Artificial intelligence-based calcium scoring using 3D transesophageal echocardiography in aortic stenosis: a pilot...

Area of Science:

• Cardiovascular imaging
• Artificial intelligence in medicine
• Echocardiography

Background:

• Computed tomography (CT) for aortic valve calcium scoring is limited by radiation and availability.
• Artificial intelligence (AI) using transthoracic echocardiography shows promise but depends on acoustic windows.
• 3D transesophageal echocardiography (TEE) offers improved visualization without radiation.

Purpose of the Study:

• To evaluate the feasibility of AI-based quantification of aortic valve calcium using 3D TEE.
• To assess the correlation between AI-derived TEE calcium scores and CT-derived Agatston scores.

Main Methods:

• A prospective pilot study included 23 patients with moderate or severe aortic stenosis.
• 3D TEE and CT scans were performed; multiplanar reconstructions generated short-axis slices.
• A computer vision AI model identified calcium speckles, and an automated TEE calcium score was calculated.

Main Results:

• The TEE calcium score significantly correlated with CT Agatston scores (r=0.65, P<0.001).
• AI-based TEE demonstrated an area under the curve of 0.87 for identifying severe calcification.
• High sensitivity (89.5%) and specificity (75.0%) were achieved for severe calcification detection.

Conclusions:

• AI-based calcium quantification using 3D TEE is feasible.
• This radiation-free approach shows strong correlation with CT-derived scores.
• 3D TEE with AI presents a promising alternative for aortic valve calcification assessment.