Scribd
Upload
6 views

Dataset Documentation

The 'heart.csv' dataset contains medical attributes for predicting and detecting cardiovascular diseases, essential for developing AI models for mortality prediction. It includes various numerical and categorical variables such as age, sex, chest pain type, and cholesterol levels. Researchers are advised to consider ethical implications and limitations when using the dataset for model development.

Uploaded by

manasbhunyabtech
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
6 views

Dataset Documentation

The 'heart.csv' dataset contains medical attributes for predicting and detecting cardiovascular diseases, essential for developing AI models for mortality prediction. It includes various numerical and categorical variables such as age, sex, chest pain type, and cholesterol levels. Researchers are advised to consider ethical implications and limitations when using the dataset for model development.

Uploaded by

manasbhunyabtech
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Dataset Documenta,on: heart.

csv

Title: AI Models for Early Detec2on and Mortality Predic2on


in Cardiovascular Diseases
1. Overview:
The `heart.csv` dataset contains medical a3ributes of individuals, providing essen:al
informa:on used for the predic:on and early detec:on of cardiovascular diseases. The
dataset is pivotal for developing AI models aimed at mortality predic:on and early detec:on
in cardiovascular diseases.

2. Variables Descrip4on:

- age:
- type: Numerical
- Descrip:on: Age of the individual.
- Range: [29, 77]

- sex:
- Type: Categorical (Binary)
- Descrip:on: Gender of the individual.
- Categories:
- `0`: Female
- `1`: Male

- cp:
- Type: Categorical
- Descrip:on: Chest pain type experienced by the individual.
- Categories:[0, 1, 2, 3]

- trestbps:
- Type: Numerical
- Descrip:on: Res:ng blood pressure (in mm Hg) upon admission to the hospital.

- chol:
- Type: Numerical
- Descrip:on: Serum cholesterol level in mg/dl.

- Vs:
- Type: Categorical (Binary)
- Descrip:on: Fas:ng blood sugar.
- Categories:
- `0`: < 120 mg/dl
- `1`: > 120 mg/dl
- restecg:
- Type: Categorical
- Descrip:on: Res:ng electrocardiographic results.
- Categories: [0, 1, 2]

- thalach:
- Type: Numerical
- Descrip:on: Maximum heart rate achieved during the Thallium stress test.

- exang:
- Type: Categorical (Binary)
- Descrip:on: Exercise-induced angina.
- Categories:
- `0`: No
- `1`: Yes

- oldpeak:
- Type: Numerical
- Descrip:on: ST depression induced by exercise rela:ve to rest.

- slope:
- Type: Categorical
- Descrip:on: Slope of the peak exercise ST segment.
- Categories: [0, 1, 2]

- ca:
- Type: Numerical
- Descrip:on: Number of major vessels colored by fluoroscopy.

- thal:
- Type: Categorical
- Descrip:on: Thalassemia type.
- Categories: [1, 2, 3]

- target:
- Type: Categorical (Binary)
- Descrip:on: Diagnosis of heart disease.
- Categories:
- `0`: Absence of heart disease
- `1`: Presence of heart disease

3. Usage:
The dataset is intended for researchers and prac::oners in the healthcare and medical
domain for developing and valida:ng AI and machine learning models aimed at predic:ng
cardiovascular diseases. It serves as a resource for the explora:on of feature importance,
model explainability, and the development of interpretable models in the cardiovascular
health domain.

4. Limita4ons and Ethical Considera4ons:


Researchers should approach the dataset considering the ethical aspects of pa:ent data,
even if anonymized. Understanding the limita:ons in terms of representa:veness, possible
biases in data collec:on, and the need for external valida:on is crucial.

5. Acknowledgments:
Researchers u:lizing this dataset should acknowledge the source, and if applicable, the
funding agencies and ins:tu:ons suppor:ng the work related to the dataset's crea:on and
distribu:on.

You might also like