Advancing a FAIR and explainable AI-based predictive model for cardiovascular diseases in hospital settings
The healthcare field is witnessing an increasing volume of data available for analysis and is undergoing pragmatic transformations due to the advent of artificial intelligence. In recent years, machine learning and its sub-field, deep learning have emerged as effective tools to help physicians with diagnosis, predictions and improved treatment plans. The paper describes a cohort of patients which were monitored during their regular visits in the cardiology department of the County Emergency Clinical Hospital of Timis in Romania on their physiological parameters and test results. After employing a wide range of conventional machine learning and advanced deep learning techniques, the experimental results revealed that the feedforward neural network performed best in predicting cardiovascular outcomes. The paper also analyzes the aspects of FAIR AI, which translates to the usage of artificial intelligence in a ethical and unbiased way in order to improve the early and precise diagnosis for cardiac patients. The work presented in this paper represents an overview of the current situation in CVDs management and treatment options, as well as offering a viable and interoperable AI-based solution that can be integrated in cardiology hospital departments in order to improve the diagnosis and experience of cardiac patients.
Authors
- Adina Niţulescu (ORCID: https://orcid.org/0009-0004-3895-4536)
Institutions
- Polytechnic University of Timişoara (RO)
Publication Details
- Journal
- Frontiers in Medicine
- Published
- 2026-11-18
- DOI
- https://doi.org/10.3389/fmed.2026.1867621
- Primary Topic
- Artificial Intelligence in Healthcare
- Type
- article
- Field-Weighted Citation Impact
- 0.00