CARDIOVASCULAR DISEASE
PREDICTION
USING AI
Introduction
Cardiovascular disease remains a leading cause of mortality worldwide, highlighting
the need for accurate and early detection methods.
This project addresses this need by employing a hybrid approach that combines
machine learning and deep learning techniques to predict cardiovascular disease.
The first component utilises machine learning algorithms Naive Bayes, K-Nearest
Neighbours (KNN), and Logistic Regression to analyse structured patient data, such
as weight, blood pressure, cholesterol levels, glucose levels, and lifestyle factors.
The second component leverages deep learning through a Convolutional Neural
Network (CNN) to analyse ECG images, providing an additional layer of predictive
accuracy.
Existing system
Existing systems for cardiovascular disease prediction often rely on either traditional
statistical methods or basic machine learning models, primarily focusing on structured
data
like patient demographics and health metrics.
These systems typically use methods such as logistic regression or decision trees to
assess risk based on inputs like blood pressure, cholesterol levels, and lifestyle factors.
While effective, these approaches may lack the ability to analyze complex,
unstructured data such as ECG images, which can provide critical insights into heart
conditions.
Additionally, many systems do not integrate multiple types of data, potentially
limiting their accuracy and comprehensiveness.
Disadvantage
s
Limited Data Integration:
Many existing systems only analyse structured data from
health metrics and demographics, missing out on valuable insights from unstructured
data like ECG images or other advanced diagnostic tests, which can lead to less
comprehensive
risk assessments.
Reduced Accuracy:
Traditional models often rely on simpler algorithms that may not
capture complex relationships or patterns in the data, potentially reducing the overall
predictive accuracy and effectiveness of the system in identifying at-risk individuals.
Proposed system
The proposed system aims to enhance cardiovascular disease prediction by integrating
both machine learning and deep learning techniques, offering a more comprehensive
and
accurate diagnostic tool.
It features two distinct components: the first uses machine learning algorithms Naive
Bayes, K-Nearest Neighbours (KNN), and Logistic Regression to analyse structured
patient
data, including weight, blood pressure, cholesterol levels, glucose levels, and lifestyle
factors
such as smoking and exercise.
The second component employs a Convolutional Neural Network (CNN) to analyze
ECG images, providing an additional layer of insight into cardiac conditions.
Advantage
s
Enhanced Predictive Accuracy:
By integrating both machine learning and deep
learning methods, the system leverages the strengths of each approach machine
learning for
structured data and deep learning for image analysis resulting in a more accurate and
reliable
prediction of cardiovascular disease.
Improved Early Detection and Personalisation:
The dual-component approach
allows for early and more precise detection of cardiovascular conditions, while also
accommodating individual variations in health data and lifestyle factors, leading to more
personalized and effective treatment strategie
Thank you!
