Update app.py

app.py

@@ -4,27 +4,22 @@ import random
 import plotly.graph_objects as go
 import plotly.express as px
 
-url = 'https://www.cms.gov/icd10m/version36-fullcode-cms/fullcode_cms/P0001.zip'
-df_icd10 = pd.read_csv(url, sep='\t', compression='zip')
-df_icd10 = df_icd10[['MS-DRG', 'MS-DRG Title', 'ICD-10-PCS', 'ICD-10-PCS Description', 'ICD-10-CM Code', 'ICD-10-CM Code Description']]
-df_icd10 = df_icd10[df_icd10['MS-DRG Title'].str.contains('Diseases of the Circulatory System', case=False)]
-
 # Define the states and conditions of interest
 states = ["Minnesota", "Florida", "California"]
 top_n = 10
 
 # Define the list dictionary of top 10 health conditions descending by cost, with emojis
 health_conditions = [
-    {"condition": "💔 Heart disease", "emoji": "💗", "icd10": ["I20", "I25", "I50"], "spending": 214.3},
-    {"condition": "🤕 Trauma-related disorders", "emoji": "🚑", "icd10": ["S00-T98"], "spending": 198.6},
-    {"condition": "🦀 Cancer", "emoji": "🎗️", "icd10": ["C00-C96"], "spending": 171.0},
-    {"condition": "🧠 Mental disorders", "emoji": "🧘", "icd10": ["F00-F99"], "spending": 150.8},
-    {"condition": "🦴 Osteoarthritis and joint disorders", "emoji": "🏥", "icd10": ["M00-M25", "M40-M54"], "spending": 142.4},
-    {"condition": "💉 Diabetes", "emoji": "🩸", "icd10": ["E10-E14"], "spending": 107.4},
-    {"condition": "🫁 Chronic obstructive pulmonary disease and asthma", "emoji": "🫀", "icd10": ["J40-J47"], "spending": 91.0},
-    {"condition": "🩺 Hypertension", "emoji": "💉", "icd10": ["I10-I15"], "spending": 83.9},
-    {"condition": "🔬 Hyperlipidemia", "emoji": "🔬", "icd10": ["E78"], "spending": 83.9},
-    {"condition": "🦴 Back problems", "emoji": "🧍", "icd10": ["M40-M54"], "spending": 67.0}
+    {"condition": "💔 Heart disease", "emoji": "💗", "spending": 214.3},
+    {"condition": "🤕 Trauma-related disorders", "emoji": "🚑", "spending": 198.6},
+    {"condition": "🦀 Cancer", "emoji": "🎗️", "spending": 171.0},
+    {"condition": "🧠 Mental disorders", "emoji": "🧘", "spending": 150.8},
+    {"condition": "🦴 Osteoarthritis and joint disorders", "emoji": "🏥", "spending": 142.4},
+    {"condition": "💉 Diabetes", "emoji": "🩸", "spending": 107.4},
+    {"condition": "🫁 Chronic obstructive pulmonary disease and asthma", "emoji": "🫀", "spending": 91.0},
+    {"condition": "🩺 Hypertension", "emoji": "💉", "spending": 83.9},
+    {"condition": "🔬 Hyperlipidemia", "emoji": "🔬", "spending": 83.9},
+    {"condition": "🦴 Back problems", "emoji": "🧍", "spending": 67.0}
 ]
 
 # Create a DataFrame from the list dictionary
@@ -34,32 +29,28 @@ df_top_conditions = pd.DataFrame(health_conditions)
 total_spending = round(df_top_conditions["spending"].sum(), 1)
 
 # Define the roll function
-def roll(icd10):
-    # Filter the CMS data to include only the given ICD-10 codes
-    df_filtered = df_icd10[df_icd10['ICD-10-CM Code'].str.contains(icd10, case=False)]
-    # Group the data by ICD--10-CM Code and sum the spending for each code
-    df_grouped = df_filtered.groupby(['ICD-10-CM Code', 'ICD-10-CM Code Description']).agg({'spending': 'sum'}).reset_index()
-    # Sort the data by spending in descending order
-    df_sorted = df_grouped.sort_values('spending', ascending=False)
-    # Return the top 3 variants
-    return df_sorted.head(3)
-
-#Define the bar chart
-fig_bar = go.Figure()
-
-#Iterate over each health condition and add a bar to the chart
-for i, condition in df_top_conditions.iterrows():
-    icd10 = '|'.join(condition['icd10'])
-    df_top_variants = roll(icd10)
-    fig_bar.add_trace(go.Bar(
-        x=df_top_variants['ICD-10-CM Code Description'],
-        y=df_top_variants['spending'],
-        name=condition['condition'],
-        marker=dict(color=f"rgba{px.colors.qualitative.Plotly[i]}")
-        ))
-
-# Customize the layout of the bar chart
-fig_bar.update_layout(barmode='group', title=f"Top 3 Variants by Spending for Each Health Condition in {', '.join(states)}", xaxis_title='ICD-10-CM Code Description', yaxis_title='Spending')
-
-# Display the bar chart in the Streamlit app
-st.plotly_chart(fig_bar)
+def roll():
+    # Generate a thousand 10-sided dice rolls for each health condition
+    rolls = [random.randint(1, 10) for _ in range(1000)]
+    # Calculate the frequency of each variant
+    frequencies = [rolls.count(i) for i in range(1, 11)]
+    return frequencies
+
+# Define the sunburst chart
+fig_sunburst = go.Figure(go.Sunburst(
+    labels=df_top_conditions["emoji"] + " " + df_top_conditions["condition"],
+    parents=[""] * top_n,
+    values=df_top_conditions["spending"],
+    maxdepth=2
+))
+
+# Customize the layout of the sunburst chart
+fig_sunburst.update_layout(title=f"Top {top_n} Health Conditions in {', '.join(states)} by Spending (Total: ${total_spending}B)")
+
+# Display the sunburst chart and variants per condition in the Streamlit app
+st.plotly_chart(fig_sunburst)
+for index, row in df_top_conditions.iterrows():
+    frequencies = roll()
+    fig_bar = px.bar(x=[f"Variant {i}" for i in range(1, 11)], y=frequencies[:10], labels={'x': 'Variant', 'y': 'Frequency'})
+    fig_bar.update_layout(title=f"Variants of {row['condition']} ({row['emoji']})")
+    st.plotly_chart(fig_bar)