app.py

import streamlit as st
import pandas as pd
import numpy as np

# Function to generate synthetic data
def generate_synthetic_data(num_members):
    unique_ids = [f"MEM_{i:05d}" for i in range(1, num_members + 1)]
    primary_keys = [f"PPK_{i:05d}" for i in range(1, num_members + 1)]

    # Synthetic Enrollments
    enrollments_data = {
        "MEM_AGE": np.random.randint(18, 80, num_members),
        "MEM_MSA_NAME": np.random.choice(["DETROIT", "HONOLULU", "LOS ANGELES"], num_members),
        "MEM_STAT": np.random.choice(["ACTIVE", "INACTIVE"], num_members),
        "MEMBER_ID": unique_ids,
        "PRIMARY_PERSON_KEY": primary_keys,
        "PAYER_LOB": np.random.choice(["MEDICAID", "COMMERCIAL", "MEDICARE"], num_members),
        "PAYER_TYPE": np.random.choice(["PPO", "HMO"], num_members),
        "PRIMARY_CHRONIC_CONDITION_ROLLUP_DESC": np.random.choice(["Cancer", "Diabetes", "Hypertension"], num_members),
        "Count of PRIMARY_CHRONIC_CONDITION_ROLLUP_ID": np.random.randint(1, 5, num_members),
        "PROD_TYPE": np.random.choice(["DENTAL", "VISION", "MEDICAL"], num_members),
        "RELATION": np.random.choice(["SUBSCRIBER", "DEPENDENT"], num_members),
        "Sum of QTY_MM_DN": np.random.randint(0, 10, num_members),
        "Sum of QTY_MM_MD": np.random.randint(0, 10, num_members),
        "Sum of QTY_MM_RX": np.random.randint(0, 10, num_members),
        "Sum of QTY_MM_VS": np.random.randint(0, 10, num_members),
        "YEARMO": np.random.randint(202201, 202412, num_members),
    }
    enrollments_df = pd.DataFrame(enrollments_data)

    # Synthetic Members
    members_data = {
        "MEM_ETHNICITY": np.random.choice(["Hispanic", "Non-Hispanic", None], num_members),
        "MEM_GENDER": ["F"] * num_members,  # Ensuring all members are female
        "MEM_MSA_NAME": enrollments_data["MEM_MSA_NAME"],
        "MEM_RACE": np.random.choice(["White", "Black", "Asian", None], num_members),
        "MEM_STATE": np.random.choice(["MI", "HI", "CA"], num_members),
        "MEM_ZIP3": np.random.randint(100, 999, num_members),
        "MEMBER_ID": unique_ids,
        "PRIMARY_PERSON_KEY": primary_keys,
    }
    members_df = pd.DataFrame(members_data)

    # Synthetic Providers
    providers_data = {
        "PROV_CLINIC_STATE": np.random.choice(["MI", "HI", "CA"], num_members),
        "PROV_CLINIC_ZIP": np.random.randint(10000, 99999, num_members),
        "PROV_KEY": [f"PK_{i:05d}" for i in range(1, num_members + 1)],
        "PROV_NPI_ORG": np.random.randint(1, 50, num_members),
        "PROV_TAXONOMY": np.random.choice(["208100000X", "207RE0101X"], num_members),
        "PROV_TYPE": np.random.choice(["Type1", "Type2"], num_members),
    }
    providers_df = pd.DataFrame(providers_data)

    # Synthetic Services
    services_data = {
        "MEMBER_ID": unique_ids,
        "PRIMARY_PERSON_KEY": primary_keys,
        "Sum of AMT_ALLOWED": np.random.uniform(1000, 10000, num_members),
        "Sum of AMT_BILLED": np.random.uniform(1000, 15000, num_members),
        "Count of AMT_PAID": np.random.randint(1, 5, num_members),
        "ATT_PROV_KEY": np.random.choice(providers_data["PROV_KEY"], num_members),
        "BILL_PROV_KEY": np.random.choice(providers_data["PROV_KEY"], num_members),
        "CLAIM_IN_NETWORK": np.random.choice(["Y", "N", None], num_members),
        "RELATION": enrollments_data["RELATION"],
        "SERVICE_SETTING": np.random.choice(["OUTPATIENT", "INPATIENT"], num_members),
        "Sum of SERVICE_LINE": np.random.randint(1, 10, num_members),
        "Sum of SV_UNITS": np.random.randint(1, 100, num_members),
        "YEARMO": enrollments_data["YEARMO"],
    }
    services_df = pd.DataFrame(services_data)

    return enrollments_df, members_df, providers_df, services_df


# Streamlit App
st.title("Synthetic Medical Billing Data Generator")

# Slider for number of members
num_members = st.slider("Select number of unique members:", min_value=10, max_value=1000, step=10, value=100)

# Generate synthetic data
enrollments_df, members_df, providers_df, services_df = generate_synthetic_data(num_members)

# Display dataframes
st.subheader("Preview of Generated Data")
st.write("Enrollments Data")
st.dataframe(enrollments_df.head())
st.write("Members Data")
st.dataframe(members_df.head())
st.write("Providers Data")
st.dataframe(providers_df.head())
st.write("Services Data")
st.dataframe(services_df.head())

# Allow downloading the generated files
st.subheader("Download Synthetic Data")
st.download_button(
    label="Download Enrollments Data",
    data=enrollments_df.to_csv(index=False),
    file_name="Synthetic_Enrollments.csv",
    mime="text/csv",
)
st.download_button(
    label="Download Members Data",
    data=members_df.to_csv(index=False),
    file_name="Synthetic_Members.csv",
    mime="text/csv",
)
st.download_button(
    label="Download Providers Data",
    data=providers_df.to_csv(index=False),
    file_name="Synthetic_Providers.csv",
    mime="text/csv",
)
st.download_button(
    label="Download Services Data",
    data=services_df.to_csv(index=False),
    file_name="Synthetic_Services.csv",
    mime="text/csv",
)