LLM-Based Multimodal Planning for TAVR: Independent Validation, Calibration, and Efficiency Gains

Announcing a new article publication for Cardiovascular Innovations and Applications journal. A new study demonstrates that a multimodal large language model (LLM) can accurately support preoperative planning for transcatheter aortic valve replacement (TAVR). By integrating CT imaging and clinical data, the system automatically generates structured surgical plans, including valve selection, access strategy, and risk alerts. Tested across 950 patients, the model showed high agreement with expert heart teams while reducing planning time by over 90%. The findings highlight the potential of LLM-driven tools to improve efficiency and decision-making in complex cardiac procedures.

Read open access article: https://www.scienceopen.com/hosted-document?doi=10.15212/CVIA.2026.0006

# # # # # #

CVIA is available on the ScienceOpen platform and at Cardiovascular Innovations and Applications. Submissions may be made using ScholarOne Manuscripts. There are no author submission or article processing fees. Cardiovascular Innovations and Applications is indexed in the EMBASE, EBSCO, ESCI, OCLC, Primo Central (Ex Libris), Sherpa Romeo, NISC (National Information Services Corporation), DOAJ, Index Copernicus, Research4Life and Ulrich’s web Databases. Follow CVIA on Twitter @CVIA_Journal; or Facebook.

Xiaodong Wang, Yadan Li and Min Jin et al. LLM-Based Multimodal Planning for TAVR: Independent Validation, Calibration, and Efficiency Gains. CVIA. 2026. Vol. 11(1). DOI: 10.15212/CVIA.2026.0006