def test_peaks_cleanecg(self):
t, s = loadtxt(path + 'cleanecg.txt')
pr = array([
20, 87, 156, 225, 291, 355, 418, 482, 550, 624, 694, 764, 834, 905,
978
])
assert_array_equal(pr, peaks(s, 75))
def BigPeaks(s, th, min_peak_distance=5, peak_return_percentage=0.1):
p = peaks(s, th)
pp = []
if len(p) == 0:
pp = []
else:
p = clean_near_peaks(s, p, min_peak_distance)
if len(p) != 0:
ars = argsort(s[p])
pp = p[ars]
num_peaks_to_return = int(ceil(len(p) * peak_return_percentage))
pp = pp[-num_peaks_to_return:]
else:
pp == []
return pp
def _bigPeaks(s, th, min_peak_distance=5, peak_return_percentage=0.1):
pp = []
if not list(s):
return pp
else:
p = ni.peaks(s, th)
if not list(p):
return pp
else:
p = ni.clean_near_peaks(s, p, min_peak_distance)
if not list(p):
return pp
else:
ars = np.argsort(s[p])
pp = p[ars]
num_peaks_to_return = int(np.ceil(len(p) * peak_return_percentage))
pp = pp[-num_peaks_to_return:]
return pp
fs = 1000.
t = arange(len(ecg)) / fs
#Para retirar o DC e colocar o referencial na origem
media = mean(ecg)
ecg = ecg - media
subplot(2, 1, 1)
plot(t, ecg)
title('ECG')
xlabel('Time(ms)')
ylabel('ECG(mV)')
grid()
counter = ni.peaks(ecg, 400)
noise = zeros(len(ecg))
noise[min(counter):max(counter)] = 1
subplot(2, 1, 2)
plot(t, noise)
title('ECG NOISE DETECTOR')
xlabel('Time(ms)')
ylabel('Detection')
axis([0, 100, -1, 1.5])
grid()
plt.tight_layout()
show()
def test_peaks_eda(self):
t, s = loadtxt(path + 'eda.txt')
pr = array([95, 146, 346])
assert_array_equal(pr, peaks(s, 600))
dnx = np.diff(nx)
dny = np.diff(ny)
# Detection of straight segments
idx = dnx != 0
idy = dny == 0
vals = (idx * idy) + (~idx * ~idy)
figure()
subplot(131)
plot(nx, ny)
scatter(nx[vals], ny[vals], color='r')
# Looping detection
maxs_ny = ni.peaks(ny)
mins_ny = ni.peaks(-ny)
maxs_nx = ni.peaks(nx)
mins_nx = ni.peaks(-nx)
subplot(132)
plot(nx, ny)
scatter(nx[maxs_ny], ny[maxs_ny], color='r')
scatter(nx[mins_ny], ny[mins_ny], color='g')
scatter(nx[maxs_nx], ny[maxs_nx], color='b')
scatter(nx[mins_nx], ny[mins_nx], color='k')
e = hstack([maxs_ny, mins_ny, maxs_nx, mins_nx])
mask = hstack([
ones(len(maxs_ny)),
t=arange(len(ecg))/fs
#Para retirar o DC e colocar o referencial na origem
media=mean(ecg)
ecg=ecg-media
subplot(2,1,1)
plot(t, ecg)
title('ECG')
xlabel('Time(ms)')
ylabel('ECG(mV)')
grid()
counter = ni.peaks(ecg,400)
noise=zeros(len(ecg))
noise[min(counter):max(counter)]=1
subplot(2,1,2)
plot(t,noise)
title('ECG NOISE DETECTOR')
xlabel('Time(ms)')
ylabel('Detection')
axis([0, 100, -1, 1.5])
grid()
plt.tight_layout()
show()
def test_peaks_eda(self):
t, s = loadtxt(path + 'eda.txt')
pr = array([95, 146, 346])
assert_array_equal(pr, peaks(s, 600))
def test_peaks_cleanecg(self):
t, s = loadtxt(path + 'cleanecg.txt')
pr = array([20, 87, 156, 225, 291, 355, 418, 482, 550, 624, 694, 764, 834, 905, 978])
assert_array_equal(pr, peaks(s, 75))