Python Health Data Visualization: Plotly, Pandas & GeoDa

You will need this additional code:

# Means and STD

distributions = [

    {"mean":62.88, "std":18.2, "color":"#1f77b4","label":"Media 62.88"},

    {"mean":68.39, "std":16.63, "color":"#ff7f0e","label":"Media 68.39"},

    {"mean":71.56, "std":13.48, "color":"#2ca02c","label":"Media 71.56"},

    {"mean":56.52, "std":14.24, "color":"#d67228","label":"Media 56.52"},

    {"mean":72.75, "std":12.4, "color":"#9467bd","label":"Media 72.88"}

]

#Define x range

x= np.linspace(30, 100.500)

plt.figure(figsize=(10,6))

# Define normal distribution formula

def normal_pfd(x, mean, std):

    return(1/(std*np.sqrt(2*np.pi)))*np.exp(-0.5*((x-mean)/std)**2)

You need this code for your Sankey Chart:

import pandas as pd

import plotly.graph_objects as go

df = pd.read_csv("DepartamentOfNotificationLeukaemias.csv")

df.columns = df.columns.str.strip().str.upper()

df['NUMBER OF PEOPLE'] = pd.to_numeric(df['NUMBER OF PEOPLE'], errors='coerce').fillna(0)

origins = df['HOUSEHOLD'].unique()

targets = df['NOTIFICATION FACILITY'].unique()

nodes = list(origins) + [t for t in targets if t not in origins]

node_dict = {node: idx for idx, node in enumerate(nodes)}

df['Source'] = df['HOUSEHOLD'].map(node_dict)

df['Target'] = df['NOTIFICATION FACILITY'].map(node_dict)

incoming = df.groupby('Target')['NUMBER OF PEOPLE'].sum()

outgoing = df.groupby('Source')['NUMBER OF PEOPLE'].sum()

node_colors = []

for node in nodes:

    idx = node_dict[node]

    in_flow = incoming.get(idx, 0)

    out_flow = outgoing.get(idx, 0)

    if in_flow > out_flow:

        node_colors.append("rgba(214, 39, 40, 0.8)")

    else:

        node_colors.append("rgba(44, 160, 101, 0.8)")

x_positions = [0.01 if node in origins else 0.99 for node in nodes]

fig = go.Figure(data=[go.Sankey(

    arrangement="snap",

    node=dict(

        pad=20,

        thickness=20,

        line=dict(color="black", width=0.5),

        label=nodes,

        color=node_colors,

        x=x_positions

    ),

    link=dict(

        source=df['Source'],

        target=df['Target'],

        value=df['NUMBER OF PEOPLE'],

        color="rgba(160,160,160,0.3)"

    )

)])

fig.update_layout(

    title_text="Notificación de casos de leucemia por departamento de origen y atención",

    font=dict(size=12),

    height=800,

    margin=dict(l=30, r=30, t=60, b=30)

)

fig.show()