Self-Explainable Interface for Disease Diagnostic Using Ensembler Machine Learning Model

Vanitha P, Aarthi R, Yasvanthika K, Purushothaman M, Shamini S

2025

Abstract

Diabetes and stroke are major chronic diseases that significantly affect global living standards. This study uses machine learning techniques to develop a system for early detection and treatment of individuals at risk. Health data, including age, blood pressure, glucose levels, and BMI, is used to predict diabetes. Models such as Gradient Boosting, K-Nearest Neighbors (KNN), Support Vector Classifier (SVC), Random Forest, XGBoost, AdaBoost, and LightGBM were employed, with an ensemble achieving 91% accuracy, establishing it as a strong predictor. For stroke prediction, models such as Extra Trees Classifier, Random Forest, XGB Classifier, and Gradient Boosting utilized factors like age, hypertension, heart disease, and glucose levels. The Random Forest Classifier achieved 99% accuracy, the highest among all models. A web interface built using Streamlit deploys these models, enabling real-time predictions by allowing users to input health attributes during examinations. This interface supports doctors and patients in identifying risks for both stroke and diabetes efficiently. This study demonstrates how machine learning can significantly aid in the early detection of chronic conditions, improving treatment outcomes and enabling timely interventions, ultimately fostering healthier lives.

Paper Citation

in Harvard Style

P V., R A., K Y., M P. and S S. (2025). Self-Explainable Interface for Disease Diagnostic Using Ensembler Machine Learning Model. In Proceedings of the 3rd International Conference on Futuristic Technology - Volume 2: INCOFT; ISBN 978-989-758-763-4, SciTePress, pages 538-544. DOI: 10.5220/0013596400004664

in Bibtex Style

@conference{incoft25,
author={Vanitha P and Aarthi R and Yasvanthika K and Purushothaman M and Shamini S},
title={Self-Explainable Interface for Disease Diagnostic Using Ensembler Machine Learning Model},
booktitle={Proceedings of the 3rd International Conference on Futuristic Technology - Volume 2: INCOFT},
year={2025},
pages={538-544},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013596400004664},
isbn={978-989-758-763-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 3rd International Conference on Futuristic Technology - Volume 2: INCOFT
TI - Self-Explainable Interface for Disease Diagnostic Using Ensembler Machine Learning Model
SN - 978-989-758-763-4
AU - P V.
AU - R A.
AU - K Y.
AU - M P.
AU - S S.
PY - 2025
SP - 538
EP - 544
DO - 10.5220/0013596400004664
PB - SciTePress