403 lines
11 KiB
Python
403 lines
11 KiB
Python
"""
|
||
心血管疾病数据仪表板
|
||
Streamlit应用程序，用于心血管疾病数据的清洗、特征工程和交互式可视化
|
||
终端启动程序命令
|
||
streamlit run D:\AI_Class\PyCharm\Work_Space\CardAI\module1_dashboard\cardio_dashboard.py
|
||
"""
|
||
|
||
import streamlit as st
|
||
import pandas as pd
|
||
import numpy as np
|
||
import plotly.express as px
|
||
import plotly.graph_objects as go
|
||
from pathlib import Path
|
||
import warnings
|
||
warnings.filterwarnings('ignore')
|
||
|
||
# 设置页面配置
|
||
st.set_page_config(
|
||
page_title="心血管疾病数据分析仪表板",
|
||
page_icon="❤️",
|
||
layout="wide",
|
||
initial_sidebar_state="expanded"
|
||
)
|
||
|
||
# 常量定义 - 使用绝对路径
|
||
DATA_PATH = Path("D:/AI_Class/PyCharm/Work_Space/CardAI/data/心血管疾病.xlsx")
|
||
# BMI分类（中国标准）
|
||
BMI_CATEGORIES = {
|
||
"偏瘦": (0, 18.5),
|
||
"正常": (18.5, 24),
|
||
"超重": (24, 28),
|
||
"肥胖": (28, float('inf'))
|
||
}
|
||
CHOLESTEROL_MAP = {1: "正常", 2: "高于正常", 3: "远高于正常"}
|
||
GLUC_MAP = {1: "正常", 2: "高于正常", 3: "远高于正常"}
|
||
GENDER_MAP = {1: "女性", 2: "男性"}
|
||
|
||
# 数据加载和清洗函数（使用缓存）
|
||
@st.cache_data
|
||
def load_and_clean_data(file_path):
|
||
"""
|
||
加载并清洗心血管疾病数据
|
||
|
||
Args:
|
||
file_path: Excel文件路径
|
||
|
||
Returns:
|
||
pandas.DataFrame: 清洗后的数据框
|
||
"""
|
||
try:
|
||
# 加载数据
|
||
df = pd.read_excel(file_path)
|
||
|
||
# 1. 特征工程
|
||
# 将年龄从天数转换为年（四舍五入）
|
||
df['age_years'] = round(df['age'] / 365.25).astype(int)
|
||
|
||
# 计算BMI
|
||
df['BMI'] = df['weight'] / ((df['height'] / 100) ** 2)
|
||
|
||
# 2. 异常值处理
|
||
# 删除舒张压 ≥ 收缩压的记录
|
||
df = df[df['ap_lo'] < df['ap_hi']]
|
||
|
||
# 删除血压极端异常值（保留收缩压∈[90,250]，舒张压∈[60,150]）
|
||
df = df[(df['ap_hi'] >= 90) & (df['ap_hi'] <= 250)]
|
||
df = df[(df['ap_lo'] >= 60) & (df['ap_lo'] <= 150)]
|
||
|
||
# 3. 类别转换
|
||
# 胆固醇水平转换
|
||
df['cholesterol_str'] = df['cholesterol'].map(CHOLESTEROL_MAP)
|
||
|
||
# 血糖水平转换
|
||
df['gluc_str'] = df['gluc'].map(GLUC_MAP)
|
||
|
||
# 性别转换
|
||
df['gender_str'] = df['gender'].map(GENDER_MAP)
|
||
|
||
# 心血管疾病状态转换
|
||
df['cardio_str'] = df['cardio'].map({0: "无心血管疾病", 1: "有心血管疾病"})
|
||
|
||
# 4. BMI分类（中国标准）
|
||
def categorize_bmi(bmi):
|
||
for category, (low, high) in BMI_CATEGORIES.items():
|
||
if low <= bmi < high:
|
||
return category
|
||
return "未知"
|
||
|
||
df['bmi_category'] = df['BMI'].apply(categorize_bmi)
|
||
|
||
# 重置索引
|
||
df = df.reset_index(drop=True)
|
||
|
||
return df
|
||
|
||
except FileNotFoundError:
|
||
st.error(f"数据文件未找到: {file_path}")
|
||
return pd.DataFrame()
|
||
except Exception as e:
|
||
st.error(f"数据加载失败: {str(e)}")
|
||
return pd.DataFrame()
|
||
|
||
@st.cache_data
|
||
def filter_data(df, age_range, gender_filter, cardio_filter):
|
||
"""
|
||
根据筛选条件过滤数据
|
||
|
||
Args:
|
||
df: 原始数据框
|
||
age_range: 年龄范围 [min, max]
|
||
gender_filter: 性别筛选列表
|
||
cardio_filter: 心血管疾病筛选列表
|
||
|
||
Returns:
|
||
pandas.DataFrame: 筛选后的数据框
|
||
"""
|
||
filtered_df = df.copy()
|
||
|
||
# 年龄筛选
|
||
filtered_df = filtered_df[
|
||
(filtered_df['age_years'] >= age_range[0]) &
|
||
(filtered_df['age_years'] <= age_range[1])
|
||
]
|
||
|
||
# 性别筛选
|
||
if "全部" not in gender_filter:
|
||
gender_values = [k for k, v in GENDER_MAP.items() if v in gender_filter]
|
||
filtered_df = filtered_df[filtered_df['gender'].isin(gender_values)]
|
||
|
||
# 心血管疾病筛选
|
||
if "全部" not in cardio_filter:
|
||
cardio_values = []
|
||
if "有心血管疾病" in cardio_filter:
|
||
cardio_values.append(1)
|
||
if "无心血管疾病" in cardio_filter:
|
||
cardio_values.append(0)
|
||
filtered_df = filtered_df[filtered_df['cardio'].isin(cardio_values)]
|
||
|
||
return filtered_df
|
||
|
||
def create_age_distribution_chart(df):
|
||
"""
|
||
创建年龄分布直方图
|
||
|
||
Args:
|
||
df: 数据框
|
||
|
||
Returns:
|
||
plotly.graph_objects.Figure: 年龄分布图表
|
||
"""
|
||
fig = px.histogram(
|
||
df,
|
||
x='age_years',
|
||
color='cardio_str',
|
||
nbins=30,
|
||
title='年龄分布（按心血管疾病状态）',
|
||
labels={'age_years': '年龄（岁）', 'count': '人数', 'cardio_str': '心血管疾病状态'},
|
||
color_discrete_map={"有心血管疾病": "#EF553B", "无心血管疾病": "#636EFA"},
|
||
opacity=0.7
|
||
)
|
||
|
||
fig.update_layout(
|
||
bargap=0.1,
|
||
xaxis_title="年龄（岁）",
|
||
yaxis_title="人数",
|
||
legend_title="心血管疾病状态",
|
||
hovermode='x unified'
|
||
)
|
||
|
||
return fig
|
||
|
||
def create_bmi_cardio_chart(df):
|
||
"""
|
||
创建BMI分类对心血管疾病影响的堆叠柱状图
|
||
|
||
Args:
|
||
df: 数据框
|
||
|
||
Returns:
|
||
plotly.graph_objects.Figure: BMI分类图表
|
||
"""
|
||
# 计算交叉表
|
||
cross_tab = pd.crosstab(
|
||
df['bmi_category'],
|
||
df['cardio_str'],
|
||
normalize='index'
|
||
).reset_index()
|
||
|
||
# 转换为长格式
|
||
cross_tab_melted = cross_tab.melt(
|
||
id_vars='bmi_category',
|
||
var_name='cardio_status',
|
||
value_name='percentage'
|
||
)
|
||
|
||
# 创建堆叠柱状图
|
||
fig = px.bar(
|
||
cross_tab_melted,
|
||
x='bmi_category',
|
||
y='percentage',
|
||
color='cardio_status',
|
||
title='BMI分类对心血管疾病的影响',
|
||
labels={
|
||
'bmi_category': 'BMI分类',
|
||
'percentage': '比例',
|
||
'cardio_status': '心血管疾病状态'
|
||
},
|
||
color_discrete_map={"有心血管疾病": "#EF553B", "无心血管疾病": "#636EFA"},
|
||
text_auto='.1%'
|
||
)
|
||
|
||
fig.update_layout(
|
||
xaxis_title="BMI分类",
|
||
yaxis_title="比例",
|
||
legend_title="心血管疾病状态",
|
||
yaxis_tickformat=',.0%',
|
||
hovermode='x unified'
|
||
)
|
||
|
||
return fig
|
||
|
||
def display_summary_stats(df):
|
||
"""
|
||
显示摘要统计信息
|
||
|
||
Args:
|
||
df: 数据框
|
||
"""
|
||
total_records = len(df)
|
||
cardio_cases = df['cardio'].sum()
|
||
cardio_rate = (cardio_cases / total_records * 100) if total_records > 0 else 0
|
||
|
||
col1, col2, col3 = st.columns(3)
|
||
|
||
with col1:
|
||
st.metric(
|
||
label="总记录数",
|
||
value=f"{total_records:,}",
|
||
delta=None
|
||
)
|
||
|
||
with col2:
|
||
st.metric(
|
||
label="心血管疾病病例数",
|
||
value=f"{cardio_cases:,}",
|
||
delta=None
|
||
)
|
||
|
||
with col3:
|
||
st.metric(
|
||
label="心血管疾病风险率",
|
||
value=f"{cardio_rate:.2f}%",
|
||
delta=None
|
||
)
|
||
|
||
def main():
|
||
"""
|
||
主函数：Streamlit应用程序入口
|
||
"""
|
||
# 标题
|
||
st.title("❤️ 心血管疾病数据分析仪表板")
|
||
st.markdown("---")
|
||
|
||
# 加载数据
|
||
with st.spinner("正在加载数据..."):
|
||
df = load_and_clean_data(DATA_PATH)
|
||
|
||
if df.empty:
|
||
st.error("数据加载失败，请检查数据文件路径和格式。")
|
||
return
|
||
|
||
# 侧边栏 - 筛选器
|
||
st.sidebar.header("🔍 数据筛选")
|
||
|
||
# 年龄范围滑块
|
||
age_min = int(df['age_years'].min())
|
||
age_max = int(df['age_years'].max())
|
||
age_range = st.sidebar.slider(
|
||
"选择年龄范围（岁）",
|
||
min_value=age_min,
|
||
max_value=age_max,
|
||
value=[20, 80],
|
||
step=1
|
||
)
|
||
|
||
# 性别筛选器
|
||
gender_options = ["女性", "男性", "全部"]
|
||
gender_filter = st.sidebar.multiselect(
|
||
"选择性别",
|
||
options=gender_options,
|
||
default=["全部"]
|
||
)
|
||
|
||
# 心血管疾病筛选器
|
||
cardio_options = ["有心血管疾病", "无心血管疾病", "全部"]
|
||
cardio_filter = st.sidebar.multiselect(
|
||
"选择心血管疾病状态",
|
||
options=cardio_options,
|
||
default=["全部"]
|
||
)
|
||
|
||
# 应用筛选
|
||
filtered_df = filter_data(df, age_range, gender_filter, cardio_filter)
|
||
|
||
# 显示筛选信息
|
||
st.sidebar.markdown("---")
|
||
st.sidebar.info(f"**筛选结果**: {len(filtered_df):,} 条记录")
|
||
|
||
# 主页面
|
||
# 1. 摘要统计
|
||
st.header("📊 数据摘要")
|
||
display_summary_stats(filtered_df)
|
||
|
||
st.markdown("---")
|
||
|
||
# 2. 年龄分布图表
|
||
st.header("📈 年龄分布分析")
|
||
col1, col2 = st.columns([3, 1])
|
||
|
||
with col1:
|
||
age_chart = create_age_distribution_chart(filtered_df)
|
||
st.plotly_chart(age_chart, use_container_width=True)
|
||
|
||
with col2:
|
||
st.markdown("### 年龄统计")
|
||
st.metric("平均年龄", f"{filtered_df['age_years'].mean():.1f} 岁")
|
||
st.metric("年龄中位数", f"{filtered_df['age_years'].median():.1f} 岁")
|
||
st.metric("年龄标准差", f"{filtered_df['age_years'].std():.1f} 岁")
|
||
|
||
st.markdown("---")
|
||
|
||
# 3. BMI分类分析
|
||
st.header("⚖️ BMI分类分析")
|
||
col1, col2 = st.columns([3, 1])
|
||
|
||
with col1:
|
||
bmi_chart = create_bmi_cardio_chart(filtered_df)
|
||
st.plotly_chart(bmi_chart, use_container_width=True)
|
||
|
||
with col2:
|
||
st.markdown("### BMI统计")
|
||
st.metric("平均BMI", f"{filtered_df['BMI'].mean():.1f}")
|
||
st.metric("BMI中位数", f"{filtered_df['BMI'].median():.1f}")
|
||
|
||
# BMI分类分布
|
||
bmi_dist = filtered_df['bmi_category'].value_counts()
|
||
st.markdown("### BMI分类分布")
|
||
for category, count in bmi_dist.items():
|
||
percentage = (count / len(filtered_df)) * 100
|
||
st.markdown(f"**{category}**: {count:,} ({percentage:.1f}%)")
|
||
|
||
st.markdown("---")
|
||
|
||
# 4. 数据预览
|
||
st.header("🔍 数据预览")
|
||
|
||
with st.expander("查看筛选后的数据"):
|
||
st.dataframe(
|
||
filtered_df[
|
||
['id', 'age_years', 'gender_str', 'height', 'weight', 'BMI',
|
||
'bmi_category', 'ap_hi', 'ap_lo', 'cholesterol_str',
|
||
'gluc_str', 'smoke', 'alco', 'active', 'cardio_str']
|
||
].head(100),
|
||
use_container_width=True
|
||
)
|
||
|
||
# 5. 数据下载
|
||
st.header("📥 数据导出")
|
||
|
||
@st.cache_data
|
||
def convert_df_to_csv(df):
|
||
return df.to_csv(index=False).encode('utf-8')
|
||
|
||
csv_data = convert_df_to_csv(filtered_df)
|
||
|
||
col1, col2 = st.columns(2)
|
||
|
||
with col1:
|
||
st.download_button(
|
||
label="📥 下载筛选数据 (CSV)",
|
||
data=csv_data,
|
||
file_name="filtered_cardio_data.csv",
|
||
mime="text/csv",
|
||
help="下载当前筛选条件下的数据"
|
||
)
|
||
|
||
with col2:
|
||
if st.button("🔄 重置筛选"):
|
||
st.rerun()
|
||
|
||
# 页脚
|
||
st.markdown("---")
|
||
st.markdown(
|
||
"""
|
||
<div style='text-align: center; color: gray;'>
|
||
<p>心血管疾病数据分析仪表板 | 数据来源: 心血管疾病.xlsx | 总记录数: 70,000</p>
|
||
</div>
|
||
""",
|
||
unsafe_allow_html=True
|
||
)
|
||
|
||
if __name__ == "__main__":
|
||
main() |