Aortic Valve (TAVR/TAVI)Friday, September 25, 2026

TAVI-TEC: an AI-based tool for procedural planning of transcatheter aortic valve implantation.

1 min read·By E. Nolan Beckett, MD·Source: Cardiovascular intervention and therapeutics

Key Numbers

77.1%

95% CI

83.9%

From The Valve Wire

TAVI-TEC is a fully automated AI pipeline that segments cardiovascular structures, extracts annular measurements, and predicts SAPIEN 3 Ultra valve size from preprocedural CTA.

In this single-center development cohort, it produced annular area and perimeter measurements with strong agreement to clinician values (CCC 0.93 and 0.91) in 2-6 minutes, and predicted the implanted valve size correctly 77% of the time, with most misses landing on the adjacent size.

This is a technical feasibility report, not a clinical validation.

Development and testing occurred at one institution with one valve platform; there is no external cohort, no comparison against inter-reader variability among clinicians, and no outcome data on paravalvular leak, annular rupture, or pacemaker rates when the algorithm's size is followed.

Source Abstract

Computed tomography angiography (CTA) is crucial for preprocedural TAVI planning, providing the anatomical information required for prosthesis sizing and vascular access assessment. As the volume of TAVI procedure increases, improving efficiency and standardizing annotations is becoming essential in clinical practice. This study presents TAVI-TEC, a fully automated artificial intelligence-based framework integrated into a web-based DICOM viewer for routine preoperative TAVI planning. Pre-procedural CTA scans from patients undergoing TAVI with SAPIEN 3 Ultra (S3U) prostheses were processed using a fully automated pipeline. Deep learning-based segmentation of cardiovascular structures, calcification detection, centerline extraction, landmark identification, and annular plane definition was implemented to quantify key annular and aortic root measurements and color-coded maps of lumen reduction and vessel diameter for vascular access. A multilayer perceptron classifier was trained to predict prosthesis size prior to the TAVI procedure. Results revealed that TAVI-TEC enabled pre-procedural measurements in approximately 2-6 min. Strong agreement with clinician-derived measurements was observed for annular area (coefficient of concordance, CCC = 0.934; interclass correlation coefficient, ICC = 0.935; R² = 0.881) and perimeter (CCC = 0.909; ICC = 0.909; R² = 0.854). The valve-size prediction model achieved a classification accuracy of 77.1% (95% CI: 70.1-83.9%) with the clinically selected and implanted valve size, with most discordances occurring between adjacent prosthesis sizes. These findings support the technical feasibility and promising initial performance of TAVI-TEC. Prospective multicenter and external validation, including additional valve platforms, will be required before routine clinical implementation and before its effects on operator variability and clinical workflow can be established.

Authors: Zerillo A, Cannata S, Bellavia D, Ciriello D, Manini S et al.

Read the Original

Full article at Cardiovascular intervention and therapeutics

Open at Cardiovascular intervention and therapeutics →

The Valve Wire summarizes; we link out to primary sources for verification.