An explainable hybrid framework for early detection of cardiovascular diseases using Categorical Boosting and Bees algorithm

Sep 28, 2026

Uploaded by Ivy

An explainable hybrid machine learning framework is developed for early detection of cardiovascular diseases (CVD) using Categorical Boosting (CatBoost) combined with the Bees optimization algorithm. To address data imbalance, Random Oversampling is applied, followed by Pearson correlation analysis and Min–Max scaling for normalization. The model achieves 98.04% accuracy with strong precision, recall, F1-score, and specificity. Interpretability is provided through LIME and SHAP to identify key attributes driving CVD occurrence, supporting targeted, timely clinical decisions.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages19

Reading Time48 min

AccessFree to download

Introduction

Methodology

Data preprocessing

Hybrid prediction model

Explainable AI

Results and evaluation

Discussion and implications

[www. nature.com/scientificreports](www. nature.com/scientificreports) OPEN An explainable hybrid framework for early detection of cardiovascular diseases using Categorical Boosting and Bees algorithm Jayanta Sen􀀍 & Sweta Bhattacharya􀀍 Cardiovascular disease (CVD) remains one of the leading causes of death worldwide, claiming millions of lives each year. The early detection ofCVD enables healthcare professionals to make informed decisions about the patient’s health. Machine learning (ML) -based frameworks have been extremely popular in predicting diseases. However, results generated from traditional ML models are “blackbox,” lacking transparency and interpretability. The objective of the present study is to develop an ML framework that detects CVD with promising accuracy and, further, provide interpretability to the generated outcomes to ensure targeted therapies. The Framingham, Massachusetts CVD dataset, which is publicly available from the Kaggle Repository, is used in this study. As part of the data preprocessing, the Random Oversampling (RO) technique is applied to overcome the data imbalance problem, followed by Pearson Correlation analysis to understand the correlation between attributes. Then, the Min–Max scaling technique is used for data normalization. The pre-processed data is fed into a hybrid ML framework incorporating the Categorical Boosting (CatBoost) and BEEs algorithms to achieve optimized CVD prediction results. The proposed Hybrid model yielded 98.04% accuracy, a Precision of 97.09%, a Recall of 98.96%, an F1-score of 98.02%, and a Specificity of 97.16%, with a total execution time of 26.6580 s. The proposed model outperformed contemporary state-of-theart algorithms, considering most evaluation metrics. Additionally, Explainable Artificial Intelligence (XAI) techniques, such as LIME and SHAP, are implemented to identify the contribution of the most significant attributes towards the occurrence ofCVD, offering valuable insights into the detection of the disease and enabling healthcare providers to make accurate and timely treatment decisions. Keywords Cardiovascular disease, Machine learning, CatBoost, BEES, Explainable AI Cardiovascular disease (CVD) is a condition caused by blood channel obstruction that causes chest pain and heart attacks, leading to major heart-related diseases like heart failure, which can also have fatal consequences1. With 15 million deaths recorded in 2015, CVD has continuously been the leading cause of mortality for the previous 15 years2. A study from January 2017 confirmed that CVD is the primary cause of global death. By 2020, the World Health Organization estimated that CVD would account for approximately 17.9 million deaths annually worldwide3. The death toll from coronary heart disease has been steadily rising, with projections indicating that global deaths from CVD could exceed 23.6 million by 2030, solidifying its position as the most significant cause of mortality worldwide. Cerebrovascular disease, peripheral arterial disease, coronary heart disease, transient ischemic attacks, pulmonary embolism, vascular illnesses, congenital heart defects, chronic heart conditions, deep vein thrombosis, and strokes are some of the prevalent cardiovascular diseases4. Diagnosing cardiac disease is further complicated by various risk variables, including irregular pulse rates, high blood pressure, abnormal cholesterol levels, and hypertension. Thus, early detection of cardiovascular diseases (CVD) is crucial for mitigating such an impact. Machine learning (ML), a rapidly advancing field within Artificial Intelligence (AI), has proven valuable in analyzing large datasets from various domains, including healthcare5. Hybrid deep learning models that utilise adaptive dropout and optimisation techniques have demonstrated improved convergence School of Computer Science Engineering and Information Systems (SCORE), Vellore Institute of Technology, Vellore, Tamil Nadu 632014, India. 􀀍

FAQs

Why is early detection of cardiovascular diseases important in this study?

Early detection enables healthcare professionals to make informed decisions and supports targeted therapies, reducing the impact of CVD. The study emphasizes improved outcomes through timely and accurate identification.

How is data imbalance handled before model training?

The study applies Random Oversampling (RO) to overcome data imbalance. It then uses Pearson Correlation analysis to understand feature relationships and Min–Max scaling for normalization.

What makes the proposed framework explainable?

Explainable AI techniques, specifically LIME and SHAP, are implemented. They identify which significant attributes contribute most to the occurrence of CVD, improving interpretability beyond black-box predictions.

Sep 28, 2026

Uploaded by Ivy