3. Introduction
Cardiovascular Disease (CVD) is a term that encompasses a range of
conditions affecting the heart and blood vessels. The CVD system of
body is responsible for the transportation of blood in the body, also
known as Circulatory System. The main elemnts of Cardiovascular
system are the heart, veins, artries, blood capillaries. If any of the
element is not able to perform its function properly, the
cardiovascular disease occurs. The failure of Cardiovascular System
can result in instant death. It is a leading cause of death globally,
accounting for millions of deaths each year.
5. CAD (Coronary Artery Disease): Narrowed or blocked coronary
arteries due to plaque buildup.
Hypertensive Heart Disease: Caused by high blood pressure.
Heart Failure: Inability to pump blood effectively.
Arrhythmias: Irregular heart rhythms.
Valvular Heart Diseases: Problems with heart valves.
Congenital Heart Diseases: Birth defects affecting heart structure.
PAD (Peripheral Artery Disease): Narrowed leg arteries.
Overview
6. Causes
Cardiovascular Disease is primarily caused by the buildup
of fatty deposits in the arteries, known as
atherosclerosis. Other causes include high blood
pressure, smoking, high cholesterol levels, and diabetes.
Symptoms
Risk Factors
Common symptoms of Cardiovascular Disease include
chest pain or discomfort, shortness of breath, fatigue,
dizziness, and irregular heartbeat. In some cases, there
may be no symptoms until a heart attack or stroke
occurs.
These include age, family history, high blood pressure,
high cholesterol levels, smoking, obesity, diabetes,
and a sedentary lifestyle.
Disease Details
7. Previous work done
Author: Akintunde O. Akinkuolie, Julie E. Buring, Paul M. Ridker, Samia Mora
Article Title: A Novel Protein Glycan Biomarker and Future Cardiovascular Disease Events (2014)
Problem: Baseline concentrations of GlycA, a protein glycan biomarker, were associated with an
increased risk of incident CVD events in women. The association of GlycA with CVD was
comparable to that of hsCRP, a biomarker of inflammation. Adjusting for hsCRP attenuated the
association of GlycA with CVD, suggesting a possible role for protein glycans in inflammation and
cardiovascular disease.
Methods: The study utilized statistical tests like Wilcoxon rank sum and χ2 tests. They correlated
GlycA with risk factors using Spearman coefficients and estimated risk with Cox regression.
Models were adjusted for factors including age, race, smoking, blood pressure, cholesterol, and
inflammatory markers.
Conclusion: Higher levels of certain types of molecules in the blood were linked to a greater risk
of cardiovascular disease (CVD). However, when they considered another marker related to
inflammation, called hsCRP, this link became less strong. The study suggests potential clinical
applications of glycobiology in CVD prevention.
8. Previous work done
Author: Emmi Tikkanen, PhD; Stefan Gustafsson, PhD; Erik Ingelsson, MD, PhD
Article Title: Associations of Fitness, Physical Activity, Strength, and Genetic Risk With
Cardiovascular Disease (2018)
Problem: The association between fitness, physical activity, genetic risk, and cardiovascular
disease (CVD) outcomes.
Methods: Analyzing objective and subjective measures of fitness and physical activity, along with
genetic risk scores for coronary heart disease and atrial fibrillation, in relation to prospective CVD
events and all-cause death in a large cohort of individuals from the UK Biobank. Cox proportional
hazards models were used for analysis.
Conclusion: Being fit and active is linked to lower risks of heart problems. Even with a genetic
predisposition to heart disease, being active still helps reduce the risk. Personalized exercise plans
might be especially beneficial for people with a higher genetic risk of heart disease.
9. Previous work done
Author: Yash Jayesh Chauhan
Article Name: Cardiovascular Disease Prediction using Classification Algorithms of Machine
Learning(2020)
Problem: The research problem addressed is the prediction of heart diseases with a high rate of
accuracy using machine learning algorithms.
Methods: The proposed solution is to compare the accuracy of different machine learning
classification algorithms to evaluate the 10-year risk of future heart diseases. Data preprocessing,
feature extraction, and classification techniques such as Logistic Regression, Random Forest, K-
Nearest Neighbors (KNN), Support Vector Machine (SVM), and Decision Tree are employed.
Conclusion: Machine learning methods like Logistic Regression, Random Forest, K-Nearest
Neighbors (KNN), Support Vector Machine (SVM), and Decision Tree are effective for predicting
cardiovascular diseases and assessing risks. Logistic regression stands out for its ability to
calculate the probability of disease based on patient attributes. The study emphasizes the
importance of using machine learning in healthcare to detect and treat cardiovascular diseases
early, leading to better patient outcomes and healthcare management.
10. Previous work done
Author: Kevin J. Clerkin, Justin A. Fried, Jayant Raikhelkar, Gabriel Sayer, Jan M. Griffin, Amirali
Masoumi, Sneha S. Jain, Daniel Burkhoff, Deepa Kumaraiah, LeRoy Rabbani, Allan Schwartz, Nir
Uriel
Article Name: COVID-19 and Cardiovascular Disease(2020)
Problem: The research investigates how COVID-19 affects individuals with pre-existing heart
conditions. It aims to comprehend the exacerbating effects of COVID-19 on heart problems and
proposes strategies for treatment and care for patients with both COVID-19 and cardiovascular
diseases.
Methods: The study involves analyzing medical records, previous studies, and COVID-19 data to
understand the relationship between COVID-19 and cardiovascular diseases. Various techniques
such as analyzing patient traits and blood markers, comparing findings with past outbreaks, meta-
analysis of multiple studies, and experimentation with drugs were employed.
Conclusion: The study highlights that COVID-19 patients with pre-existing heart conditions face a
higher risk of severe outcomes. Understanding the connections between COVID-19 and heart
problems is crucial for improving patient care.
11. Previous work done
Author: Muhammad Zakir Hossin, Lorena Fernandez de la Cruz, Kyla A McKay, Tim F Oberlander,
Anna Sandström, and Neda Razaz
Article Title: Association of pre-existing maternal cardiovascular diseases with
neurodevelopmental disorders in offspring: a cohort study in Sweden and British Columbia,
Canada (2024)
Problem: Investigating the association between maternal cardiovascular disease (CVD) and
neurodevelopmental disorders (ADHD, ASD, intellectual disability) in offspring.
Methods: Cox regression, mediation analysis, Poisson model, counterfactual mediation analysis.
Conclusion: Children of mothers with cardiovascular disease have a slightly higher risk of ADHD
and ASD compared to those without, with varying degrees of association based on different types
of maternal CVD. Paternal CVD generally doesn't show such associations except for
cerebrovascular disease. Factors occurring during pregnancy might play a more significant role
than genetics in connecting heart problems and neurodevelopmental disorders in children. Future
research should focus on understanding the mechanisms behind this association, exploring long-
term outcomes, and finding ways to prevent neurodevelopmental disorders in children of
mothers with CVD.
12. Source
The dataset used in this research was
obtained from Kaggle.
Size
Variables
The dataset consists of records from
over 70,000 individuals.
The dataset includes a wide range of
variables related to cardiovascular
health, including age, gender, height,
weight, cholesterol levels, and
lifestyle factors such as smoking and
physical activity.
Dataset Information
13. Data Cleaning: Dropping
unnecessary columns
('Unnamed: 0' and 'id') from the
DataFrame.
Feature Engineering: Calculating
BMI (Body Mass Index) from
weight and height.
Handling Outliers: Filtering
extreme BMI values.
Preprocessing Steps
14. Logistic Regression Accuracy: 0.9811
Decision Tree Accuracy: 0.9762
Random Forest Accuracy: 0.9815
Support Vector Machine Accuracy: 0.9813
K-Nearest Neighbors Accuracy: 0.9786
Results of the Code