-
Notifications
You must be signed in to change notification settings - Fork 0
/
preprocessing PTDBD.py
178 lines (133 loc) · 6.1 KB
/
preprocessing PTDBD.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
import numpy as np
from scipy import signal
import matplotlib.pyplot as plt
import os
from scipy import signal
def butter_highpass(cutoff, fs, order=5):
nyq = 0.5 * fs
normal_cutoff = cutoff / nyq
b, a = signal.butter(order, normal_cutoff, btype='high', analog=False)
return b, a
def butter_lowpass(cutoff, fs, order=5):
nyq = 0.5 * fs
normal_cutoff = cutoff / nyq
b, a = signal.butter(order, normal_cutoff, btype='low', analog=False)
return b, a
def butter_highpass_filter(data, cutoff, fs, order=5):
b, a = butter_highpass(cutoff, fs, order=order)
y = signal.filtfilt(b, a, data)
return y
def butter_lowpass_filter(data, cutoff, fs, order=5):
b, a = butter_lowpass(cutoff, fs, order=order)
y = signal.filtfilt(b, a, data)
return y
def get_ecg_names():
data = list()
infarction = list()
localization = list()
for root, dirs, files in os.walk("PTBDB"):
for file in files:
if file.endswith(".hea"):
#print(os.path.join(root, file))
fp = open(os.path.join(root, file))
diagnosisline = [line for line in fp if line.startswith('# Reason for admission:')]
diagnosis = diagnosisline[0].rstrip("\n")
diagnosis = diagnosis[24:]
fname = os.path.join(root, file.rstrip("hea")+"csv")
if diagnosis == "Myocardial infarction":
localizationline = [line for line in fp if line.startswith('# Acute infarction (localization):')]
localization = localizationline[0].rstrip("\n")
print(file.rstrip(".hea")+"csv", "has an infarction")
data.append(fname)
infarction.append(1)
localization = localization[30:]
localization.append()
print(file.rstrip(".hea")+"csv", "has an infarction in: ",localization)
elif diagnosis == "Healthy control":
#print(file.rstrip(".hea") + "csv", "is a healthy control")
data.append(fname)
infarction.append(0)
#print(,file.rstrip(".hea")+".csv")
break
return data,infarction
def load_PTBDB_ecg(file):
# ECG format
# Elapsed time','i','ii','iii','avr','avl','avf','v1','v2','v3','v4','v5','v6'
rawarray = np.loadtxt(file,delimiter=',',skiprows=2,usecols=(1,2,3,4,5,6,7,8,9,10,11,12))
b = list()
for item in rawarray.T:
##Highpass to get rid of baseline wander and lowpass to get rid of high frequency noise
#y = butter_highpass_filter(item, 1, 1500, 5)
#y = butter_lowpass_filter(item, 25, 1000, 5)
y = butter_lowpass_filter(butter_highpass_filter(item, 1, 1500, 5), 50, 1000, 5)
b.append(y)
a = np.asarray(b).T
# in principle the line below is Pan-Tompkins from derivative step to moving window step
c = np.convolve(np.square(np.gradient(a[:,1],1)),np.ones(50))
refractory_period = 200 # to have a QRS after less than 200 ms is physiologically impossible
threshold = max(c)/3 #Threshold should be one 1/3 of the maximum peak in registration
# Pan-Tompkins continues
peaks = list()
for idx, val in enumerate(c):
# Unpythonic
refractory_period+=1
if idx - 1 > 0 and idx + 1 < len(c) and c[idx - 1] < val and refractory_period>200 and c[idx + 1] < val and val > threshold :
# plt.axvline(x=idx,linewidth=0.5,color = 'k')
refractory_period = 0
peaks.append(idx)
peaks = np.asarray(peaks)
#meanrr = np.mean(np.diff(peaks, 1, 0))
#fig, ax = plt.subplots()
#ax.plot(c*20,linewidth=0.5,color='r')
#ax.plot(b,linewidth=0.5,color='r')
#ax.plot(a,linewidth=0.5)
#for num in peaks:
# plt.axvline(x=num, linewidth=0.5, color='k')
beats = list() # A list to store our individual beats
#meanrr = np.mean(np.diff(rpeaks, 1, 0)) # Calculate new more accurate RR-interval
for idx, val in enumerate(peaks):
if idx > 0 and idx < len(peaks) - 1:
slice = a[int(val - 200):int(val + 400):1]
beats.append(slice)
barr = np.asarray(beats) # Make into a numpy array for convenience
#barr = np.take(barr, (0, 6, 1, 7, 2, 8, 3, 9, 4, 10, 5, 11), 2) # reorganize columns for subplots
#fig, ax = plt.subplots()
#ax.plot(barr[1])
#plt.show()
return barr
#lead_names = np.take(lead_names, (0, 6, 1, 7, 2, 8, 3, 9, 4, 10, 5, 11)) # Reorganize the columns for subplots
fnames,infarction = get_ecg_names()
ecgs = list()
for idx,f in enumerate(fnames[1:len(fnames)]): #len(fnames)len(fnames)
ecg = load_PTBDB_ecg(f)
ecgs.append((ecg,np.full(len(ecg),infarction[idx]),np.full(len(ecg),idx),idx))
print(idx)
ecgs = np.asarray(ecgs)
save = False
if save == True:
np.save('alsoecgnr123.npy',ecgs,allow_pickle=True)
print("saved")
ecgs = np.asarray(ecgs)
plot = True
if plot == True:
lead_names = np.asarray(['i', 'ii', 'iii', 'aVr', 'aVl', 'aVf', 'v1', 'v2', 'v3', 'v4', 'v5', 'v6'])
fig, ax_list = plt.subplots(6, 2,sharex='all')
#ax_list = ax_list.flatten()
for ecg in ecgs:
for idx,ax in enumerate(ax_list.T.flatten()):
#print(idx)
ax.plot(ecg[0][:,:,idx].T,linewidth=0.1,alpha=0.1,color='black')
ax.set_ylabel(lead_names[idx])
#ax_list[idx].axvline(200, linewidth=0.8, color='r')
#ax_list[idx].set_ylabel(lead_names[idx])
#ax_list[idx].set_autoscaley_on(False)
#ax_list[idx].set_autoscalex_on(True)
#ax_list[idx].set_ylim([-2, 2])
#ax_list[idx].grid(True,'both','both')
# ax_list[idx].yaxis.set_major_locator(MultipleLocator(1))
# ax_list[idx].yaxis.set_minor_locator(MultipleLocator(0.2))
# ax_list[idx].xaxis.set_major_locator(MultipleLocator(200))
# ax_list[idx].xaxis.set_minor_locator(MultipleLocator(40))
plt.tight_layout()
plt.show()
print("done")