EchoTab improves echocardiographic diagnosis using measurement tokens
EchoTab: Learning Measurement-Grounded Representations for Echocardiographic Diagnosis through Tabular Reconstruction
Machine Learning
Summary
Echocardiography captures the heart from many views to understand its structure and function. The authors noticed that doctors first use specific heart measurements from these views to make diagnoses, but current computer methods do not reflect this step well. They created EchoTab, which learns to represent the heart by focusing on key measurements before making a diagnosis. This approach helps computers better understand echocardiograms and supports more accurate heart disease detection.
What this means in practice
- •For medical imaging developers: Improve automated echocardiogram diagnostic software by integrating measurement-focused representations for better accuracy.
- •For clinical workflow designers: Design clinical decision support tools that use intermediate cardiac measurements as explicit inputs to mirror cardiologist reasoning.
Authors
Michelle Espranita Liman, Ioana Ionac, Alexander Müller, Eimo Martens, Daniel Rueckert, Philip Müller
Abstract
Diagnostic information in multi-view medical imaging studies, like echocardiography, is distributed across views, each capturing different aspects of cardiac structure and function. In clinical practice, these views are first interpreted through quantitative measurements, which then support specific diagnoses. However, existing echo representation learning methods do not explicitly model measurements as an intermediate representation between views and diagnosis. As a result, individual measurements are not explicitly represented in a form that downstream diagnostic tasks can selectively use, hindering accurate diagnosis. Inspired by this hierarchical clinical workflow, we introduce EchoTab, a measurement-grounded representation learning framework that represents multi-view echocardiography studies using measurement-specific tokens as an intermediate representation between echo views and diagnosis. Each measurement token selectively attends to the views relevant to its estimation and is trained through tabular reconstruction to predict the corresponding measurement value. The resulting tokens form a factorized echo representation, which encode clinically meaningful structural and functional properties and can be selectively used for downstream diagnosis. Unlike the commonly used contrastive approaches, EchoTab does not require negative samples or large batch sizes. Experiments on two datasets demonstrate that EchoTab learns effective measurement-grounded representations that improve diagnosis classification in multi-view echocardiography and its learned attention aligns with established measurement-view relationships used by cardiologists. Our code is accessible at https://github.com/michelleespranita/EchoTab.