def get_resampled_data(subject_id, visit_number):
calibration_date_pers, is_calibrated_pm_pers, airspeck_raw = load_personal_airspeck_file(
subject_id,
subject_visit_number=visit_number,
project_name='peeps',
upload_type='manual',
is_minute_averaged=False,
calibrate_pm_and_gas=False,
return_calibration_flag=False)
# If upload type is automatic -- Change to 'gpsLatitude':'gpsLongitude'
# airspeck_raw['gpsAccuracy'] = pd.to_numeric(airspeck_raw['gpsAccuracy'])
airspeck_raw.loc[airspeck_raw['gpsAccuracy'] > 1000,
'gpsLongitude':'gpsLatitude'] = np.nan
airspeck_raw.loc[airspeck_raw['gpsLatitude'] < 28.4,
'gpsLongitude':'gpsLatitude'] = np.nan
airspeck_raw.loc[airspeck_raw['gpsLatitude'] > 28.9,
'gpsLongitude':'gpsLatitude'] = np.nan
airspeck_raw.loc[airspeck_raw['gpsLongitude'] < 76.8,
'gpsLongitude':'gpsLatitude'] = np.nan
airspeck_raw.loc[airspeck_raw['gpsLongitude'] > 77.6,
'gpsLongitude':'gpsLatitude'] = np.nan
airspeck = airspeck_raw.resample('1min').mean()
return calibration_date_pers, is_calibrated_pm_pers, airspeck
def get_resampled_data(subject_id):
calibration_date_pers, is_calibrated_pm_pers, airspeck_raw = load_personal_airspeck_file(subject_id, 'british-heart', upload_type='automatic', calibrate_pm_and_gas=True, return_calibration_flag=True)
#airspeck_raw = load_personal_airspeck_file(subject_id, project_name='british-heart', upload_type='automatic')
airspeck_raw['gpsAccuracy'] = pd.to_numeric(airspeck_raw['gpsAccuracy'])
airspeck_raw.loc[airspeck_raw['gpsAccuracy'] > 1000, 'gpsLatitude':'gpsLongitude'] = np.nan
airspeck_raw.loc[airspeck_raw['gpsLatitude'] < 49.88, 'gpsLatitude':'gpsLongitude'] = np.nan
airspeck_raw.loc[airspeck_raw['gpsLatitude'] > 55.79, 'gpsLatitude':'gpsLongitude'] = np.nan
airspeck_raw.loc[airspeck_raw['gpsLongitude'] < -5.9, 'gpsLatitude':'gpsLongitude'] = np.nan
airspeck_raw.loc[airspeck_raw['gpsLongitude'] > 1.8, 'gpsLatitude':'gpsLongitude'] = np.nan
airspeck = airspeck_raw.resample('1min').mean()
return airspeck
def plot_philap_report_graphs_for_subject(subject_id, graphs_dir):
# Download raw data if not present
participant_details = load_philap_participant_details()
try:
airspeck_raw = load_personal_airspeck_file(subject_id,
upload_type='manual',
is_minute_averaged=False)
airspeck = airspeck_raw.resample('1min').mean()
respeck = load_respeck_file(subject_id, upload_type='manual')
except:
print(
"Please download all Peeps data via download_all_philap_data(raw_airspeck=True) "
"before calling this function")
# Delete incorrect GPS. These coordinates are just outside the larger area of Delhi
airspeck_raw.loc[airspeck_raw['gpsAccuracy'] > 1000,
'gpsLongitude':'gpsLatitude'] = np.nan
airspeck_raw.loc[airspeck_raw['gpsLatitude'] < 10,
'gpsLongitude':'gpsLatitude'] = np.nan
airspeck_raw.loc[airspeck_raw['gpsLatitude'] > 40,
'gpsLongitude':'gpsLatitude'] = np.nan
airspeck_raw.loc[airspeck_raw['gpsLongitude'] < 10,
'gpsLongitude':'gpsLatitude'] = np.nan
airspeck_raw.loc[airspeck_raw['gpsLongitude'] > 80,
'gpsLongitude':'gpsLatitude'] = np.nan
home_gps = airspeck.loc[(1 < airspeck.index.hour)
& (airspeck.index.hour <= 3)].mean()
# If there was no personal data during the night, fall back on the GPS coordinates the researchers provided
if pd.isnull(home_gps['gpsLatitude']):
home_gps = get_home_gps_for_subject(subject_id, participant_details)
# Select locations near home
radius_home = 0.002
correction_factor = airspeck['gpsAccuracy'] * 0.00001
home_mask = (np.abs(airspeck['gpsLatitude'] - home_gps['gpsLatitude']) < radius_home + correction_factor) & \
(np.abs(airspeck['gpsLongitude'] - home_gps['gpsLongitude']) < radius_home + correction_factor)
##################################
# Draw detailed exposure plot
##################################
sns.set_style('darkgrid', {'xtick.bottom': True, 'xtick.major.size': 5})
fig, ax = plt.subplots(figsize=(15, 5))
if np.count_nonzero(home_mask) > 0:
for ts in airspeck.loc[home_mask].index:
ax.axvspan(ts,
ts + pd.DateOffset(minutes=1),
facecolor=CB_color_cycle[0],
alpha=0.3,
zorder=1)
ax.scatter(airspeck.index, airspeck['pm2_5'], s=2, color='black', zorder=2)
# Plot stationary airspeck home
home_airspeck = load_static_airspeck_file(subject_id,
suffix_filename='_home')
if home_airspeck is not None and len(home_airspeck) > 0:
ax.scatter(home_airspeck.index,
home_airspeck['pm2_5'],
s=2,
color='blue')
ax.set_ylabel("PM2.5 (μg/m³)")
start = airspeck.index[0].replace(hour=0, minute=0, second=0)
end = airspeck.index[-1].replace(hour=0, minute=0,
second=0) + pd.DateOffset(days=1)
ax.set_xlim(start, end)
formatter = mdates.DateFormatter('%d.%m %Hh',
tz=dateutil.tz.gettz(
project_mapping['philap'][1]))
ax.xaxis.set_major_formatter(formatter)
ax.set_title(
"Continuous PM2.5 personal exposure levels and ambient concentrations")
fig.autofmt_xdate()
home_patch = mpatches.Patch(color=CB_color_cycle[0],
label='Home',
alpha=0.3)
airs_home_patch = Line2D(range(1),
range(1),
marker='o',
color='#00000000',
markerfacecolor="blue",
label='Home sensor')
airp_patch = Line2D(range(1),
range(1),
marker='o',
color='#00000000',
markerfacecolor="black",
label='Personal sensor')
plt.legend(handles=[home_patch, airp_patch, airs_home_patch])
plt.tight_layout()
plt.savefig(graphs_dir + "{}_detailed_exposure.png".format(subject_id),
dpi=300)
plt.show()
##################################
# Draw summary bar graph
##################################
sns.set_style('darkgrid', {'xtick.bottom': False, 'xtick.major.size': 0.0})
home = airspeck.loc[home_mask, 'pm2_5'].mean()
other = airspeck.loc[~home_mask, 'pm2_5'].mean()
overall = airspeck['pm2_5'].mean()
if home_airspeck is not None:
home_ambient = home_airspeck['pm2_5'].mean()
else:
home_ambient = np.nan
mean_values = [home, other, overall, home_ambient]
fig, ax = plt.subplots(figsize=(8, 5))
ax.set_title(
"Mean PM2.5 personal exposure levels and ambient concentrations")
ax.bar(np.arange(len(mean_values)),
mean_values,
width=0.5,
color=CB_color_cycle,
edgecolor="none")
ax.set_ylabel("PM2.5 (μg/m³)")
ax.set_xlim(-0.5, len(mean_values) - 0.5)
plt.xticks(np.arange(len(mean_values)), [
"Home\npersonal", "Other\npersonal", "Overall\npersonal",
"Home\nambient"
])
plt.savefig(graphs_dir +
"{}_mean_exposure.png".format(subject_id, subject_id),
dpi=300)
plt.show()
##################################
# Draw map
##################################
get_maps_image(airspeck_raw,
graphs_dir + "{}_airspeck_map.png".format(subject_id),
zoom=13)
##################################
# Other statistics
##################################
# Create new empty file
open(graphs_dir + "{}_stats.txt".format(subject_id), 'w').close()
# Append stats to this file
with open(graphs_dir + "{}_stats.txt".format(subject_id), 'a') as f:
f.write("Step count: {}\n".format(respeck['step_count'].sum()))
f.write(
"Mean breathing rate during night: {:.2f} breaths per minute\n".
format(respeck.loc[(0 < respeck.index.hour) &
(respeck.index.hour < 6),
'breathing_rate'].mean()))
f.write("Mean breathing rate during day: {:.2f} breaths per minute\n".
format(respeck.loc[(6 <= respeck.index.hour) &
(respeck.index.hour <= 23),
'breathing_rate'].mean()))
f.write("\nStart of recording: {}\n".format(
airspeck.index[0].replace(tzinfo=None)))
f.write("End of recording: {}\n".format(
airspeck.index[-1].replace(tzinfo=None)))
f.write("Total duration: {}\n".format(airspeck.index[-1] -
airspeck.index[0]))
f.write("Total recording time at home: {:.1f} h\n".format(
np.count_nonzero(home_mask) / 60.))
def create_dublin_pixelgram_for_subject(
subject_id, overwrite_pixelgram_if_already_exists=False):
download_respeck_and_personal_airspeck_data(subject_id,
upload_type='manual')
respeck_data = load_respeck_file(subject_id, upload_type='manual')
airspeck_data = load_personal_airspeck_file(subject_id,
upload_type='manual')
# Load correction factors for timezone
corrections = pd.read_excel(dublin_timezones_correction_filepath).replace(
np.nan, 0).set_index('subject_id')
participant_details = load_dublin_participant_details()
row = participant_details.loc[subject_id]
# Load exposure period
from_time = row['start_of_exposure_time_to_shs']
to_time = row['end_of_exposure_time_to_shs']
start_exposure = row['date_of_exposure_to_shs'].replace(
hour=from_time.hour, minute=from_time.minute,
second=from_time.second).to_pydatetime() + timedelta(
hours=int(corrections.loc[subject_id, 'shs_times_difference']))
if not pd.isnull(row['end_date_of_exposure_to_shs']):
end_exposure = row['end_date_of_exposure_to_shs'].replace(
hour=to_time.hour, minute=to_time.minute,
second=to_time.second).to_pydatetime() + timedelta(
hours=int(corrections.loc[subject_id, 'shs_times_difference']))
else:
end_exposure = row['date_of_exposure_to_shs'].replace(
hour=to_time.hour, minute=to_time.minute,
second=to_time.second).to_pydatetime() + timedelta(
hours=int(corrections.loc[subject_id, 'shs_times_difference']))
# Load recording period
from_time = row['start_time_of_monitoring']
start_recording = row['start_date_of_monitoring'].replace(
hour=from_time.hour, minute=from_time.minute,
second=from_time.second).to_pydatetime() + timedelta(
hours=int(corrections.loc[subject_id,
'recording_times_difference']))
to_time = row['end_time_of_monitoring']
end_recording = row['end_date_of_monitoring'].replace(
hour=to_time.hour, minute=to_time.minute,
second=to_time.second).to_pydatetime() + timedelta(
hours=int(corrections.loc[subject_id,
'recording_times_difference']))
# Look up timezone
tz = timezone(project_mapping[subject_id[:3]][1])
print("Creating pixelgram for subject {}".format(subject_id))
plot_combined_pixelgram_dublin(
subject_id,
respeck_data,
airspeck_data,
exposure_period=[
tz.localize(start_exposure),
tz.localize(end_exposure)
],
recording_period=[
tz.localize(start_recording),
tz.localize(end_recording)
],
overwrite_if_already_exists=overwrite_pixelgram_if_already_exists)
def download_data_and_plot_combined_pixelgram(
subject_id,
timeframe=None,
filter_out_not_worn_respeck=True,
overwrite_pixelgram_if_already_exists=False,
subject_visit_number=None,
overwrite_data_if_already_exists=False,
upload_type='automatic'):
project_name = get_project_for_subject(subject_id)
plot_dir = project_mapping[project_name][3]
if subject_visit_number is None:
label_files = "{}".format(subject_id)
else:
label_files = "{}({})".format(subject_id, subject_visit_number)
pixelgram_filepath = plot_dir + "{}_combined_pixelgram.png".format(
label_files)
# Check if pixelgram already exists
if not overwrite_pixelgram_if_already_exists and os.path.isfile(
pixelgram_filepath):
print("Pixelgram for subject {} already exists. Skipping subject.".
format(label_files))
return
# Download data if not present
download_respeck_and_personal_airspeck_data(
subject_id,
upload_type=upload_type,
timeframe=timeframe,
overwrite_if_already_exists=overwrite_data_if_already_exists,
subject_visit_number=subject_visit_number)
# Load data and create plot
respeck_data = load_respeck_file(
subject_id,
project_name=project_name,
upload_type=upload_type,
subject_visit_number=subject_visit_number,
filter_out_not_worn=filter_out_not_worn_respeck)
airspeck_data = load_personal_airspeck_file(
subject_id,
project_name=project_name,
upload_type=upload_type,
subject_visit_number=subject_visit_number)
if len(respeck_data) == 0:
print("RESpeck data for subject {} empty. Skipping subject.".format(
label_files))
return
if len(airspeck_data) == 0:
print("Airspeck data for subject {} empty. Skipping subject.".format(
label_files))
return
if timeframe is not None:
tz = timezone(project_mapping[project_name][1])
if timeframe[0].tzinfo is None:
start_time = tz.localize(timeframe[0])
end_time = tz.localize(timeframe[1])
else:
start_time = timeframe[0]
end_time = timeframe[1]
plot_combined_pixelgram(
subject_id,
respeck_data[start_time:end_time],
airspeck_data[start_time:end_time],
pixelgram_filepath,
overwrite_if_already_exists=overwrite_pixelgram_if_already_exists,
subject_visit_number=subject_visit_number)
else:
plot_combined_pixelgram(
subject_id,
respeck_data,
airspeck_data,
pixelgram_filepath,
overwrite_if_already_exists=overwrite_pixelgram_if_already_exists,
subject_visit_number=subject_visit_number)