Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM68913
Volume:
10 |
11 |
Nov 2026
Pages:
1-6
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.

Department of Computer Science and Engineering
Don Bosco Institute of Technology

Department of Computer Science and Engineering
Don Bosco Institute of Technology

Department of Computer Science and Engineering
Don Bosco Institute of Technology

Department of Computer Science and Engineering
Don Bosco Institute of Technology
Department of Computer Science and Engineering
Don Bosco Institute of Technology
Judging cardiovascular risk means reading blood pres-sure, cholesterol, glucose, chest-pain type and exercise response together, and a bare score from a model rarely helps the person who has to act on it. This paper describes CardioMind AI, a prototype decision-support web application that wraps a super-vised risk model in a workflow for entering clinical attributes, inspecting the estimated probability and per-feature contribu-tions, saving a report, finding cardiology centres, and asking a retrieval-grounded assistant for background information. Four classifiers, namely Logistic Regression, Random Forest, XGBoost and an SVM, were compared on the UCI Cleveland benchmark using a stratified 80:20 split and sigmoid (Platt) calibration. Lo-gistic Regression gave the best accuracy (85.25%) and specificity (87.88%), while Random Forest gave the best ROC-AUC (0.9340), PR-AUC (0.9290) and Brier score (0.1199). The reported test par-tition appears to hold only 61 patients, so the differences between models are within sampling noise. The work is a prototype and a benchmark study. It is not clinically validated and must not be used for diagnosis.
KEYWORDS
This paper presented CardioMind AI, a prototype that places a calibrated heart-disease risk estimate inside a workflow of ex-planation, reporting, service discovery and grounded question answering. In the reported comparison, Logistic Regression led on accuracy (85.25%) and specificity (87.88%), and Random Forest led on ROC-AUC (0.9340), PR-AUC (0.9290) and Brier score (0.1199). Because the test partition most likely contains 61 patients, these gaps are small relative to their uncertainty, and the ranking is provisional. The system is a reasonable starting point for studying how model output can be presented responsibly. Cross-validated evaluation, independent data and clinical review are the main requirements before it can be considered for any real-world use.
[1] M. T. García-Ordás et al., “Heart disease risk prediction using deep learning techniques with feature augmentation,” Multime-dia Tools and Applications, vol. 82, pp. 31759–31773, 2023. [2] R. Detrano et al., “International application of a new probability algorithm for the diagnosis of coronary artery disease,” Am. J. Cardiology, vol. 64, no. 5, pp. 304–310, 1989. [3] A. Asuncion and D. J. Newman, “UCI Machine Learning Repos-itory,” Univ. of California, Irvine, 2007. [4] L. Breiman, “Random forests,” Machine Learning, vol. 45, no. 1, pp. 5–32, 2001. [5] C. Cortes and V. Vapnik, “Support-vector networks,” Machine Learning, vol. 20, no. 3, pp. 273–297, 1995. [6] T. Chen and C. Guestrin, “XGBoost: A scalable tree boosting system,” in Proc. 22nd ACM SIGKDD, 2016, pp. 785–794. [7] F. Pedregosa et al., “Scikit-learn: Machine learning in Python,”J. Mach. Learn. Res., vol. 12, pp. 2825–2830, 2011. [8] J. Platt, “Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods,” in Advances in Large Margin Classifiers, MIT Press, 1999, pp. 61–74. [9] A. Niculescu-Mizil and R. Caruana, “Predicting good probabili-ties with supervised learning,” in Proc. 22nd ICML, 2005, pp. 625–632. [10] G. W. Brier, “Verification of forecasts expressed in terms of probability,” Monthly Weather Review, vol. 78, no. 1, pp. 1–3, 1950.
Publication Date
11/12/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
2
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...