-
Notifications
You must be signed in to change notification settings - Fork 14
/
utils.py
200 lines (181 loc) · 7.43 KB
/
utils.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
import numpy as np
from tqdm import tqdm
from glob import glob
import wfdb as wf
import scipy.signal
from sklearn.preprocessing import StandardScaler
import pickle
import torch
import torch.nn as nn
from torch.autograd import Variable
from torch.utils.data import DataLoader
from torch.utils.data import TensorDataset
from network import IncUNet
def obtain_data(args):
#path = '../../Datasets/ECG/mitdb/'
dat_path = args.datapath + '*.atr'
paths = glob(dat_path)
p2 = [path[:-4] for path in paths]
fs = wf.rdsamp(p2[0])[1]['fs']
windowed_data = []
windowed_beats = []
count = 0
count1 = 0
for path in tqdm(p2):
ann = wf.rdann(path,'atr')
record = wf.io.rdrecord(path)
beats = ann.sample
labels = ann.symbol
len_beats = len(beats)
data = record.p_signal[:,0]
if(args.dataset == 'mitdb' or args.dataset == 'nstdb'):
### Sampling rate in MITDB and NSTDB is 360. Therefore, for 10s data the window size is 3600 ###
if (path[-2:] == '_6' or (path[-2:] != str(args.db) and args.evaluate_nstdb)):#(path[-2:] == '_6' or path[-2:] == '12' or path[-2:] == '06' or path[-2:] == '18' or path[-2:] == '24'):
print('Skip')
continue
else:
ini_index = 0
final_index = 0
### Checking for Beat annotations
non_required_labels = ['[','!',']','x','(',')','p','t','u','`',"'",'^','|','~','+','s','T','*','D','=','"','@']
for window in range(len(data) // 3600):
count += 1
for r_peak in range(ini_index,len_beats):
if beats[r_peak] > (window+1) * 3600:
final_index = r_peak
#print('FInal index:',final_index)
break
record_anns = list(beats[ini_index: final_index])
record_labs = labels[ini_index: final_index]
to_del_index = []
for actual_lab in range(len(record_labs)):
for lab in range(len(non_required_labels)):
if(record_labs[actual_lab] == non_required_labels[lab]):
to_del_index.append(actual_lab)
print('To del Indices are:',to_del_index)
for indice in range(len(to_del_index)-1,-1,-1):
print(indice)
del record_anns[to_del_index[indice]]
windowed_beats.append(np.asarray(record_anns) - (window) * 3600)
windowed_data.append(data[window * 3600 : (window+1) * 3600])
ini_index = final_index
elif(args.dataset == 'mit_bih_Exercise_ST_change'):
ini_index = 0
final_index = 0
for window in range(len(data) // 3600):
count += 1
windowed_data.append(data[window * 3600 : (window+1) * 3600])
for r_peak in range(ini_index,len_beats):
if beats[r_peak] > (window+1) * 3600:
final_index = r_peak
#print('FInal index:',final_index)
break
windowed_beats.append(beats[ini_index: final_index] - (window) * 3600)
ini_index = final_index
### Scaling and Resampling
scaler = StandardScaler()
scaled_data = scaler.fit_transform(np.asarray(windowed_data).transpose()).transpose()
resampled_beat = []
for record in range(len(windowed_data)):
resampled_beat.append(scipy.signal.resample(scaled_data[record],5000))
patient_ecg = np.asarray(resampled_beat)
return patient_ecg,windowed_beats
def score(r_ref, r_ans, fs_, thr_):
for record in range(len(r_ref)):
r_ref[record] = r_ref[record][(r_ref[record] >= 0.5*fs_) & (r_ref[record] <= 9.5*fs_)]
all_TP = 0
all_FN = 0
all_FP = 0
failed_record = []
fp = []
fn = []
zer = []
HR_score = 0
record_flags = np.ones(len(r_ref))
failed_record = []
fp = []
fn = []
zer = []
for i in range(len(r_ref)):
FN = 0
FP = 0
TP = 0
# if math.isnan(hr_ans[i]):
# hr_ans[i] = 0
# hr_der = abs(int(hr_ans[i]) - int(hr_ref[i]))
# if hr_der <= 0.02 * hr_ref[i]:
# HR_score = HR_score + 1
# elif hr_der <= 0.05 * hr_ref[i]:
# HR_score = HR_score + 0.75
# elif hr_der <= 0.1 * hr_ref[i]:
# HR_score = HR_score + 0.5
# elif hr_der <= 0.2 * hr_ref[i]:
# HR_score = HR_score + 0.25
for j in range(len(r_ref[i])):
loc = np.where(np.abs(r_ans[i] - r_ref[i][j]) <= thr_*fs_)[0]
if j == 0:
err = np.where((r_ans[i] >= 0.5*fs_ + thr_*fs_) & (r_ans[i] <= r_ref[i][j] - thr_*fs_))[0]
elif j == len(r_ref[i])-1:
err = np.where((r_ans[i] >= r_ref[i][j]+thr_*fs_) & (r_ans[i] <= 9.5*fs_ - thr_*fs_))[0]
else:
err = np.where((r_ans[i] >= r_ref[i][j]+thr_*fs_) & (r_ans[i] <= r_ref[i][j+1]-thr_*fs_))[0]
FP = FP + len(err)
if len(loc) >= 1:
TP += 1
FP = FP + len(loc) - 1
elif len(loc) == 0:
FN += 1
if FN + FP > 1:
record_flags[i] = 0
elif FN == 1 and FP == 0:
record_flags[i] = 0.3
elif FN == 0 and FP == 1:
record_flags[i] = 0.7
##Custom
if(record_flags[i] != 1):
failed_record.append(i)
if(record_flags[i] == 0.7):
fp.append(i)
elif(record_flags[i] == 0.3):
fn.append(i)
elif(record_flags[i] == 0):
zer.append(i)
all_FP += FP
all_FN += FN
all_TP += TP
print("Failed_records:",failed_record)
print("FP's:",fp)
print("FN's:",fn)
print("Zeros's:",zer)
##
rec_acc = round(np.sum(record_flags) / len(r_ref), 4)
hr_acc = round(HR_score / len(r_ref), 4)
print( 'QRS_acc: {}'.format(rec_acc))
print('HR_acc: {}'.format(hr_acc))
print('Scoring complete.')
Recall = all_TP / (all_FN + all_TP)
Precision = all_TP / (all_FP + all_TP)
F1_score = 2 * Recall * Precision / (Recall + Precision)
print("TP's:{} FN's:{} FP's:{}".format(all_TP,all_FN,all_FP))
print("REcall:{}, Precision(FNR):{}, F1-Score:{}".format(Recall,Precision,F1_score))
return rec_acc,all_FP,all_FN,all_TP
def load_model_CNN(SAVED_MODEL_PATH,test_loader,device='cpu'):
C,H,W = 1,1,5000
loaded_model = IncUNet(in_shape=(C,H,W))
loaded_model.load_state_dict(torch.load(SAVED_MODEL_PATH, map_location = lambda storage, loc: storage, pickle_module=pickle))
loaded_model.to(device)
loaded_model.eval()
print("...........Evaluation..........")
loaded_model.eval()
### Need to change after this ###
net_test_loss = 0
y_pred = []
batch_length = 64
y_pred = []
with torch.no_grad():
for step,x in enumerate(test_loader):
print('Step = ',step)
x = Variable(x[0].to(device))
y_predict_test = loaded_model(x)
y_pred.append(y_predict_test[:,0,:])
return y_pred