1Meenakshi College of Allied Health Sciences, Meenakshi Academy of Higher Education and Research
2Department of Microbiology, Meenakshi Medical College and Hospital Research Institute, Meenakshi Academy of Higher Education and Research
3Department of Biochemistry, Meenakshi Ammal Dental College and Hospital, Meenakshi Academy of Higher Education and Research
4Meenakshi College of Arts & Science, Meenakshi Academy of Higher Education and Research
5Arulmigu Meenakshi College of Nursing, Meenakshi Academy of Higher Education and Research
6Meenakshi College of Pharmacy, Meenakshi Academy of Higher Education and Research
Background: Heart Failure with Preserved Ejection Fraction (HFpEF) is a very heterogeneous disease that is not easily diagnosed in the early phases because of its heterogeneity complexity and the lack of biomarkers which are sensitive to and universal. The developments in artificial intelligence (AI) provide opportunities to examine complicated physiological and molecular data to reveal what were previously overlooked signs of early disease.
Objective: To assess whether AI-driven analytic algorithms detect new biomarkers to predict HFpEF at early stages with the help of multimodal clinical, imaging and molecular data.
Method: The effective inclusion of echocardiographic parameters, natriuretic peptides levels, proteomic profiles and wearable-generated physiological signals in a multi-centre cohort were used in this study. Machine-learning algorithms derived (expressing gradient boosting and deep neural networks) to categorize early-stage HFpEF were used to rank candidate biomarkers. Cross-validation and area under the curve (AUC) were used to estimate model performance.
Results: The AI models were very discriminative at early HFpEF (AUC 0.87). The application of feature-importance analysis revealed a number of potentially advantageous biomarkers, such as left-atrial strain, inflammatory protein-signature, microRNA cluster, and continuous hemodynamic patterns that are detected by wearables. These indicators enhanced the accuracy of an early detection over and above conventional clinical predictors and natriuretic peptides.
Conclusion: Biomarkers detectable much earlier in the stages of HFpEF can be identified through the help of AI-respected analyses, which have a clinical relevance. The combination of advanced analytics and multimodal diagnostics can significantly help to identify the risks and categorize them into levels of timely treatment in this multifaceted condition.
Keywords: Heart failure preserved ejection, left atrial strain, microRNA, early diagnosis, wearable monitoring, AI.
How to cite this article: Durga B, Subha VJ, Anandhi D, Shanthi V, Vasanthapriya J, Pugazhendi S. Artificial Intelligence to Identify Novel Biomarkers for Early Detection of Heart Failure with Preserved Ejection Fraction. Int J Drug Deliv Technol. 2026;16(10s): 96-101; DOI: 10.25258/ijddt.16.10s.13
Source of support: Nil.
Conflict of interest: None