The implementation of artificial intelligence (AI) in cardiology presents significant challenges despite its potential to revolutionize clinical practice. Key barriers include issues related to data acquisition, ethical concerns, regulatory frameworks, and economic considerations. In imaging, the variability in data acquisition and the lack of structured, labeled datasets limit the development and validation of AI-driven tools. Similarly, AI applications in electrocardiography (ECG) are hindered by the need for large, well-annotated datasets to ensure generalizability across populations. The integration of AI in wearable devices introduces privacy concerns and necessitates robust regulatory pathways. Electronic health records (EHR) offer valuable opportunities for AI-driven insights, but differences in data quality and biases across healthcare settings pose challenges. Risk prediction models, while highly sensitive, often suffer from low specificity, increasing the risk of overdiagnosis and unnecessary testing. Additionally, ethical, legal, and regulatory considerations remain central to AI adoption, particularly in defining liability, informed consent, and data privacy. Despite these challenges, AI holds great promise for improving cardiovascular care and is expected to play an essential role in future clinical practice.
Artificial Intelligence in Cardiology: Ethical, Regulatory Considerations, and Conclusions
D'Ascenzo, Fabrizio;Dusi, Veronica;de Filippo, Ovidio
2025-01-01
Abstract
The implementation of artificial intelligence (AI) in cardiology presents significant challenges despite its potential to revolutionize clinical practice. Key barriers include issues related to data acquisition, ethical concerns, regulatory frameworks, and economic considerations. In imaging, the variability in data acquisition and the lack of structured, labeled datasets limit the development and validation of AI-driven tools. Similarly, AI applications in electrocardiography (ECG) are hindered by the need for large, well-annotated datasets to ensure generalizability across populations. The integration of AI in wearable devices introduces privacy concerns and necessitates robust regulatory pathways. Electronic health records (EHR) offer valuable opportunities for AI-driven insights, but differences in data quality and biases across healthcare settings pose challenges. Risk prediction models, while highly sensitive, often suffer from low specificity, increasing the risk of overdiagnosis and unnecessary testing. Additionally, ethical, legal, and regulatory considerations remain central to AI adoption, particularly in defining liability, informed consent, and data privacy. Despite these challenges, AI holds great promise for improving cardiovascular care and is expected to play an essential role in future clinical practice.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
