Spaces:
Sleeping
Sleeping
Commit
·
1737659
1
Parent(s):
35af352
Added sleep trains options on the tool
Browse files- portiloop/src/demo/demo.py +7 -1
- portiloop/src/demo/offline.py +10 -3
- portiloop/src/demo/utils.py +49 -0
portiloop/src/demo/demo.py
CHANGED
@@ -29,10 +29,15 @@ def main():
|
|
29 |
# Threshold value
|
30 |
threshold = gr.Slider(0, 1, value=0.82, step=0.01, label="Threshold", interactive=True)
|
31 |
# Detection Channel
|
|
|
|
|
32 |
detect_channel = gr.Dropdown(choices=["1", "2", "3", "4", "5", "6", "7", "8"], value="2", label="Detection Channel in XDF recording", interactive=True)
|
33 |
# Frequency
|
34 |
freq = gr.Dropdown(choices=["100", "200", "250", "256", "500", "512", "1000", "1024"], value="250", label="Sampling Frequency (Hz)", interactive=True)
|
35 |
|
|
|
|
|
|
|
36 |
with gr.Row():
|
37 |
output_array = gr.File(label="Output CSV File")
|
38 |
output_table = gr.Markdown(label="Output Table")
|
@@ -45,7 +50,8 @@ def main():
|
|
45 |
detect_filter,
|
46 |
threshold,
|
47 |
detect_channel,
|
48 |
-
freq
|
|
|
49 |
outputs=[output_array, output_table])
|
50 |
|
51 |
demo.queue()
|
|
|
29 |
# Threshold value
|
30 |
threshold = gr.Slider(0, 1, value=0.82, step=0.01, label="Threshold", interactive=True)
|
31 |
# Detection Channel
|
32 |
+
|
33 |
+
with gr.Row():
|
34 |
detect_channel = gr.Dropdown(choices=["1", "2", "3", "4", "5", "6", "7", "8"], value="2", label="Detection Channel in XDF recording", interactive=True)
|
35 |
# Frequency
|
36 |
freq = gr.Dropdown(choices=["100", "200", "250", "256", "500", "512", "1000", "1024"], value="250", label="Sampling Frequency (Hz)", interactive=True)
|
37 |
|
38 |
+
# Detect trains dropdown
|
39 |
+
detect_trains = gr.Dropdown(choices=["All Spindles", "Isolated & First", "Trains"], value="All Spindles", label="Detection mode:", interactive=True)
|
40 |
+
|
41 |
with gr.Row():
|
42 |
output_array = gr.File(label="Output CSV File")
|
43 |
output_table = gr.Markdown(label="Output Table")
|
|
|
50 |
detect_filter,
|
51 |
threshold,
|
52 |
detect_channel,
|
53 |
+
freq,
|
54 |
+
detect_trains],
|
55 |
outputs=[output_array, output_table])
|
56 |
|
57 |
demo.queue()
|
portiloop/src/demo/offline.py
CHANGED
@@ -2,11 +2,11 @@ import numpy as np
|
|
2 |
from portiloop.src.detection import SleepSpindleRealTimeDetector
|
3 |
from portiloop.src.stimulation import UpStateDelayer
|
4 |
from portiloop.src.processing import FilterPipeline
|
5 |
-
from portiloop.src.demo.utils import compute_output_table, sleep_stage, xdf2array, offline_detect, offline_filter, OfflineSleepSpindleRealTimeStimulator
|
6 |
import gradio as gr
|
7 |
|
8 |
|
9 |
-
def run_offline(xdf_file, detect_filter_opts, threshold, channel_num, freq, stimulation_phase="Fast", buffer_time=0.25):
|
10 |
# Get the options from the checkbox group
|
11 |
offline_filtering = 0 in detect_filter_opts
|
12 |
lacourse = 1 in detect_filter_opts
|
@@ -76,7 +76,14 @@ def run_offline(xdf_file, detect_filter_opts, threshold, channel_num, freq, stim
|
|
76 |
# Create the detector
|
77 |
if online_detection:
|
78 |
detector = SleepSpindleRealTimeDetector(threshold=threshold, channel=1) # always 1 because we have only one channel
|
79 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
80 |
if stimulation_phase != "Fast":
|
81 |
stimulation_delayer = UpStateDelayer(freq, stimulation_phase == 'Peak', time_to_buffer=buffer_time, stimulate=lambda: None)
|
82 |
stimulator.add_delayer(stimulation_delayer)
|
|
|
2 |
from portiloop.src.detection import SleepSpindleRealTimeDetector
|
3 |
from portiloop.src.stimulation import UpStateDelayer
|
4 |
from portiloop.src.processing import FilterPipeline
|
5 |
+
from portiloop.src.demo.utils import OfflineIsolatedSpindleRealTimeStimulator, OfflineSpindleTrainRealTimeStimulator, compute_output_table, sleep_stage, xdf2array, offline_detect, offline_filter, OfflineSleepSpindleRealTimeStimulator
|
6 |
import gradio as gr
|
7 |
|
8 |
|
9 |
+
def run_offline(xdf_file, detect_filter_opts, threshold, channel_num, freq, detect_trains, stimulation_phase="Fast", buffer_time=0.25):
|
10 |
# Get the options from the checkbox group
|
11 |
offline_filtering = 0 in detect_filter_opts
|
12 |
lacourse = 1 in detect_filter_opts
|
|
|
76 |
# Create the detector
|
77 |
if online_detection:
|
78 |
detector = SleepSpindleRealTimeDetector(threshold=threshold, channel=1) # always 1 because we have only one channel
|
79 |
+
|
80 |
+
if detect_trains == "All Spindles":
|
81 |
+
stimulator = OfflineSleepSpindleRealTimeStimulator()
|
82 |
+
elif detect_trains == "Trains":
|
83 |
+
stimulator = OfflineSpindleTrainRealTimeStimulator()
|
84 |
+
elif detect_trains == "Isolated & First":
|
85 |
+
stimulator = OfflineIsolatedSpindleRealTimeStimulator()
|
86 |
+
|
87 |
if stimulation_phase != "Fast":
|
88 |
stimulation_delayer = UpStateDelayer(freq, stimulation_phase == 'Peak', time_to_buffer=buffer_time, stimulate=lambda: None)
|
89 |
stimulator.add_delayer(stimulation_delayer)
|
portiloop/src/demo/utils.py
CHANGED
@@ -39,6 +39,7 @@ def sleep_stage(data, threshold=150, group_size=2):
|
|
39 |
return unmasked_indices
|
40 |
|
41 |
|
|
|
42 |
class OfflineSleepSpindleRealTimeStimulator(Stimulator):
|
43 |
def __init__(self):
|
44 |
self.last_detected_ts = time.time()
|
@@ -70,6 +71,54 @@ class OfflineSleepSpindleRealTimeStimulator(Stimulator):
|
|
70 |
self.delayer = delayer
|
71 |
self.delayer.stimulate = lambda: True
|
72 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
73 |
def xdf2array(xdf_path, channel):
|
74 |
xdf_data, _ = pyxdf.load_xdf(xdf_path)
|
75 |
|
|
|
39 |
return unmasked_indices
|
40 |
|
41 |
|
42 |
+
|
43 |
class OfflineSleepSpindleRealTimeStimulator(Stimulator):
|
44 |
def __init__(self):
|
45 |
self.last_detected_ts = time.time()
|
|
|
71 |
self.delayer = delayer
|
72 |
self.delayer.stimulate = lambda: True
|
73 |
|
74 |
+
|
75 |
+
class OfflineSpindleTrainRealTimeStimulator(OfflineSleepSpindleRealTimeStimulator):
|
76 |
+
def __init__(self):
|
77 |
+
super().__init__()
|
78 |
+
self.max_spindle_train_t = 6.0
|
79 |
+
|
80 |
+
def stimulate(self, detection_signal):
|
81 |
+
self.index += 1
|
82 |
+
stim = False
|
83 |
+
for sig in detection_signal:
|
84 |
+
# We detect a stimulation
|
85 |
+
if sig:
|
86 |
+
# Record time of stimulation
|
87 |
+
ts = self.index
|
88 |
+
|
89 |
+
elapsed = ts - self.last_detected_ts
|
90 |
+
# Check if time since last stimulation is long enough
|
91 |
+
if self.wait_timesteps < elapsed < int(self.max_spindle_train_t * 250):
|
92 |
+
if self.delayer is not None:
|
93 |
+
# If we have a delayer, notify it
|
94 |
+
self.delayer.detected()
|
95 |
+
stim = True
|
96 |
+
|
97 |
+
self.last_detected_ts = ts
|
98 |
+
return stim
|
99 |
+
|
100 |
+
class OfflineIsolatedSpindleRealTimeStimulator(OfflineSpindleTrainRealTimeStimulator):
|
101 |
+
def stimulate(self, detection_signal):
|
102 |
+
self.index += 1
|
103 |
+
stim = False
|
104 |
+
for sig in detection_signal:
|
105 |
+
# We detect a stimulation
|
106 |
+
if sig:
|
107 |
+
# Record time of stimulation
|
108 |
+
ts = self.index
|
109 |
+
|
110 |
+
elapsed = ts - self.last_detected_ts
|
111 |
+
# Check if time since last stimulation is long enough
|
112 |
+
if int(self.max_spindle_train_t * 250) < elapsed:
|
113 |
+
if self.delayer is not None:
|
114 |
+
# If we have a delayer, notify it
|
115 |
+
self.delayer.detected()
|
116 |
+
stim = True
|
117 |
+
|
118 |
+
self.last_detected_ts = ts
|
119 |
+
return stim
|
120 |
+
|
121 |
+
|
122 |
def xdf2array(xdf_path, channel):
|
123 |
xdf_data, _ = pyxdf.load_xdf(xdf_path)
|
124 |
|