AnyECG-Lab: An Exploration Study of Fine-tuning an ECG Foundation Model to Estimate Laboratory Values from Single-Lead ECG Signals

2025-10-25

AI Summary

Single-lead wearable ECG signals suffer from low signal-to-noise ratio, high inter-individual variability, and poor generalizability, limiting their clinical utility for noninvasive biomarker estimation. Method: This study pioneers the adaptation of ECGFounder—a large-scale pretrained ECG foundation model—to single-lead ECG for joint noninvasive prediction of 33 clinical biochemical markers. Training leveraged 20 million standardized 10-second ECG segments; fine-tuning and validation were conducted on the MC-MED dataset. Results: The model achieved AUC > 0.65 (good discrimination) for 33 markers, 0.55–0.65 (moderate) for 59, and < 0.55 (limited) for 16—demonstrating that single-lead ECG encodes observable latent biochemical features. This work extends the clinical applicability of ECG to real-time, noninvasive auxiliary diagnosis, establishing a novel paradigm for time-critical settings such as emergency care.

Technology Category

• Machine Learning: Other Foundations of Machine Learning
• Cognitive Modeling & Cognitive Systems: Other Foundations of Cognitive Modeling & Systems
• Humans and AI: Other Foundations of Human Computation & AI

Application Category

• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Security and Privacy: Large-scale security measurements

Abstract

Timely access to laboratory values is critical for clinical decision-making, yet current approaches rely on invasive venous sampling and are intrinsically delayed. Electrocardiography (ECG), as a non-invasive and widely available signal, offers a promising modality for rapid laboratory estimation. Recent progress in deep learning has enabled the extraction of latent hematological signatures from ECGs. However, existing models are constrained by low signal-to-noise ratios, substantial inter-individual variability, limited data diversity, and suboptimal generalization, especially when adapted to low-lead wearable devices. In this work, we conduct an exploratory study leveraging transfer learning to fine-tune ECGFounder, a large-scale pre-trained ECG foundation model, on the Multimodal Clinical Monitoring in the Emergency Department (MC-MED) dataset from Stanford. We generated a corpus of more than 20 million standardized ten-second ECG segments to enhance sensitivity to subtle biochemical correlates. On internal validation, the model demonstrated strong predictive performance (area under the curve above 0.65) for thirty-three laboratory indicators, moderate performance (between 0.55 and 0.65) for fifty-nine indicators, and limited performance (below 0.55) for sixteen indicators. This study provides an efficient artificial-intelligence driven solution and establishes the feasibility scope for real-time, non-invasive estimation of laboratory values.

Problem

• Estimating laboratory values from single-lead ECG signals non-invasively
• Addressing low signal-to-noise ratios and inter-individual variability in ECG analysis
• Enhancing generalization of ECG models for wearable device applications

Innovation

• Fine-tuned ECG foundation model for lab estimation
• Used transfer learning on large clinical dataset
• Generated standardized ECG segments to enhance sensitivity