app.py

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

# Seed for reproducibility
np.random.seed(42)

# Function to generate synthetic Enrollments
def generate_enrollments(num_members):
    primary_keys = [f"PPK_{i+1:05d}" for i in range(num_members)]
    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": primary_keys,
        "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),
        "YEARMO": np.random.randint(202201, 202412, num_members),
    }
    return pd.DataFrame(enrollments_data)

# Function to generate synthetic Members
def generate_members(num_members):
    primary_keys = [f"PPK_{i+1:05d}" for i in range(num_members)]
    members_data = {
        "MEM_ETHNICITY": np.random.choice(["Hispanic", "Non-Hispanic", None], num_members),
        "MEM_GENDER": ["F"] * num_members,
        "MEM_MSA_NAME": np.random.choice(["DETROIT", "HONOLULU", "LOS ANGELES"], num_members),
        "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": primary_keys,
        "PRIMARY_PERSON_KEY": primary_keys,
    }
    return pd.DataFrame(members_data)

# Function to generate synthetic Providers
def generate_providers(num_providers):
    providers_data = {
        "PROV_CLINIC_STATE": np.random.choice(["MI", "HI", "CA"], num_providers),
        "PROV_CLINIC_ZIP": np.random.randint(10000, 99999, num_providers),
        "PROV_KEY": [f"PK_{i+1:05d}" for i in range(num_providers)],
        "Sum of PROV_NPI_ORG": np.random.randint(1, 50, num_providers),
        "PROV_TAXONOMY": np.random.choice(["208100000X", "207RE0101X"], num_providers),
        "PROV_TYPE": np.random.choice(["Type1", "Type2"], num_providers),
    }
    return pd.DataFrame(providers_data)

# Function to generate synthetic Services
def generate_services(num_services, primary_keys):
    services_data = {
        "PRIMARY_PERSON_KEY": np.random.choice(primary_keys, num_services),
        "Sum of AMT_ALLOWED": np.random.uniform(1000, 10000, num_services),
        "Sum of AMT_BILLED": np.random.uniform(1000, 15000, num_services),
        "Count of AMT_PAID": np.random.randint(1, 5, num_services),
        "ATT_PROV_KEY": [f"PK_{i+1:05d}" for i in np.random.randint(1, len(primary_keys), num_services)],
        "BILL_PROV_KEY": [f"PK_{i+1:05d}" for i in np.random.randint(1, len(primary_keys), num_services)],
        "CLAIM_IN_NETWORK": np.random.choice(["Y", "N", None], num_services),
        "RELATION": np.random.choice(["SUBSCRIBER", "DEPENDENT"], num_services),
        "SERVICE_SETTING": np.random.choice(["OUTPATIENT", "INPATIENT"], num_services),
        "Sum of SERVICE_LINE": np.random.randint(1, 10, num_services),
        "Sum of SV_UNITS": np.random.randint(1, 100, num_services),
        "YEARMO": np.random.randint(202201, 202412, num_services),
    }
    return pd.DataFrame(services_data)

# Function to generate synthetic BreastCancer data
def generate_breast_cancer_data(num_patients):
    patient_ids = [f"PPK_{i+1:05d}" for i in range(num_patients)]
    breast_cancer_data = {
        "Patient ID": patient_ids,
        "Age": np.random.randint(30, 80, num_patients),
        "Menopausal Status": np.random.choice(["Post-menopausal", "Pre-menopausal"], num_patients),
        "Tumor Size (cm)": np.round(np.random.lognormal(mean=0.7, sigma=0.5, size=num_patients), 2),
        "Lymph Node Involvement": np.random.choice(["Positive", "Negative"], num_patients),
        "Tumor Grade": np.random.choice([1, 2, 3], num_patients),
        "Tumor Stage": np.random.choice(["I", "II", "III", "IV"], num_patients),
        "ER Status": np.random.choice(["Positive", "Negative"], num_patients),
        "PR Status": np.random.choice(["Positive", "Negative"], num_patients),
        "HER2 Status": np.random.choice(["Positive", "Negative"], num_patients),
        "Ki-67 Level": np.random.choice(["High", "Low"], num_patients),
        "TNBC Status": np.random.choice(["Positive", "Negative"], num_patients),
        "BRCA Mutation": np.random.choice(["Positive", "Negative"], num_patients),
        "Overall Health": np.random.choice(["Good", "Poor"], num_patients),
        "Genomic Recurrence Score": np.random.choice(["Low", "Intermediate", "High", "N/A"], num_patients),
        "Treatment": np.random.choice(["Surgery", "Chemotherapy", "Radiation Therapy"], num_patients),
    }
    return pd.DataFrame(breast_cancer_data)

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

# Input parameters
num_members = st.slider("Number of Members to Generate", 10, 1000, 100)
num_providers = st.slider("Number of Providers to Generate", 10, 500, 100)
num_services = st.slider("Number of Services to Generate", 10, 2000, 500)
num_patients = st.slider("Number of Breast Cancer Patients to Generate", 10, 500, 100)

if st.button("Generate Data"):
    enrollments_df = generate_enrollments(num_members)
    members_df = generate_members(num_members)
    providers_df = generate_providers(num_providers)
    services_df = generate_services(num_services, enrollments_df["PRIMARY_PERSON_KEY"].tolist())
    breast_cancer_df = generate_breast_cancer_data(num_patients)

    # Display data
    st.subheader("Enrollments Data")
    st.dataframe(enrollments_df.head())
    st.download_button("Download Enrollments", enrollments_df.to_csv(index=False), "enrollments.csv")

    st.subheader("Members Data")
    st.dataframe(members_df.head())
    st.download_button("Download Members", members_df.to_csv(index=False), "members.csv")

    st.subheader("Providers Data")
    st.dataframe(providers_df.head())
    st.download_button("Download Providers", providers_df.to_csv(index=False), "providers.csv")

    st.subheader("Services Data")
    st.dataframe(services_df.head())
    st.download_button("Download Services", services_df.to_csv(index=False), "services.csv")

    st.subheader("Breast Cancer Data")
    st.dataframe(breast_cancer_df.head())
    st.download_button("Download Breast Cancer Data", breast_cancer_df.to_csv(index=False), "breast_cancer.csv")