Artificial Intelligence in Echocardiography

Artificial Intelligence (AI) holds tremendous potential to revolutionize cardiovascular imaging by enhancing diagnostic accuracy, improving workflow efficiency, and reducing healthcare disparities. However, the lack of standardized AI terminology and clear regulatory frameworks creates barriers to innovation and appropriate reimbursement. Current payment structures incorrectly assume AI reduces clinician workload when it may actually introduce new responsibilities for interpretation and oversight. Additionally, liability concerns regarding algorithm inaccuracies discourage clinicians from integrating AI tools, while inadequate diversity in AI training datasets risks perpetuating healthcare disparities. 

ASE evaluates AI-related policies, legislation, and regulations to ensure they enhance cardiovascular care while maintaining patient safety and clinical integrity. The Society supports standardized terminology for AI applications, rigorous FDA validation processes, payment models that recognize essential clinician expertise, and balanced liability frameworks that don't overburden frontline providers. These evidence-based positions enable AI to enhance cardiovascular care while preserving the fundamental importance of clinical judgment and patient-centered decision-making. 

ASE supports the AMA’s AI taxonomy for medical services & procedures: 

• Assistive - The work performed by the machine for the physician is assistive when the machine detects clinically relevant data without analysis or generated conclusions. Requires physician interpretation and report. 
• Augmentative - The work performed by the machine for the physician is augmentative when the machine analyzes and/or quantifies data to yield clinically meaningful output. Requires physician interpretation and report. 
• Autonomous - The work performed by the machine for the physician is autonomous when the machine automatically interprets data and independently generates clinically meaningful conclusions without concurrent physician involvement. Autonomous medical services and procedures include interrogating and analyzing data. The work of the algorithm may or may not include acquisition, preparation, and/or transmission of data.  

ASE AI CPT Designation Collaboration Process 

If developers of AI technology are interested in collaborating with ASE for reimbursement for their technology, developers are required to notify the Society before pursuing CPT designation. Full information can be found in the linked policy, “ASE AI Procedures for CPT Designation”.  

In summary, ASE strongly encourages engagement through the Industry Roundtable Partners structure. Applications must be submitted at least 8 weeks before AMA deadlines, and developers must have initiated FDA discussions, offer their service at one clinical site, and provide US clinical trial efficacy data.