Cardiac Output Prediction from Echocardiograms: Self-Supervised Learning with Limited Data

2026-02-14

AI Summary

This study addresses the limitations of traditional cardiac output measurement, which relies on invasive right heart catheterization, by proposing a non-invasive prediction method based on echocardiography. Leveraging only a small-scale dataset of apical four-chamber echocardiographic videos from the same downstream task and under limited annotation, the approach employs a SimCLR self-supervised learning framework integrated with deep convolutional networks and temporal modeling for pretraining. This strategy effectively enhances feature representation and generalization capability. On the test set, the method achieves an average Pearson correlation coefficient of 0.41, outperforming the PanEcho model pretrained on over one million echocardiographic images. These results demonstrate the efficacy and innovation of few-shot self-supervised pretraining in medical image analysis.

Technology Category

• Machine Learning: Unsupervised & Self-Supervised Learning
• Computer Vision: Multi-modal Vision
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Large pretrained models with web data
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Cardiac Output (CO) is a key parameter in the diagnosis and management of cardiovascular diseases. However, its accurate measurement requires right-heart catheterization, an invasive and time-consuming procedure, motivating the development of reliable non-invasive alternatives using echocardiography. In this work, we propose a self-supervised learning (SSL) pretraining strategy based on SimCLR to improve CO prediction from apical four-chamber echocardiographic videos. The pretraining is performed using the same limited dataset available for the downstream task, demonstrating the potential of SSL even under data scarcity. Our results show that SSL mitigates overfitting and improves representation learning, achieving an average Pearson correlation of 0.41 on the test set and outperforming PanEcho, a model trained on over one million echocardiographic exams. Source code is available at https://github.com/EIDOSLAB/cardiac-output.

Problem

• Cardiac Output
• Echocardiography
• Self-Supervised Learning
• Limited Data
• Non-invasive Prediction

Innovation

• Self-Supervised Learning
• Cardiac Output Prediction
• Echocardiography
• SimCLR
• Limited Data