detresult)
lesion_center_predictions = nd.prediction_centering_fast(
result_predictions)
#lesion_center_predictions, prediction_labels = nd.prediction_cluster(result_predictions)
if 'vision_path' in dir() and vision_path is not None:
lesions = []
for nc in range(len(lesion_center_predictions)):
lesions.append(np.int_(lesion_center_predictions[nc][0:3]))
volume_predicted = cvm.view_coordinates(result_predictions *
1000,
lesions,
window_size=10,
reverse=False,
slicewise=False,
show=False)
np.save(vision_path + "/" + uid + "_prediction.npy",
volume_predicted)
if 'prediction_labels' in dir():
prediction_cluster_vision = lc.segment_color_vision(
prediction_labels)
np.save(
vision_path + "/" + uid + "_prediction_clusters.npy",
prediction_cluster_vision)
print('Detection Done. time:{}s'.format(time.time() - start_time))
'''
#randomly create a result for testing
lesion_center_predictions = []
for nc in range(10):
lesion_center_predictions.append([random.randint(0,image.shape[0]-1), random.randint(0,image.shape[1]-1), random.randint(0,image.shape[2]-1), random.random()])
'''
if len(lesion_center_predictions) < 1000:
print('Nodule coordinations:')
if len(lesion_center_predictions) <= 0:
print('none')
for nc in range(len(lesion_center_predictions)):
print('{} {} {} {}'.format(lesion_center_predictions[nc][0],
lesion_center_predictions[nc][1],
lesion_center_predictions[nc][2],
lesion_center_predictions[nc][3]))
for nc in range(len(lesion_center_predictions)):
#the output coordination order is [x,y,z], while the order for volume image should be [z,y,x]
result = [uid]
result.extend(
mt.coord_conversion(lesion_center_predictions[nc][:3],
origin,
old_spacing,
full_scan.shape,
image.shape,
dir_array=False)[::-1])
if label is not None: result.append(label)
result.append(lesion_center_predictions[nc][3])
results.append(result)
#results.append([uid, (lesion_center_predictions[nc][2]*new_spacing[2])+origin[2], (lesion_center_predictions[nc][1]*new_spacing[1])+origin[1], (lesion_center_predictions[nc][0]*new_spacing[0])+origin[0], lesion_center_predictions[nc][3]])
#if len(lesion_center_predictions)<1000:
#print('{} {} {} {}' .format(lesion_center_predictions[nc][0], lesion_center_predictions[nc][1], lesion_center_predictions[nc][2], lesion_center_predictions[nc][3]))
columns = ['seriesuid', 'coordX', 'coordY', 'coordZ', 'probability']
if label is not None:
columns.insert(4, 'class')
result_frame = pd.DataFrame(data=results, columns=columns)
result_frame.to_csv(result_file, index=False, float_format='%.4f')
np.save(evaluation_path + '/result.npy', np.array(results))
if 'annotation_file' in dir():
assessment = eva.detection_assessment(results,
annotation_file,
label=label)
if assessment is None:
print('assessment failed')
#patient_evaluations.write('%d/%d patient %s assessment failed\n' %(p+1, len(test_patients), uid))
continue
#num_scans, FPsperscan, sensitivities, CPMscore, FPsperscan2, sensitivities2, CPMscore2, lesions_detected = assessment
num_scans = assessment['num_scans']
FPsperscan, sensitivities = assessment['FROC']
CPMscore = assessment['CPM']
prediction_order = assessment['prediction_order']
lesions_detected = assessment['detection_cites']
if len(FPsperscan) <= 0 or len(sensitivities) <= 0:
print("No results to evaluate, continue")
else:
eva.evaluation_vision(CPMs,
num_scans,
FPsperscan,
sensitivities,
CPMscore,
lesions_detected,
output_path=evaluation_path)
#patient_evaluations.write('%d/%d patient %s CPM score:%f\n' %(p+1, len(test_patients), uid, single_assessment[6]))
print('Evaluation Done. time:{}s'.format(time.time() - start_time))
num_positive = (lesions_detected >= 0).nonzero()[0].size
for ndi in range(len(lesions_detected)):
if results[prediction_order[ndi]][-1] <= 0.5 or (
lesions_detected[:ndi] >=
0).nonzero()[0].size == num_positive:
break
if lesions_detected[ndi] == -1:
hard_negatives.append(results[prediction_order[ndi]])
hard_negatives_frame = pd.DataFrame(data=hard_negatives,
columns=columns)
hard_negatives_frame.to_csv(hard_negatives_file,
index=False,
float_format='%.4f')
print('Hard Negatives Extracted. time:{}s'.format(time.time() -
start_time))
print('Overall Detection Done')
#net_init_name = "lidc_3D_multi_crop_net_64_aug3" if os.access(net_store_path, os.F_OK): shutil.rmtree(net_store_path) os.makedirs(net_store_path) if os.access(tensorboard_path, os.F_OK): shutil.rmtree(tensorboard_path) #data arrangement if 'net_init_name' in dir(): net_init_path = load_path + "/" + net_init_name net_init_file = net_init_path + "/epoch12/epoch12" filelist_train_path = net_init_path + "/filelist_train.log" filelist_val_path = net_init_path + "/filelist_val.log" filelist_test_path = net_init_path + "/filelist_test.log" train_files = bt.filelist_load(filelist_train_path) val_files = bt.filelist_load(filelist_val_path) test_files = bt.filelist_load(filelist_test_path) else: lidc_dir="../data_samples/lidc_cubes_64_overbound_ipris" files_lists, _ = lt.filelist_training(lidc_dir, shuffle=True, cross_fold=5, test_fold=5) train_files = files_lists['train'] val_files = files_lists['val'] test_files = files_lists['test'] bt.filelist_store(train_files, net_store_path + '/' + 'filelist_train.log') bt.filelist_store(val_files, net_store_path + '/' + 'filelist_val.log') bt.filelist_store(test_files, net_store_path + '/' + 'filelist_test.log') train_num = len(train_files) val_num = len(val_files) #train_num = 32
def get_filelists_patientwise(patient_uids=None,
filelists=None,
fileext='npy',
datasetidx='',
config={}):
save_path = config['model_root'] + '/' + config['env']
if filelists is None:
filelistdict = {}
filelistdict['train'] = bt.filelist_load(
config['filelists' + str(datasetidx)]['train'])
filelistdict['val'] = bt.filelist_load(config['filelists' +
str(datasetidx)]['val'])
bt.filelist_store(filelistdict['train'],
save_path + '/filelist_train.log')
bt.filelist_store(filelistdict['val'], save_path + '/filelist_val.log')
if 'test' in config['filelists' + str(datasetidx)].keys():
filelistdict['test'] = bt.filelist_load(
config['filelists' + str(datasetidx)]['test'])
bt.filelist_store(filelistdict['test'],
save_path + '/filelist_test.log')
print("filelist loaded")
else:
#filelists=["/home/fyl/datasets/luna_64/train", "/home/fyl/datasets/luna_64/test", "/home/fyl/datasets/npy_non_set"]
#filelist = []
if patient_uids is None:
patient_uids = []
for filelist in filelists:
files = glob.glob(filelist + '/*.' + fileext)
#filelist.extend(files)
for file in files:
filename = os.path.basename(file)
filenamenoext = os.path.splitext(filename)[0]
fileinfos = filenamenoext.split('_')
patient_uid = fileinfos[0]
if patient_uid not in patient_uids:
patient_uids.append(patient_uid)
elif type(patient_uids) == str:
patient_uids = bt.filelist_load(patient_uids)
if config['filelist_shuffle']: random.shuffle(patient_uids)
bt.filelist_store(
patient_uids,
save_path + '/patientlist' + str(datasetidx) + '.log')
#bt.filelist_store('luna16samplelist.log', filelist)
patient_folds, folddict = bt.foldlist(
patient_uids, config['num_cross_folds'], {
'val': config['val_fold' + str(datasetidx)],
'test': config['test_fold' + str(datasetidx)]
})
filelist_overall = []
filelistdict = {}
for setname in patient_folds.keys():
filelistdict[setname] = []
'''
print("filelist {} generating" .format(setname))
for patient_uid in tqdm(patient_folds[setname]):
for filelist in filelists:
if os.path.isfile(filelist):
files = bt.filelist_load(filelist)
else:
files = glob.glob(filelist+'/%s*.%s' %(patient_uid, fileext))
if 'lidc' in filelist and hasattr(opt, 'remove_uncertain') and opt.remove_uncertain:
files = lt.filelist_remove_uncertain(files)
filelist_overall.extend(files)
filelistdict[setname].extend(files)
bt.filelist_store(filelistdict[setname], save_path+'/filelist'+str(datasetidx)+'_'+setname+'_fold'+str(folddict[setname])+'.log')
'''
for filelist in filelists:
if os.path.isfile(filelist):
files = bt.filelist_load(filelist)
else:
files = glob.glob(filelist + '/*.%s' % (fileext))
filelist_overall.extend(files)
if 'lidc' in filelist and 'remove_uncertain' in config.keys(
) and config['remove_uncertain']:
filelist_overall = lt.filelist_remove_uncertain(
filelist_overall)
if config['filelist_shuffle']: random.shuffle(filelist_overall)
for file in filelist_overall:
patient_uid = os.path.basename(file).split('_')[0]
for setname in patient_folds.keys():
if patient_uid in patient_folds[setname]:
filelistdict[setname].append(file)
for setname in patient_folds.keys():
bt.filelist_store(
filelistdict[setname],
save_path + '/filelist' + str(datasetidx) + '_' + setname +
'_fold' + str(folddict[setname]) + '.log')
bt.filelist_store(filelist_overall,
save_path + '/filelist' + str(datasetidx) + '.log')
print("filelist generated")
return filelistdict
