-
Notifications
You must be signed in to change notification settings - Fork 0
/
utils.py
798 lines (675 loc) · 29.4 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
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
from IPython.display import display
import matplotlib.pyplot as plt
import numpy as np
import os
import json
import pywt
import wfdb
from wfdb import processing
from scipy.signal import argrelextrema
from scipy.signal import savgol_filter
sampling_rate = 250 # sampling rate is 250Hz
k = int(np.ceil(np.log2(sampling_rate))) # decomposition level
ROOT_PATH = '/Users/sumuzhao/ETHZ/Semester_project_2/'
def generate_atr_list(atr_path, is_edb=True):
"""generate the effective annotation file list"""
atr_list = []
for dirpath, _, filenames in os.walk(atr_path):
for filename in filenames:
basename = os.path.splitext(filename)[0]
if is_edb:
if filename.endswith('.atr') and basename not in ['e0133', 'e0155', 'e0509', 'e0611', 'e0163',
'e0405', 'e0409', 'e0704']:
atr_list.append(basename)
else:
if filename.endswith('.atr'):
atr_list.append(basename)
atr_list.sort(key=lambda x: int(os.path.splitext(x)[0][1:]))
return atr_list
def pick_s_symbols(atr_name, is_edb=True, same_st_episodes='stb'):
"""pick up all the effective 's' symbols"""
if is_edb:
ann = wfdb.rdann('./edb_records/{}'.format(atr_name), 'atr')
symbol_list = ann.symbol
sample_list = ann.sample
aux_list = ann.aux_note
sym_sam_list = list(zip(sample_list, aux_list))
st_idx = []
for j in range(len(symbol_list)):
if symbol_list[j] == 's':
st_idx.append(sym_sam_list[j])
else:
ann = wfdb.rdann('./ltst_records/{}'.format(atr_name), same_st_episodes)
symbol_list = ann.symbol
sample_list = ann.sample
aux_list = ann.aux_note
sym_sam_list = list(zip(sample_list, aux_list))
st_idx = []
for j in range(len(symbol_list)):
if symbol_list[j] == 's':
st_idx.append(sym_sam_list[j])
return st_idx
def pick_st_pairs(st_idx, is_edb=True):
"""pick up all the 'ST' symbol pairs for each episodes"""
if is_edb:
dict = {'ST0+': [], 'ST1+': [], 'ST0-': [], 'ST1-': []}
for i in range(len(st_idx) - 1):
for j in range(i + 1, len(st_idx)):
if st_idx[i][1] == '(ST0+\x00' and st_idx[j][1] == 'ST0+)\x00':
dict['ST0+'].append((int(st_idx[i][0]), int(st_idx[j][0])))
break
elif st_idx[i][1] == '(ST1+\x00' and st_idx[j][1] == 'ST1+)\x00':
dict['ST1+'].append((int(st_idx[i][0]), int(st_idx[j][0])))
break
elif st_idx[i][1] == '(ST0-\x00' and st_idx[j][1] == 'ST0-)\x00':
dict['ST0-'].append((int(st_idx[i][0]), int(st_idx[j][0])))
break
elif st_idx[i][1] == '(ST1-\x00' and st_idx[j][1] == 'ST1-)\x00':
dict['ST1-'].append((int(st_idx[i][0]), int(st_idx[j][0])))
break
else:
dict = {'ST0': [], 'ST1': [], 'ST2': []}
for i in range(len(st_idx) - 1):
for j in range(i + 1, len(st_idx)):
if (st_idx[i][1][:4] == '(st0' and st_idx[j][1][:3] == 'st0') or \
(st_idx[i][1][:6] == '(rtst0' and st_idx[j][1][:5] == 'rtst0'):
dict['ST0'].append((int(st_idx[i][0]), int(st_idx[j][0])))
break
elif (st_idx[i][1][:4] == '(st1' and st_idx[j][1][:3] == 'st1') or \
(st_idx[i][1][:6] == '(rtst1' and st_idx[j][1][:5] == 'rtst1'):
dict['ST1'].append((int(st_idx[i][0]), int(st_idx[j][0])))
break
elif (st_idx[i][1][:4] == '(st2' and st_idx[j][1][:3] == 'st2') or \
(st_idx[i][1][:6] == '(rtst2' and st_idx[j][1][:5] == 'rtst2'):
dict['ST2'].append((int(st_idx[i][0]), int(st_idx[j][0])))
break
for key in list(dict.keys()):
if not dict[key]:
dict.pop(key)
return dict
def create_st_dictionary(record_path='./edb_records', is_edb=True, same_st_episodes='stb'):
"""
create a dictionary for all the records
format: is_edb=True,
{'e0103':
{'ST0+': [(start, end), ...],
'ST1+': [(start, end), ...],
'ST0-': [(start, end), ...],
'ST1-': [(start, end), ...]},
'e0104':
{},
...
}
is_edb=False
{'s20011':
{'ST0': [(start, end), ...],
'ST1': [(start, end), ...],
'ST2': [(start, end), ...]},
's20011':
{},
...
}
"""
dict = {}
atr_list = generate_atr_list(record_path, is_edb)
for i in range(len(atr_list)):
st_idx = pick_s_symbols(atr_list[i], is_edb, same_st_episodes)
st_dict= pick_st_pairs(st_idx, is_edb)
print(i, st_dict)
if st_dict:
dict[atr_list[i]] = st_dict
else:
print("Drop empty record {}".format(atr_list[i]))
return atr_list, dict
def st_histogram(atr_list, st_dict):
st0_p = 0
st0_n = 0
st1_p = 0
st1_n = 0
for atr in atr_list:
for key in list(st_dict[atr].keys()):
if key == 'ST0+':
st0_p += 1
elif key == 'ST0-':
st0_n += 1
elif key == 'ST1+':
st1_p += 1
elif key == 'ST1-':
st1_n += 1
plt.bar(range(4), [st0_p, st0_n, st1_p, st1_n], color='lightsteelblue')
plt.xlabel("ST episode categories")
plt.ylabel("Number of ST episodes")
for x, y in zip(range(4), [st0_p, st0_n, st1_p, st1_n]):
plt.text(x, y, y, ha='center', va='bottom')
plt.xticks(range(4), ['ST0+', 'ST0-', 'ST1+', 'ST1-'])
plt.show()
def smooth(y, box_pts):
"""smooth function"""
box = np.ones(box_pts)/box_pts
y_smooth = np.convolve(y, box, mode='same')
return y_smooth
def get_normal_part(atr_name, dict, data_path='./edb_records/'):
"""
get the normal parts for each record (eliminate the ST episodes)
return normal_dict: dict{}
format: {'N0': [np.array()],
'N1': [np.array()],
'N2': [np.array()]}
"""
normal_signal_dict = {'N0': [], 'N1': [], 'N2': []}
parse_list = parse_st_sub_dict(atr_name, dict)
signal, _ = wfdb.rdsamp(data_path + atr_name)
st0_length = 0
st1_length = 0
st2_length = 0
for pl in parse_list:
part_list = dict[atr_name][pl[0]]
for part in part_list:
if pl[1] == 0:
st0_length += part[1] - part[0]
elif pl[1] == 1:
st1_length += part[1] - part[0]
else:
st2_length += part[1] - part[0]
signal[part[0] - 1000:part[1] + 1000, pl[1]] = 0
signal0 = signal[:, 0]
signal1 = signal[:, 1]
signal0 = signal0[signal0 != 0]
signal1 = signal1[signal1 != 0]
if st0_length != 0:
normal_signal_dict['N0'].append(signal0[0:st0_length])
if st1_length != 0:
normal_signal_dict['N1'].append(signal1[0:st1_length])
if atr_name[1] == '3':
signal2 = signal[:, 2]
signal2 = signal2[signal2 != 0]
if st2_length != 0:
normal_signal_dict['N2'].append(signal2[0:st2_length])
return normal_signal_dict
def load_st_signal(atr_name, dict, data_path='./edb_records/', is_edb=True):
"""
load st episodes signals
return st_signal_dict: dict{}
format: {'ST0-': [np.array()],
'ST1-': [np.array()],
'ST0+': [np.array()],
'ST1+': [np.array()],
'ST0': [np.array()],
'ST1': [np.array()]}
"""
if is_edb:
st_signal_dict = {'ST0-': [], 'ST1-': [], 'ST0+': [], 'ST1+': []}
else:
st_signal_dict = {'ST0': [], 'ST1': [], 'ST2': []}
parse_list = parse_st_sub_dict(atr_name, dict)
for pl in parse_list:
if dict[atr_name][pl[0]]:
for st_pair in dict[atr_name][pl[0]]:
st_signal, _ = wfdb.rdsamp(data_path + atr_name, sampfrom=st_pair[0], sampto=st_pair[1])
st_signal_dict[pl[0]].append(st_signal[:, pl[1]])
signal_concat = st_signal_dict[pl[0]][0]
for i in st_signal_dict[pl[0]][1:]:
signal_concat = np.concatenate((signal_concat, i))
if signal_concat.shape[0] % 2 == 1:
signal_concat = signal_concat[0:signal_concat.shape[0] - 1]
st_signal_dict[pl[0]] = [signal_concat]
return st_signal_dict
def parse_st_sub_dict(atr_name, dict):
"""
parse each record, return signal channel and whether it is elevation or depression
return parse_list: [(st episode category, signal channel, elevation or depression)]
"""
parse_list = []
sub_dict = dict[atr_name]
key = list(sub_dict.keys())
for k in key:
if k in ['ST0+', 'ST0-', 'ST0']:
channel = 0
elif k in ['ST1+', 'ST1-', 'ST1']:
channel = 1
else:
channel = 2
if k in ['ST0+', 'ST1+']:
is_elevation = True
else:
is_elevation = False
parse_list.append((k, channel, is_elevation))
return parse_list
def remove_baseline(signal, wavelet='bior2.6', is_plot=False):
"""
Removal of baseline wandering using wavelet
wavelet: bior2.6
level: 8
"""
A8, D8, D7, D6, D5, D4, D3, D2, D1 = pywt.wavedec(signal, wavelet=pywt.Wavelet(wavelet), level=8)
A8 = np.zeros_like(A8[0]) # low frequency info
RA7 = pywt.idwt(A8, D8[0], wavelet)
RA6 = pywt.idwt(RA7[0:len(D7[0])], D7[0], wavelet)
RA5 = pywt.idwt(RA6[0:len(D6[0])], D6[0], wavelet)
RA4 = pywt.idwt(RA5[0:len(D5[0])], D5[0], wavelet)
RA3 = pywt.idwt(RA4[0:len(D4[0])], D4[0], wavelet)
RA2 = pywt.idwt(RA3[0:len(D3[0])], D3[0], wavelet)
D2 = np.zeros_like(D2[0]) # high frequency noise
RA1 = pywt.idwt(RA2[0:len(D2)], D2, wavelet)
D1 = np.zeros_like(D1[0]) # high frequency noise
DenoisingSignal = pywt.idwt(RA1[0:len(D1)], D1, wavelet)
if is_plot:
plt.plot(signal[0], 'b')
plt.plot(DenoisingSignal, 'g')
plt.show()
return DenoisingSignal
def binary_spline_wavelet_filter(swa, swd, signal, points, level=4, is_plot=False):
"""
Binary spline wavelet filtering, level: 4,
low-pass filter:[1/4 3/4 3/4 1/4], high-pass filter:[-1/4 -3/4 3/4 1/4]
"""
for i in range(0, points - 3):
swa[0, i + 3] = 1 / 4 * signal[i + 3 - 2 ** 0 * 0] + \
3 / 4 * signal[i + 3 - 2 ** 0 * 1] + \
3 / 4 * signal[i + 3 - 2 ** 0 * 2] + \
1 / 4 * signal[i + 3 - 2 ** 0 * 3]
swd[0, i + 3] = - 1 / 4 * signal[i + 3 - 2 ** 0 * 0] - \
3 / 4 * signal[i + 3 - 2 ** 0 * 1] + \
3 / 4 * signal[i + 3 - 2 ** 0 * 2] + \
1 / 4 * signal[i + 3 - 2 ** 0 * 3]
for j in range(1, level):
for i in range(0, points - 24):
swa[j, i + 24] = 1 / 4 * swa[j - 1, i + 24 - 2 ** (j - 1) * 0] \
+ 3 / 4 * swa[j - 1, i + 24 - 2 ** (j - 1) * 1] \
+ 3 / 4 * swa[j - 1, i + 24 - 2 ** (j - 1) * 2] \
+ 1 / 4 * swa[j - 1, i + 24 - 2 ** (j - 1) * 3]
swd[j, i + 24] = - 1 / 4 * swa[j - 1, i + 24 - 2 ** (j - 1) * 0] \
- 3 / 4 * swa[j - 1, i + 24 - 2 ** (j - 1) * 1] \
+ 3 / 4 * swa[j - 1, i + 24 - 2 ** (j - 1) * 2] \
+ 1 / 4 * swa[j - 1, i + 24 - 2 ** (j - 1) * 3]
if is_plot:
# draw the original signal and all the approximation and details coefficients.
plt.figure(1)
ax1 = plt.subplot2grid((5, 2), (0, 0), colspan=2)
ax1.plot(signal)
ax1.set_title("Original signal")
for i in range(level):
ax_a = plt.subplot2grid((5, 2), (i + 1, 0))
ax_a.plot(swa[i, :])
ax_d = plt.subplot2grid((5, 2), (i + 1, 1))
ax_d.plot(swd[i, :])
plt.show()
return swa, swd
def get_maximal_value_pairs(swd, points, level, is_plot=False):
"""calculate the positive and negative maximal value pairs"""
pos_idx = np.zeros_like(swd, dtype=np.int32)
neg_idx = np.zeros_like(swd, dtype=np.int32)
pos_w = swd * (swd > 0)
pos_dw = ((pos_w[:, 0:points - 1] - pos_w[:, 1:points]) < 0)
pos_idx[:, 1:points - 1] = ((pos_dw[:, 0:points - 2] - pos_dw[:, 1:points - 1]) > 0)
neg_w = swd * (swd < 0)
neg_dw = ((neg_w[:, 0:points - 1] - neg_w[:, 1:points]) > 0)
neg_idx[:, 1:points - 1] = ((neg_dw[:, 0:points - 2] - neg_dw[:, 1:points - 1]) > 0)
pos_neg_idx = np.bitwise_or(pos_idx, neg_idx)
pos_neg_idx[:, 0] = 1
pos_neg_idx[:, points - 1] = 1
w_peak = pos_neg_idx * swd
w_peak[:, 0] += 1e-10
w_peak[:, points - 1] += 1e-10
if is_plot:
# ECG信号在j=1,2,3,4尺度下的小波系数的模极大值点
plt.figure(2)
for i in range(level):
ax = plt.subplot2grid((4, 1), (i, 0))
ax.plot(w_peak[i, :])
ax.set_title('level_{}'.format(i))
plt.show()
return w_peak
def get_difference(w_peak, points, level):
"""get difference to determine the locations of maximal values"""
sig = w_peak[level - 1, :]
pos_i = sig * (sig > 0)
thpos_i = (max(pos_i[0:round(points / 4)]) +
max(pos_i[round(points / 4):2 * round(points / 4)]) +
max(pos_i[2 * round(points / 4):3 * round(points / 4)]) +
max(pos_i[3 * round(points / 4):4 * round(points / 4)])) / 4
pos_i = 1 * (pos_i > thpos_i / 3)
neg_i = sig * (sig < 0)
thneg_i = (min(neg_i[0:round(points / 4)]) +
min(neg_i[round(points / 4):2 * round(points / 4)]) +
min(neg_i[2 * round(points / 4):3 * round(points / 4)]) +
min(neg_i[3 * round(points / 4):4 * round(points / 4)])) / 4
neg_i = - 1 * (neg_i < thneg_i / 4)
interval = pos_i + neg_i
loca = np.where(interval != 0)[0]
diff = []
for i in range(0, loca.shape[0] - 1):
if abs(loca[i] - loca[i + 1]) < 80:
diff.append(interval[loca[i]] - interval[loca[i + 1]])
else:
diff.append(0)
return interval, loca, diff
def qrs_delineation(w_peak, points, interval2):
"""detect Q, R, S, onset, offset for each QRS complex"""
countR = np.zeros(1, dtype=np.int32)
countQ = np.zeros(1, dtype=np.int32)
countS = np.zeros(1, dtype=np.int32)
count_onset = np.zeros(1, dtype=np.int32)
count_offset = np.zeros(1, dtype=np.int32)
Mj1 = w_peak[0, :]
Mj3 = w_peak[2, :]
i = 0
Rnum = 0
R_results = []
while i < points - 1:
if interval2[i] == -1:
mark1 = i
i += 1
while i < points - 1 and interval2[i] == 0:
i += 1
mark2 = i
mark3 = int(round((abs(Mj3[mark2]) * (mark1 + 1) + (mark2 + 1) * abs(Mj3[mark1])) /
(abs(Mj3[mark2]) + abs(Mj3[mark1]))))
R_results.append(mark3 - 5)
countR = np.concatenate((countR, np.zeros(mark3 - 5 - countR.shape[0]), np.ones(1)))
kqs = mark3 - 5
mark_q = 0
while kqs > 0 and mark_q < 3:
if Mj1[kqs] != 0:
mark_q += 1
kqs -= 1
countQ = np.concatenate((countQ, np.zeros(kqs - countQ.shape[0]), -1 * np.ones(1)))
kqs = mark3 - 5
mark_s = 0
while kqs < points - 1 and mark_s < 3:
if Mj1[kqs] != 0:
mark_s += 1
kqs += 1
countS = np.concatenate((countS, np.zeros(kqs - countS.shape[0]), -1 * np.ones(1)))
kqs = mark3 - 5
mark_q = 0
while kqs > 0 and mark_q < 5:
if Mj1[kqs] != 0:
mark_q += 1
kqs -= 1
count_onset = np.concatenate((count_onset, np.zeros(kqs - count_onset.shape[0]), -1 * np.ones(1)))
kqs = mark3 - 5
mark_s = 0
while kqs < points - 1 and mark_s < 5:
if Mj1[kqs] != 0:
mark_s += 1
kqs += 1
count_offset = np.concatenate((count_offset, np.zeros(kqs - count_offset.shape[0]), -1 * np.ones(1)))
i += 60
Rnum += 1
i += 1
R_located = np.nonzero(countR)[0]
S_located = np.nonzero(countS)[0]
Q_located = np.nonzero(countQ)[0]
onset_located = np.nonzero(count_onset)[0]
offset_located = np.nonzero(count_offset)[0]
return Q_located, R_located, S_located, onset_located, offset_located
def get_fecg(signal, wavelet=pywt.Wavelet('db8')):
"""calculate the flatted ecg data"""
# get the baseline and fecg
coeffs_1 = pywt.wavedec(signal, wavelet, 'periodization', k)
new_coeffs_1 = []
new_coeffs_1.append(coeffs_1[0])
for coeff in coeffs_1[1:]:
new_coeffs_1.append(np.zeros(coeff.shape))
baseline = pywt.waverec(new_coeffs_1, wavelet, 'periodization')
if len(signal) > len(baseline):
signal = signal[0:len(baseline)]
elif len(signal) < len(baseline):
baseline = baseline[0:len(signal)]
fecg = signal - baseline
plt.plot(fecg, 'g-')
return fecg
def find_r_peak(fecg, wavelet=pywt.Wavelet('db8')):
coeffs_2 = pywt.wavedec(fecg, wavelet, 'periodization', k)
new_coeffs_2 = []
for i in range(k + 1):
new_coeffs_2.append(np.zeros(coeffs_2[i].shape))
score = []
for i in range(k, 1, -1):
new_coeffs_2[i] = coeffs_2[i]
pulse = pywt.waverec(new_coeffs_2, wavelet, 'periodization')
new_coeffs_2[i] = np.zeros(coeffs_2[i].shape)
sum_pulse = np.sum(np.abs(pulse))
score.append(np.abs(np.sum(fecg * np.abs(pulse) / sum_pulse)))
score_diff = [score[i] - score[i + 1] for i in range(1, len(score) - 1)]
chosen_scale = int(np.argmax(score_diff) + 2)
new_coeffs_2[-chosen_scale] = coeffs_2[-chosen_scale]
pulse = pywt.waverec(new_coeffs_2, wavelet, 'periodization')
needle = np.abs(fecg * pulse)
peak_idx = processing.find_local_peaks(needle, 150)
return peak_idx
def find_onset_offset(fecg, peak_idx):
"""find the positions of QRS onset, peak and offset points"""
fecg_smooth = savgol_filter(fecg, 21, 3)
# plt.plot(fecg_smooth, 'r-')
nbeat = len(peak_idx)
print("Number of beats:", nbeat)
onset_idx = []
offset_idx = []
pop_list = []
for i in range(nbeat):
# if fecg[peak_idx[i]] > 0:
#
# j = 1
# while peak_idx[i] - j >= 0 and fecg[peak_idx[i] - j] <= fecg[peak_idx[i] - j + 1]:
# j += 1
# while peak_idx[i] - j >= 0 and fecg[peak_idx[i] - j] > fecg[peak_idx[i] - j + 1]:
# j += 1
# if peak_idx[i] - j >= 0:
# onset_idx.append(peak_idx[i] - j)
# else:
# pop_list.append(i)
#
# j = 1
# while peak_idx[i] + j < len(fecg) and fecg[peak_idx[i] + j - 1] >= fecg[peak_idx[i] + j]:
# j += 1
# while peak_idx[i] + j < len(fecg) and fecg[peak_idx[i] + j - 1] < fecg[peak_idx[i] + j]:
# j += 1
# if peak_idx[i] + j < len(fecg):
# offset_idx.append(peak_idx[i] + j)
# else:
# pop_list.append(i)
#
# else:
#
# j = 1
# while peak_idx[i] - j >= 0 and fecg[peak_idx[i] - j] >= fecg[peak_idx[i] - j + 1]:
# j += 1
# while peak_idx[i] - j >= 0 and fecg[peak_idx[i] - j] < fecg[peak_idx[i] - j + 1]:
# j += 1
# if peak_idx[i] - j >= 0:
# onset_idx.append(peak_idx[i] - j)
# else:
# pop_list.append(i)
#
# j = 1
# while peak_idx[i] + j < len(fecg) and fecg[peak_idx[i] + j - 1] <= fecg[peak_idx[i] + j]:
# j += 1
# while peak_idx[i] + j < len(fecg) and fecg[peak_idx[i] + j - 1] > fecg[peak_idx[i] + j]:
# j += 1
# if peak_idx[i] + j < len(fecg):
# offset_idx.append(peak_idx[i] + j)
# else:
# pop_list.append(i)
if fecg_smooth[peak_idx[i]] > 0:
j = 1
while peak_idx[i] - j >= 0 and fecg_smooth[peak_idx[i] - j] <= fecg_smooth[peak_idx[i] - j + 1]:
j += 1
while peak_idx[i] - j >= 0 and fecg_smooth[peak_idx[i] - j] > fecg_smooth[peak_idx[i] - j + 1]:
j += 1
if peak_idx[i] - j >= 0:
onset_idx.append(peak_idx[i] - j)
else:
pop_list.append(i)
j = 1
while peak_idx[i] + j < len(fecg_smooth) and \
fecg_smooth[peak_idx[i] + j - 1] >= fecg_smooth[peak_idx[i] + j]:
j += 1
while peak_idx[i] + j < len(fecg_smooth) and \
fecg_smooth[peak_idx[i] + j - 1] < fecg_smooth[peak_idx[i] + j]:
j += 1
if peak_idx[i] + j < len(fecg_smooth):
offset_idx.append(peak_idx[i] + j)
else:
pop_list.append(i)
else:
j = 1
while peak_idx[i] - j >= 0 and fecg_smooth[peak_idx[i] - j] >= fecg_smooth[peak_idx[i] - j + 1]:
j += 1
while peak_idx[i] - j >= 0 and fecg_smooth[peak_idx[i] - j] < fecg_smooth[peak_idx[i] - j + 1]:
j += 1
if peak_idx[i] - j >= 0:
onset_idx.append(peak_idx[i] - j)
else:
pop_list.append(i)
j = 1
while peak_idx[i] + j < len(fecg_smooth) and \
fecg_smooth[peak_idx[i] + j - 1] <= fecg_smooth[peak_idx[i] + j]:
j += 1
while peak_idx[i] + j < len(fecg_smooth) and \
fecg_smooth[peak_idx[i] + j - 1] > fecg_smooth[peak_idx[i] + j]:
j += 1
if peak_idx[i] + j < len(fecg_smooth):
offset_idx.append(peak_idx[i] + j)
else:
pop_list.append(i)
print("Pop idx:", pop_list)
if pop_list:
if len(pop_list) % 2 == 0:
peak_idx = np.delete(peak_idx, pop_list)
onset_idx.pop(-1)
offset_idx.pop(0)
elif len(pop_list) % 2 == 1:
peak_idx = np.delete(peak_idx, pop_list)
offset_idx.pop(0)
plt.plot(peak_idx, fecg[peak_idx], 'rx', marker='x', color='#8b0000', label='R_Peak', markersize=8)
plt.plot(onset_idx, fecg[onset_idx], 'rx', marker='o', color='#8b0000', label='Onset', markersize=8)
plt.plot(offset_idx, fecg[offset_idx], 'rx', marker='v', color='#8b0000', label='Offset', markersize=8)
print("R peaks locations:", peak_idx, len(peak_idx))
print("Onset locations:", onset_idx, len(onset_idx))
print("Offset locations:", offset_idx, len(offset_idx))
return onset_idx, peak_idx, offset_idx
def get_reference_voltage(fecg, peak_idx, onset_idx, offset_idx):
nbeat = len(peak_idx)
mean_idx_diff_1 = int(np.sum(peak_idx - onset_idx) / nbeat)
mean_idx_diff_2 = int(np.sum(-peak_idx + offset_idx) / nbeat)
reference_voltage = np.sum(fecg[onset_idx]) / nbeat
print("Reference voltage:", reference_voltage)
return mean_idx_diff_1, mean_idx_diff_2, reference_voltage
def find_t_peak(signal, R_R_interval, offset_located):
"""find the positions of T wave peaks"""
countT = np.zeros(1, dtype=np.int32)
pop_list = []
for i in range(offset_located.shape[0]):
offset_T_interval = []
try:
idx = processing.find_local_peaks(signal[offset_located[i]:offset_located[i] + int(R_R_interval / 2)],
int(R_R_interval / 2) - 10)
offset_T_interval.append(idx[0])
idx = idx[0] + offset_located[i]
countT = np.concatenate((countT, np.zeros(idx - countT.shape[0]), -1 * np.ones(1)))
except:
pop_list.append(i)
average_interval = np.mean(np.array(offset_T_interval)) if offset_T_interval else 0
if average_interval != 0:
countT = np.concatenate((countT, np.zeros(offset_located[i] + average_interval
- countT.shape[0]), -1 * np.ones(1)))
print(" Add a T peak at {}".format(offset_located[i] + average_interval))
else:
print(" No T peak at {}".format(offset_located[i]))
T_located = np.nonzero(countT)[0]
return T_located, pop_list
def find_p_peak(signal, R_R_interval, onset_located):
"""find the positions of P wave peaks"""
countP = np.zeros(1, dtype=np.int32)
pop_list = []
for i in range(onset_located.shape[0]):
onset_P_interval = []
try:
idx = processing.find_local_peaks(signal[onset_located[i] - int(R_R_interval / 5):onset_located[i]],
int(R_R_interval / 5) - 1)
onset_P_interval.append(idx[0])
idx = idx[0] + onset_located[i] - int(R_R_interval / 5)
countP = np.concatenate((countP, np.zeros(idx - countP.shape[0]), -1 * np.ones(1)))
except:
pop_list.append(i)
average_interval = np.mean(np.array(onset_P_interval)) if onset_P_interval else 0
if average_interval != 0:
countP = np.concatenate((countP, np.zeros(onset_located[i] - int(R_R_interval / 5) +
average_interval - countP.shape[0]),
-1 * np.ones(1)))
print(" Add a P peak at {}".format(onset_located[i] - int(R_R_interval / 5) +
average_interval))
else:
print(" No P peak at {}".format(onset_located[i]))
P_located = np.nonzero(countP)[0]
return P_located, pop_list
def find_f_point(signal, offset_located, T_located, reference_voltage):
"""find the positions of F points"""
countF = np.zeros(1, dtype=np.int32)
pop_list = []
c_F = 0
for i in range(offset_located.shape[0]):
try:
if np.min(signal[offset_located[i]:T_located[i]]) < reference_voltage < \
np.max(signal[offset_located[i]:T_located[i]]):
idx, = np.where(signal[offset_located[i]:T_located[i]] >= reference_voltage)
idx = idx[0] + offset_located[i]
countF = np.concatenate((countF, np.zeros(idx - countF.shape[0]), -1 * np.ones(1)))
else:
idx = int((offset_located[i] + T_located[i]) / 2)
countF = np.concatenate((countF, np.zeros(idx - countF.shape[0]), np.ones(1)))
except:
c_F += 1
pop_list.append(i)
print(" Pop {} points from total {} points!".format(c_F, offset_located.shape[0]))
F_located = np.nonzero(countF)[0]
F_located_val = countF[F_located]
return F_located, F_located_val, pop_list
def generate_features(fecg, peak_idx, t_peak_idx, f_point_idx, f_point_val,
mean_idx_diff_1, mean_idx_diff_2, reference_voltage):
"""extract some features"""
"""features from paper Ischemia episode detection......"""
nbeat = len(peak_idx)
feature_1 = []
feature_2 = []
feature_3 = []
for i in range(nbeat):
k = peak_idx[i] + mean_idx_diff_2
feature_1.append(np.sum(np.abs(fecg[k:t_peak_idx[i] + 1] - reference_voltage)))
if f_point_val[i] == -1:
feature_2.append(np.sum(fecg[k:f_point_idx[i] + 1] - reference_voltage) / np.abs(fecg[peak_idx[i]]))
else:
feature_2.append(10)
m = peak_idx[i] - mean_idx_diff_1
feature_3.append(np.abs((fecg[k] - fecg[m]) / (k - m)))
feature_2 = np.asarray(feature_2)
f2_f_idx = np.where(feature_2 == 10)[0]
f2_normal_idx = np.where(feature_2 != 10)[0]
average_f2 = np.sum(feature_2[f2_normal_idx]) / f2_normal_idx.shape[0]
feature_2[f2_f_idx] = average_f2
feature_2.tolist()
feature_1 = np.asarray(feature_1).reshape([len(feature_1), 1])
feature_2 = np.asarray(feature_2).reshape([len(feature_2), 1])
feature_3 = np.asarray(feature_3).reshape([len(feature_3), 1])
return feature_1, feature_2, feature_3
# x_1 = []
# x_2 = []
# x_3 = []
# for i in range(int(nbeat / 5)):
# x_1.append(sum(feature_1[5 * i:5 * i + 5]) / 5)
# x_2.append(sum(feature_2[5 * i:5 * i + 5]) / 5)
# x_3.append(sum(feature_3[5 * i:5 * i + 5]) / 5)
# if nbeat % 5 != 0:
# x_1.append(sum(feature_1[int(nbeat / 5) * 5 - len(feature_1):]) / (len(feature_1) - int(nbeat / 5) * 5))
# x_2.append(sum(feature_2[int(nbeat / 5) * 5 - len(feature_2):]) / (len(feature_2) - int(nbeat / 5) * 5))
# x_3.append(sum(feature_3[int(nbeat / 5) * 5 - len(feature_3):]) / (len(feature_3) - int(nbeat / 5) * 5))
# x_1 = np.asarray(x_1).reshape([len(x_1), 1])
# x_2 = np.asarray(x_2).reshape([len(x_2), 1])
# x_3 = np.asarray(x_3).reshape([len(x_3), 1])
#
# return x_1, x_2, x_3
atr_list = generate_atr_list('./edb_records')
with open('./edb_st_episodes.json') as infile:
st_dict = json.load(infile)