Skip to content

Added code to assign each individual facilities at different levels #1614

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
RachelMurray-Watson wants to merge 1 commit into from
Closed

Added code to assign each individual facilities at different levels #1614

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
142 changes: 140 additions & 2 deletions src/tlo/methods/demography.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,8 +11,27 @@
from types import MappingProxyType
from typing import Union

"""
The core demography module and its associated events.
* Sets initial population size
* Determines the 'is_alive', age-related, and residential location properties.
* Runs the OtherDeathPoll which represents the deaths due to causes other than those represented by disease modules.
"""

import math
from collections import defaultdict
from pathlib import Path
from types import MappingProxyType
from typing import Union
try:
from shapely.geometry import Point
except ImportError:
shapely.geometry = None
import random
import numpy as np
import pandas as pd
import geopandas as gpd
from scipy.spatial import cKDTree

from tlo import (
DAYS_IN_MONTH,
Expand Down Expand Up @@ -68,7 +87,6 @@ def age_at_date(
# Assume a fixed number of days in all years, ignoring variations due to leap years
return (date - date_of_birth) / pd.Timedelta(days=DAYS_IN_YEAR)


class Demography(Module):
"""
The core demography module.
Expand Down Expand Up @@ -114,6 +132,8 @@ def __init__(self, name=None, resourcefilepath=None, equal_allocation_by_distric
'gbd_causes_of_death_data': Parameter(Types.DATA_FRAME,
'Proportion of deaths in each age/sex group attributable to each possible'
' cause of death in the GBD dataset.'),
'possible_facilities':Parameter(Types.DATA_FRAME,
'Possible facilities for individuals to be assigned'),
}

# Next we declare the properties of individuals that this module provides.
Expand All @@ -134,7 +154,7 @@ def __init__(self, name=None, resourcefilepath=None, equal_allocation_by_distric
),

'district_num_of_residence': Property(
Types.CATEGORICAL,
Types.CATEGORICAL,
'The district number in which the person is resident',
categories=['SET_AT_RUNTIME']
),
Expand All @@ -151,6 +171,21 @@ def __init__(self, name=None, resourcefilepath=None, equal_allocation_by_distric
categories=['SET_AT_RUNTIME']
),

'level_0': Property(Types.CATEGORICAL,
'Which level 0 facility is "used" by individual.',
categories=['SET_AT_RUNTIME']),
'level_1a': Property(Types.CATEGORICAL,
'Which level 1a facility is "used" by individual.',
categories = ['SET_AT_RUNTIME']),
'level_1b': Property(Types.CATEGORICAL,
'Which level 1b facility is "used" by individual.',
categories=['SET_AT_RUNTIME']),
'level_2': Property(Types.CATEGORICAL,
'Which level 2 facility is "used" by individual.',
categories=['SET_AT_RUNTIME']),
'level_3': Property(Types.CATEGORICAL,
'Which level 3 facility is "used" by individual.',
categories=['SET_AT_RUNTIME']),
# Age calculation is handled by demography module
'age_exact_years': Property(Types.REAL, 'The age of the individual in exact years'),
'age_years': Property(Types.INT, 'The age of the individual in years'),
Expand Down Expand Up @@ -205,6 +240,9 @@ def read_parameters(self, data_folder):
Path(self.resourcefilepath) / 'gbd' / 'ResourceFile_CausesOfDeath_GBD2019.csv'
).set_index(['Sex', 'Age_Grp'])

# possible facilities for all levels
self.parameters['possible_facilities'] = pd.read_csv(Path(self.resourcefilepath) / 'climate_change_impacts'/'facilities_with_lat_long_region.csv')['Fname']

def pre_initialise_population(self):
"""
1) Store all the cause of death represented in the imported GBD data
Expand Down Expand Up @@ -242,6 +280,38 @@ def pre_initialise_population(self):
categories=self.parameters['pop_2010']['Region'].unique().tolist()
)

self.PROPERTIES['level_0'] = Property(
Types.CATEGORICAL,
'Which level 0 facility is "used" by individual.',
categories= self.parameters['possible_facilities'].unique().tolist()
)

self.PROPERTIES['level_1a'] = Property(
Types.CATEGORICAL,
'Which level 1a facility is "used" by individual.',
categories= self.parameters['possible_facilities'].unique().tolist()
)

self.PROPERTIES['level_1b'] = Property(
Types.CATEGORICAL,
'Which level 1b facility is "used" by individual.',
categories= self.parameters['possible_facilities'].unique().tolist()
)

self.PROPERTIES['level_2'] = Property(
Types.CATEGORICAL,
'Which level 2 facility is "used" by individual.',
categories= self.parameters['possible_facilities'].unique().tolist()
)

self.PROPERTIES['level_3'] = Property(
Types.CATEGORICAL,
'Which level 3 facility is "used" by individual.',
categories= self.parameters['possible_facilities'].unique().tolist()
)



def initialise_population(self, population):
"""Set properties for this module and compute the initial population scaling factor"""
df = population.props
Expand Down Expand Up @@ -293,6 +363,12 @@ def initialise_population(self, population):
df.loc[df.is_alive, 'age_days'] = (
self.sim.date - demog_char_to_assign['date_of_birth']
).dt.days
self.assign_closest_facility_level()
df.loc[df.is_alive, 'facility_used_level_0'] = self.sim.population.props.loc[df.is_alive, "level_0"]
df.loc[df.is_alive, 'facility_used_level_1a'] = self.sim.population.props.loc[df.is_alive, "level_1a"]
df.loc[df.is_alive, 'facility_used_level_1b'] = self.sim.population.props.loc[df.is_alive, "level_1b"]
df.loc[df.is_alive, 'facility_used_level_2'] = self.sim.population.props.loc[df.is_alive, "level_2"]
df.loc[df.is_alive, 'facility_used_level_3'] = self.sim.population.props.loc[df.is_alive, "level_3"]

# Ensure first individual in df is a man, to safely exclude person_id=0 from selection of direct birth mothers.
# If no men are found in df, issue a warning and proceed with female individual at person_id = 0.
Expand Down Expand Up @@ -395,6 +471,68 @@ def _edit_init_pop_to_prevent_persons_greater_than_max_age(self, df, max_age: in
_df.prob = _df.prob / _df.prob.sum() # Rescale `prob` so that it sums to 1.0
return _df.reset_index(drop=True)

def assign_closest_facility_level(self):
"""Function that assigns an individual coordinates,
and then the facilities at which they recieve care at each level."""
# Load district polygons from shapefile
malawi_admin2 = gpd.read_file(Path(self.resourcefilepath/'mapping'/'ResourceFile_mwi_admbnda_adm2_nso_20181016.shp'))
worldpop_gdf = gpd.read_file(Path(self.resourcefilepath / 'climate_change_impacts' / 'worldpop_density_with_districts.shp'))
worldpop_gdf['Z_prop'] = pd.to_numeric(worldpop_gdf['Z_prop'], errors='coerce') # even when saved as numeric, read in as string

# Create a lookup from district name to its polygon
district_to_geometry = {row["ADM2_EN"]: row["geometry"] for _, row in malawi_admin2.iterrows()}
def sample_point_within_polygon(polygon):
"""Randomly sample a point within a given polygon."""
minx, miny, maxx, maxy = polygon.bounds
while True:
random_point = Point(random.uniform(minx, maxx), random.uniform(miny, maxy))
if polygon.contains(random_point):
return random_point

def assign_coordinate_by_population_weight(district_name):
"""Assigns a coordinate within the district, weighted by population density (Z_proportion)."""
subset = worldpop_gdf[worldpop_gdf["ADM2_EN"] == district_name]
if subset.empty:
return None
chosen_point = subset.sample(weights="Z_prop").iloc[0]["geometry"]
return chosen_point


# Assign unique coordinates to each individual based on their district
df = self.sim.population.props.copy() # take copy of dataframe
df["coordinate_of_residence"] = df["district_of_residence"].apply(
assign_coordinate_by_population_weight)
facility_info =pd.read_csv(Path(self.resourcefilepath) / 'climate_change_impacts' / "facilities_with_lat_long_region.csv")# these are ones that were included in the regression model

facility_levels = {
"level_0": ["Health Post", "Village Health Committee", "Community Health Station", "Village Clinic", "Mobile Clinic", "Outreach Clinic"],
"level_1a": [
"Dispensary", "Rural Health Centre", "Urban Health Centre", "Private Clinic", "Special Clinic",
"Antenatal Clinic", "Maternity Clinic", "Maternity Facility"
],
"level_1b": ["Community Hospital", "Rural Hospital", "CHAM Hospital"],
"level_2": ["District Hospital", "District Health Office"],
"level_3": ["Kamuzu Central Hospital", "Mzuzu Central Hospital", "Zomba Central Hospital",
"Queen Elizabeth Central Hospital"],
"level_4": ["Zomba Mental Hospital"]
}

individual_coords = np.array([
(point.x, point.y) if point else (np.nan, np.nan)
for point in df["coordinate_of_residence"]
])

for level, facility_types in facility_levels.items():
relevant_facilities = facility_info[facility_info["Ftype"].isin(facility_types)]
if not relevant_facilities.empty:
facility_coords = list(
zip(relevant_facilities["A109__Longitude"], relevant_facilities["A109__Latitude"]))
facility_tree = cKDTree(facility_coords)
distances, indices = facility_tree.query(individual_coords, k=1, workers=-1)
df[level] = relevant_facilities.iloc[indices].reset_index(drop=True)["Fname"]
df.drop("coordinate_of_residence", inplace=True, axis=1)
self.sim.population.props = df

@staticmethod
def _edit_init_pop_so_that_equal_number_in_each_district(df) -> pd.DataFrame:
"""Return an edited version of the `pd.DataFrame` describing the probability of persons in the population being
Expand Down