Early Cardiovascular Disease Detection Using Predictive Machine-Learning Models - Evaluation and Insights

Aug 3, 2026

Cardiovascular diseases remain the leading cause of mortality, making early diagnosis essential for reducing risk and improving outcomes. Machine learning enables data-driven vulnerability assessment through pattern recognition, supporting risk evaluation, early problem detection, and treatment plan personalization. This study builds a predictive machine-learning paradigm for nascent detection of heart disease by training and comparing eight classifiers on the UCI dataset, focusing on robustness, complexity, and performance.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages18

Reading Time45 min

AccessFree to download

Introduction

Cardiovascular disease burden and risk factors

Motivation for machine learning in early detection

Methods

Classifiers evaluated

Dataset and evaluation protocol

Results and Insights

Performance metrics and model comparison

Conclusion

Best-performing models for early detection

Early Cardiovascular Disease Detection Using Predictive Machine-Learning Models: Evaluation and Insights | Received 19 March 2025; Revised 1 September 2025; Accepted 1 September 2025 | | | --- | --- | | Nahed Tawfik1 Nourhan Zayed 2* | Abstract:As the foremost global cause of mortality, cardiovascular diseases (CVDs) underscore the critical importance of early diagnosis. Machine learning (ML) has emerged as a transformative tool in healthcare, enabling precise vulnerability assessment through data-driven pattern recognition. Implementing this technology in cardiology is critical for assessing risks, detecting nascent problems, and customizing treatment plans. This study intended to cultivate a | | Keywords
Cardiovascular disease; machine learning; heart illnesses; prediction model | predictive machine-learning paradigm for the nascent detection of heart disease. Eight classifiers, namely, k-nearest-neighbors, support-vector machine, logistic regression, random forest, decision tree, artificial neural networks, gradient boosting, and CN2 rule induction, were utilized to calibrate cardiovascular disease diagnosis predictions in the field of ML. One of the contributions of the proposed method is its enhanced nascent detection of cardiovascular disease, matched up with existing models using the same dataset in terms of both robustness and complexity. This study appraised assorted classifiers and their efficacies, offering helpful information for the development of trustworthy prediction models for coronary diseases. These models were scrutinized for heart illness using the UCI dataset and achieved improved accuracy and performance metrics. From the perspective of performance indicators (AUC, accuracy, F1-score, precision, & recall), the conclusions ofthis study demonstrate the efficacy of the CN2 rule induction and random forest models for detecting cardiovascular disorders. | 1. Introduction The heart portrays a decisive contribution in nurturing life by efficiently channeling oxygen-rich blood and modulating imperative hormones to keep blood pressure within a healthy range. Variations from the typical functioning of the heart can lead to heart-related ailments, ordinally termed to as cardiovascular diseases (CVDs) [1] . CVDs encompass assorted ailments that impact the heart and blood vessels. These ailments include coronary artery disease, peripheral arterial problems, rheumatic heart illnesses, congenital malformations, pulmonary embolisms, arrhythmias, and cardiomyopathies affecting the heart muscle [2] . Coronary heart disease, an important subtype of CVD, is responsible for 64% of substantial cases of cardiovascular illnesses [3]. Although it mainly affects males, females 1 Researcher, Computers and Systems Department, Electronics Research Institute (ERI), Cairo, [Egypt.](Egypt. nahedtawfik@eri.sci.eg)[ ](Egypt. nahedtawfik@eri.sci.eg)[nahedtawfik@eri.sci.eg](Egypt. nahedtawfik@eri.sci.eg)[ ](Egypt. nahedtawfik@eri.sci.eg)[2](2)* Assoc. Professor, Computers and Systems Department, Electronics Research Institute (ERI)/ Mechatronics Engineering, British University in Egypt, Cairo, [Egypt.](Egypt. nourhan@eri.sci.eg)[ ](Egypt. nourhan@eri.sci.eg)[nourhan@eri.sci.eg](Egypt. nourhan@eri.sci.eg) , [nourhan.zayed@bue.edu.eg](nourhan.zayed@bue.edu.eg) are equally susceptible. Among all cardiovascular disorders, coronary artery disease is of particular concern because of its strong correlation with the global death toll. In line with WHO (the World Health Organization), cardiovascular illnesses have severe consequences and cause approximately eighteen million deaths worldwide annually. These alarming statistics emphasize the need for scientific investigations and advancements in medicine directed at treating and mitigating the impact of CVDs on a global scale [4] . Multiple vulnerability factors, such as hypertension, obesity, abnormal lipid profiles, diabetes, tobacco use, physical inactivity, alcohol intake, and c

FAQs

Why is early detection of cardiovascular disease important?

Cardiovascular diseases cause substantial global mortality, and early diagnosis helps enable timely preventive interventions and improve treatment planning.

Which machine-learning classifiers were evaluated in the study?

The study used eight classifiers: k-nearest-neighbors, support-vector machine, logistic regression, random forest, decision tree, artificial neural networks, gradient boosting, and CN2 rule induction.

How were the models assessed and which metrics were used?

Models were evaluated using the UCI dataset with performance indicators including AUC, accuracy, F1-score, precision, and recall.

Aug 3, 2026

Upload by Theodora