The Amyloidosis Channel is supported through an educational grant from Alexion Pharma GmbH.
VJHemOnc is an independent medical education platform. Supporters, including channel supporters, have no influence over the production of content. The levels of sponsorship listed are reflective of the amount of funding given to support the channel.
ESC 2026 | Addressing the risk of false positives with an AI-guided ATTR-CM screening model
Ilia Davarashvili, MD, Leumit Health Services and Leumit Start, Tel Aviv, Israel, discusses the clinical utility and addresses the risk of false positives associated with an artificial intelligence (AI)-guided screening model for cardiac transthyretin amyloidosis (ATTR-CM). He highlights that, as a screening and assistance tool rather than a diagnostic tool, it is challenging to assess false positives or negatives directly. This interview took place during the 2026 European Society of Cardiology (ESC) Congress in Munich, Germany.
These works are owned by Magdalen Medical Publishing (MMP) and are protected by copyright laws and treaties around the world. All rights are reserved.
Transcript
Saying the truth, as I’ve told before, AI is not a diagnostic tool. It’s a screening tool and an assist tool. So it’s very hard to say straightforwardly that we can assess the false positive or false negative issue. But we can assess the sensitivity of identifying across a big healthcare database, screening tens of thousands of records, 210 patients were identified with a high suspicion of heart failure, knowing that we can recognize them earlier before clinical overt symptoms appear...
Saying the truth, as I’ve told before, AI is not a diagnostic tool. It’s a screening tool and an assist tool. So it’s very hard to say straightforwardly that we can assess the false positive or false negative issue. But we can assess the sensitivity of identifying across a big healthcare database, screening tens of thousands of records, 210 patients were identified with a high suspicion of heart failure, knowing that we can recognize them earlier before clinical overt symptoms appear. So, as I’ve told before, 10 times high yield is very impressive. And we are going forward to learn more with AI.
This transcript is AI-generated. While we strive for accuracy, please verify this copy with the video.
Read more...
