README.md

metadata

license: apache-2.0
task_categories:
  - feature-extraction
language:
  - en
tags:
  - biology
pretty_name: test
size_categories:
  - 100K<n<1M
dataset_info:
  config_name: mydata
  features:
    - name: Year
      dtype: int32
    - name: LocationAbbr
      dtype: string
    - name: LocationDesc
      dtype: string
    - name: Latitude
      dtype: float32
    - name: Longitude
      dtype: float32
    - name: Disease_Type
      dtype: int32
    - name: Data_Value_Type
      dtype: int32
    - name: Data_Value
      dtype: float32
    - name: Break_Out_Category
      dtype: string
    - name: Break_Out_Details
      dtype: string
    - name: Break_Out_Type
      dtype: int32
    - name: Life_Expectancy
      dtype: float32

Dataset Description

This dataset encompasses mortality rates for cardiovascular (CVD) and heart diseases across the United States, covering both state-specific and national levels, from 2000 to 2020. The mortality rate is quantified as the number of deaths per 100,000 individuals annually in the US. The dataset is structured to classify mortality rates according to various demographic factors, including overall rates, gender (female, male), race (white, black, Hispanic, other), and age groups (18-24, 25-44, 45-65, 65+). Additionally, life expectancy data for each state is incorporated in the dataset. For ease of use, I combined the data on a five-year interval rather than an annual basis.

Dataset Sources

• CVD Mortality Data: Centers for Disease Control and Prevention(CDC) National Vital Statistics System
  • https://data.cdc.gov/Heart-Disease-Stroke-Prevention/National-Vital-Statistics-System-NVSS-National-Car/kztq-p2jf/about_data
• Life Expectancy Data: Institute for Health Metrics and Evaluation
  • https://ghdx.healthdata.org/record/ihme-data/united-states-life-expectancy-by-county-race-ethnicity-2000-2019

Uses

This dataset serves as a valuable resource for researchers and individuals interested in examining and identifying patterns related to cardiovascular diseases in the United States. It can be utilized to forecast future fatalities caused by heart diseases by leveraging similar features present in the dataset. Additionally, the dataset enables users to gain insights into identifying states that require assistance and support in reducing mortality rates. Below are example use cases and corresponding codes:

• Analyzing the comprehensive picture of mortality and conducting time series analysis on mortality rates
  • https://colab.research.google.com/drive/1ulygrSt9jt3x_4WIGD6QdK0TcGZlpuYF
• Building regression models
  • https://colab.research.google.com/drive/1DhIni026qz5qqjfWwKXnqoQXDy-HzroC
• Developing a web application for users to quickly understand and compare mortality rates among states, along with relevant information like state population
  • https://github.com/jiwonny29/Exploring_US_Cardiovascular_Mortality_Trends_via_Streamlit

Dataset Structure

This dataset contains

• Year (int32): This column contains the year of the data record, with values ranging from 2000 to 2020
• LocationAbbr (String): Abbreviation representing the location, typically a state
• LocationDesc (String): The full name or detailed description of the location
• Latitude (float32) : Geographic coordinate that specifies the north-south position of a point on the Earth's surface
• Longitude (float32) : Geographic coordinate that specifies the east-west position of a point on the Earth's surface
• Geolocation (Tuple): A pair of latitude and longitude coordinates, formatted as (latitude, longitude), providing the geolocation or geocode of the location
• Disease_Type (int32): A key column in the dataset, representing eight unique types of cardiovascular diseases, numbered from 0 to 7. The values correspond to the following diseases:
  • 0: Major Cardiovascular Disease
  • 1: Diseases of the Heart (Heart Disease)
  • 2: Acute Myocardial Infarction (Heart Attack)
  • 3: Coronary Heart Disease
  • 4: Heart Failure
  • 5: Cerebrovascular Disease (Stroke)
  • 6: Ischemic Stroke
  • 7: Hemorrhagic Stroke
• Data_Value_Type (int32): Represents the type of data value. "Age-Standardized" is represented by 1, and "Crude" is represented by 0, indicating the measurement methods for the data value columns
• Data_Value (float32): This column represents the number of deaths per 100,000 population
• Break_Out_Category (string): This category is used for breaking down the data and includes four unique values: "Overall," "Gender," "Age," and "Race."
• Break_Out_Details (string): Specific subcategories within the Break_Out_Category. This column includes values like "Overall," six age categories (e.g., "18-24," "25-44"), two gender categories (e.g., "Female," "Male"), and four race categories (e.g., "Hispanic," "Non-Hispanic Black," "Non-Hispanic White," "Other").
• Break_Out_Type (int32): A numerical transformation of the Break_Out_Details column. In this system, "Overall" is represented as 0, "Male" and "Female" as 1 and 2, respectively; age groups "18-24," "25-44," "45-64," "65+" as 1, 2, 3, 4, respectively; and racial categories "Non-Hispanic White," "Non-Hispanic Black," "Hispanic," "Other" as 1, 2, 3, 4, respectively.
• Life_Expectancy (float32): Represents the life expectancy for the applicable year and state