SCHOOL OF COMPUTER SCIENCE ENGINEERING &

INFORMATION SYSTEMS (SCORE)

Winter Semester 2024-25

ARTIFICIAL INTELLIGENCE
SWE4010

Review-1

TITLE: HEART DISEASE PREDICTION SYSTEM

Name :N. Sakthivel

RegNo: 22MIS0576
ABSTRACT:
Heart disease is a disorder that affects the cardiovascular system and remains one of the
leading causes of mortality worldwide. Early detection and risk assessment are crucial for
preventing severe cardiac events. Machine learning techniques, specifically Support Vector
Machine (SVM) classification, can analyze various clinical parameters to predict heart disease
risk. The system utilizes multiple factors including age, blood pressure, cholesterol levels, and
ECG results. Through analysis of these parameters, along with chest pain characteristics and
exercise-induced changes, the system can effectively predict the likelihood of heart disease
presence.

A heart disease prediction system refers to a technology designed to assess an individual's

risk of having heart disease based on various clinical and demographic parameters. These
systems leverage data from multiple sources, including patient demographics, clinical
measurements, and exercise test results. The primary types of parameters considered are blood
pressure, cholesterol levels, and ECG readings.

Symptoms of heart disease can vary depending on the severity and type. Common indicators
include different types of chest pain (typical angina, atypical angina, non-anginal pain), exercise
induced angina, and ECG abnormalities. Early detection through these parameters is crucial for
effective treatment and management

Machine learning models, particularly Support Vector Machines (SVM), have demonstrated high
accuracy in predicting heart disease by analyzing large datasets and identifying complex patterns
that may not be immediately apparent to human experts. These models can be trained on
historical patient data to recognize correlations between various clinical factors and the
likelihood of developing heart disease. Additionally, modern heart disease prediction systems
integrate electronic health records (EHRs) and real-time monitoring devices, such as wearable
fitness trackers and smartwatches, to continuously assess cardiovascular health. By utilizing
these technologies, patients at risk can be identified early, allowing healthcare professionals to
take preventive measures such as recommending lifestyle modifications, prescribing medications,
or conducting further diagnostic tests. Moreover, the incorporation of artificial intelligence (AI)
and deep learning enhances the ability of prediction models to refine their accuracy over time,
making heart disease screening more efficient and accessible. The development of such systems
plays a crucial role in reducing the global burden of cardiovascular diseases by enabling timely
interventions and improving patient outcomes.

TOOL/Programming Languages:
 Google Colab (Cloud-based Jupyter Notebook for coding & model training)
 Python
Libraries Used:
 Scikit-learn (Machine Learning models)
 Pandas & NumPy (Data Processing)
 Matplotlib & Seaborn (Data Visualization)

Some Reference Articles from journals:

1) Yadav(3) (2023) focus on leveraging machine learning techniques for heart disease prediction,
utilizing algorithms such as Random Forest, Decision Trees, and Gradient Boosting to analyze a
large dataset. The study highlights the effectiveness of these models in achieving high accuracy
for early detection and diagnosis of cardiovascular conditions. The results emphasize the critical
role of feature optimization and parameter tuning to enhance prediction reliability, offering a
significant contribution to aiding medical professionals in clinical decision-making

2) Duraisamy(1) et al (2023) have discussed Heart disease prediction through online

consultation using machine learning refers to the application of advanced algorithms and
techniques to analyze medical data collected during online consultations to predict the likelihood
of an individual developing heart disease. Machine learning models are trained using historical
data that includes various risk factors such as age, gender, blood pressure, cholesterol levels, and
medical history. These models then utilize the input provided by patients during online
consultations, such as symptoms, lifestyle habits, and additional medical tests, to generate
personalized predictions about the probability of heart disease occurrence. By leveraging the
power of machine learning, this approach aims to assist healthcare professionals in making more
accurate diagnoses and providing timely recommendations for preventive measures or further
medical intervention, ultimately improving patient outcomes and reducing the burden on
healthcare systems. In this paper, a machine learning technique called Support Vector Machine
(SVM) is used for heart disease prediction.

3) Arifuddin and Buana(2023) have discussed Heart failure is a significant global cause of
mortality and morbidity. This research article aims to evaluate the performance of decision tree
(DTree) and support vector machine (SVM) methods in predicting heart disease. The study
utilizes a dataset with diverse features and employs exploratory data analysis (EDA), clustering,
and classification techniques to gain insights and evaluate the performance of the two methods.
The results demonstrate the effectiveness of both DTree and SVM in predicting heart disease.
Notably, SVM outperforms DTree in terms of accuracy, precision, recall, and F1-score. However,
the performance of these methods is influenced by the preprocessing steps applied, indicating the
importance of selecting appropriate data preprocessing techniques for optimal performance with
specific machine learning algorithms.

4) Mondal(4) et al (2023) have discussed Heart-related illnesses, often known as cardiovascular

diseases, have been the leading cause of mortality globally over the past several decades and are
now recognized as the most major illness in both India and the rest of the globe. The severity out
of the disease can be avoided with proper care at proper stage. This disease claims early and
accurate prediction to avoid causalities. As proper medical support is not adequate, diseases are
not being identified at the proper time and treatment cannot be started. Machine learning
algorithms have shown promise in predicting heart disease risk based on patient data. In this
study, a machine learning-based heart disease prediction model has been presented. The
objective of the work is to build a machine learning-based model for early and adequate
prediction of heart disease. The proposed model has utilized support vector machine and
artificial intelligence with an accuracy of 81.6% and 86.6%, respectively.

5)AbdulHussein and Bilgin(5) (2024) have discussed Machine learning, one of the most well
known applications of artificial intelligence, is altering the world of research. The aim of this
study is to generate predictions for Heart Disease Prediction (HDP) by employing effective
machine learning approaches and to predict whether an individual has heart disease. The primary
objective is to evaluate the predictive accuracy of various machine learning algorithms in
predicting the presence or absence of heart disease. The KNIME data analysis program has been
selected, and overall accuracy is chosen as the primary indicator to assess the effectiveness of
these strategies. Utilizing details such as chest pain, cholesterol levels, age, and other factors,
along with different machine learning technologies such as K Nearest Neighbor (KNN), Naive
Bayes, and Logistic Regression, a dataset of 319,796 patient records with 18 attributes was
utilized. Naive Bayes, K Nearest Neighbor (KNN), and Logistic Regression were employed as
machine learning techniques, and their prediction accuracies were compared. The application
results indicate that the logistic regression approach outperforms the K Nearest Neighbor method
and the Naive Bayes method in terms of predicting accuracy for heart disease.

References:
[1] Balakrishnan Duraisamy, Ramesh Sunku, Karthikeyan Selvaraj, Venkata Vara Ramana Pilla,
and Manikandan Sanikala (2023). Heart Disease Prediction Using Support Vector Machine.
Multidisciplinary Science Journal, 6(1):504-600.

[2] Akhdhan Arifuddin and Gandhi Surya Buana (2023). Performance Comparison of Decision
Tree and Support Vector Machine Algorithms for Heart Failure Prediction. Procedia Comput.
Sci., 234(C), 10(1):628–636.

[3] Ajay Kumar Yadav, Krishna Soni, and Shraddha Khare(2023). Heart Disease Prediction
Using Machine Learning. IEEE Transactions on Biomedical Engineering, 14(2):150-157

[4] Aishwarya Mondal, Biswajit Mondal, Anirban Chakraborty, Animesh Kar, Abhishek Biswas,
and Arindam Bhattacharya Majumder (2023). Heart Disease Prediction Using Support Vector
Machine and Artificial Neural Network. Artificial Intelligence and Applications, 2(1):59–65.
[5] Ayat Bahaa AbdulHussein and Tulin Tezcan Bilgin (2024). Comparison of Machine
Learning Algorithms for Heart Disease Prediction. Journal of Technology and Applied Sciences,
7(1):133–146.

Time Line:

Week Task Details

Feb 19
Project Proposal & Finalize the project idea, research relevant papers,
- Feb
Research and understand the dataset.
25
Feb 26 Data Collection & Load the UCI Heart Disease dataset, clean missing
- Mar 3 Preprocessing values, and perform exploratory data analysis (EDA).
Feature
Mar 4 - Identify important features, normalize/standardize
Engineering &
Mar 10 data, and prepare for model training.
Selection
Mar 11
Model Selection & Train different machine learning models (Logistic
- Mar
Implementation Regression, Random Forest, SVM, etc.).
17
Mar 25
Model Evaluation & Test accuracy, precision, recall, and F1-score;
- Mar
Comparison compare different models.
31
Apr 8 - Result Analysis & Create graphs, confusion matrices, and performance
Apr 14 Visualization visualizations.
Apr 15
Report Writing & Prepare the final report with methodology, results,
- Apr
Documentation and findings.
21

Significance :
 Helps in early detection of heart disease using AI.
 Can be used by healthcare professionals to analyze patient risk.
 Reduces human error in diagnosis.
 Provides a low-cost, automated diagnostic tool for clinics and hospitals.