import os
import os.path as osp
from src.tools.utils import get_logger
from tools.utils import mkdir_p
from src.tools.plot import ClipPlotSeriesWithBack
from src.tools.utils import read_file_contents_list, save_file_contents_list, get_logger
from src.tools.data_io import ScanWrapper
import numpy as np
import yaml
import pandas as pd
import matplotlib.pyplot as plt
import yaml
from tools.plot import mean_diff_plot, scatter_plot
logger = get_logger('Analyze Round FOV')
in_native_folder = '/nfs/masi/xuk9/SPORE/CAC_class/data/s14_ori_final_resample'
file_list_txt = '/nfs/masi/xuk9/SPORE/CAC_class/file_lists/result_temporal'
out_folder_axial_clip = '/nfs/masi/xuk9/SPORE/CAC_class/axial_clip_native'
mkdir_p(out_folder_axial_clip)
def axial_clip_plot_native():
file_name_list = read_file_contents_list(file_list_txt)
for file_name in file_name_list:
in_img_path = os.path.join(in_native_folder, file_name)
cliper_obj = ClipPlotSeriesWithBack(in_img_path, None, None, 10, 35,
15, 1, -3000, 1000, None, None,
None)
cliper_obj.clip_plot_img_only(out_folder_axial_clip)
trainloader = DataLoader(dataset=trainset, batch_size=args.batch_size, shuffle=True, **train_kws)
valiloader = DataLoader(dataset=valiset, batch_size=args.batch_size, shuffle=False, **test_kws)
testloader = DataLoader(dataset=testset, batch_size=args.batch_size, shuffle=False, **test_kws)
model = make_model(H=args.I_size, W=args.I_size, input_num=5, target_num=10, input_channel=1, d_channel=5, d_channel_ff=10, dropout=args.dropout) \
.to(device=args.device, dtype=args.value_dtype)
optimizer = torch.optim.Adam(model.parameters(), lr=args.lr, weight_decay=args.weight_decay)
lr_scheduler = torch.optim.lr_scheduler.ExponentialLR(optimizer, gamma=0.9)
loss_df = pd.DataFrame([],index=pd.Index(range(args.max_epochs), name='Epoch'), columns=['Train_loss', 'Vali_loss', 'Vali_infer_loss'])
log_file = os.path.join(args.result_folder, 'log.txt')
loss_file = os.path.join(args.result_folder, 'loss.csv')
logger = get_logger(log_file)
for epoch in range(args.max_epochs):
lr = optimizer.param_groups[0]['lr']
logger.debug('[{:s}] Epoch {:03d}, Learning rate: {}'.format(args.model, epoch+1, lr))
loss_df.iloc[epoch,0] = train_epoch(model, trainloader, optimizer, args, logger)
loss_df.iloc[epoch,1] = eval_epoch(model, valiloader, args, logger)
loss_df.iloc[epoch,2] = infer_epoch(model, valiloader, args, logger)
if (epoch+1) > 10:
lr_scheduler.gamma = 0.95
lr_scheduler.step()
if (epoch+1) % 10 == 0:
save_model(epoch, optimizer, model, args)
from src.tools.paral_average import AverageValidRegion
from src.tools.data_io import ScanWrapper, DataFolder
import argparse
import yaml
import pandas as pd
import os.path as osp
import os
import numpy as np
from src.tools.utils import get_logger, read_file_contents_list, mkdir_p
logger = get_logger('grad_cam_analysis')
affine_img_dir = '/nfs/masi/xuk9/SPORE/CAC_class/data/affine/s2_no_nan'
file_list_txt = '/nfs/masi/xuk9/SPORE/CAC_class/file_lists/complete_list'
out_folder = '/nfs/masi/xuk9/SPORE/CAC_class/data/affine/s2_no_nan_average'
mkdir_p(out_folder)
def main():
parser = argparse.ArgumentParser()
parser.add_argument('--yaml-config',
type=str,
default='simg_bmi_regression_0_cam.yaml')
args = parser.parse_args()
file_list = read_file_contents_list(file_list_txt)
# folder_obj = DataFolder(ori_img_dir, file_list)
folder_obj = DataFolder(affine_img_dir, file_list)
ave_obj = AverageValidRegion(folder_obj, 10)
import pandas as pd
import numpy as np
import argparse
import os
import os.path as osp
from src.tools.utils import get_logger, read_file_contents_list
import yaml
from sklearn.metrics import mean_squared_error
from math import sqrt
import statsmodels.api as sm
import matplotlib.pyplot as plt
from sklearn.metrics import r2_score
from tools.utils import mkdir_p
logger = get_logger('Regression analysis')
lower_bound = 12
upper_bound = 50
def mean_diff_plot(pred_list, gt_list, rmse_list, out_png):
f, ax = plt.subplots(figsize=(8, 6))
sm.graphics.mean_diff_plot(pred_list, gt_list, ax=ax)
ax.set_title(
f'RMSE: {np.mean(rmse_list):.4f}, R2: {r2_score(gt_list, pred_list):.4f}'
)
logger.info(f'Save png to {out_png}')
plt.savefig(out_png, bbox_inches='tight', pad_inches=0.1)
plt.close()
import os
import os.path as osp
from src.tools.utils import get_logger
from tools.utils import mkdir_p
from src.tools.plot import ClipPlotSeriesWithBack
from src.tools.utils import read_file_contents_list, save_file_contents_list, get_logger
from src.tools.data_io import ScanWrapper
import numpy as np
import yaml
import pandas as pd
import matplotlib.pyplot as plt
import yaml
from tools.plot import mean_diff_plot, scatter_plot
logger = get_logger('Analyze missing ROI')
in_body_mask = '/nfs/masi/xuk9/SPORE/CAC_class/data/atlas/atlas_body_mask/body_seg_resampled.nii.gz'
body_mask_overlap_roi_folder = '/nfs/masi/xuk9/SPORE/CAC_class/data/atlas/valid_region/s6_body_mask_intersect'
file_list_txt = '/nfs/masi/xuk9/SPORE/CAC_class/file_lists/result_temporal'
out_analyze_missing_roi_folder = '/nfs/masi/xuk9/SPORE/CAC_class/clinical/missing_roi'
mkdir_p(out_analyze_missing_roi_folder)
out_missing_ratio_csv = os.path.join(out_analyze_missing_roi_folder,
'missing_ratio.csv')
def _get_sess_missing_ratio(file_name):
body_mask_data = ScanWrapper(in_body_mask).get_data()
in_mask_data = ScanWrapper(
import numpy as np
import argparse
import os
import os.path as osp
from src.tools.utils import get_logger
import yaml
from sklearn.metrics import mean_squared_error
from math import sqrt
import statsmodels.api as sm
import matplotlib.pyplot as plt
from sklearn.metrics import r2_score
from tools.utils import mkdir_p
from src.tools.clinical import ClinicalDataReaderSPORE
from tools.utils import read_file_contents_list, save_file_contents_list
logger = get_logger('Run temporary consistency test')
in_file_list_txt = '/nfs/masi/xuk9/SPORE/CAC_class/file_lists/complete_list'
in_csv_file = '/nfs/masi/xuk9/SPORE/CAC_class/clinical/label_full_combined.csv'
in_raw_label_file_xlsx = '/nfs/masi/xuk9/SPORE/clustering/registration/20200512_corrField/male/clinical/label.xlsx'
out_height_weight_added_csv = '/nfs/masi/xuk9/SPORE/CAC_class/clinical/label_add_H_W.csv'
valid_bmi_file_list = '/nfs/masi/xuk9/SPORE/CAC_class/file_lists/bmi_valid'
out_diff_hist_png = '/nfs/masi/xuk9/SPORE/CAC_class/clinical/diff_hist.png'
out_long_exclude_list = '/nfs/masi/xuk9/SPORE/CAC_class/clinical/bmi_exclude_long_sess_list.txt'
out_include_bmi_list = '/nfs/masi/xuk9/SPORE/CAC_class/clinical/bmi_include_list.txt'
out_exclude_bmi_list = '/nfs/masi/xuk9/SPORE/CAC_class/clinical/bmi_exclude_list.txt'
out_inconsistency_list = '/nfs/masi/xuk9/SPORE/CAC_class/clinical/bmi_inconsistent_list.txt'
out_inconsist_subj_data = '/nfs/masi/xuk9/SPORE/CAC_class/clinical/bmi_inconsistent_subj_data.txt'