Using Foundation Models as Pseudo-Label Generators for Pre-Clinical 4D Cardiac CT Segmentation

2025-05-14

AI Summary

Domain shift arising from cross-species anatomical differences and severe annotation scarcity hinder accurate 4D cardiac CT segmentation in porcine models. Method: We propose a zero-shot self-training framework that first adapts human-pretrained medical foundation models (e.g., SAM) to porcine cardiac CT for high-quality initial pseudo-label generation; then enforces temporal consistency constraints and iterative retraining to mitigate inter-frame jitter inherent in 4D sequences. Contribution/Results: Without requiring any manually annotated porcine data, our method significantly improves segmentation accuracy for cardiac chambers and myocardium while enhancing temporal stability across the 4D volume. It establishes a generalizable, cross-species adaptation paradigm for preclinical cardiac imaging analysis.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Cognitive Modeling & Cognitive Systems: Other Foundations of Cognitive Modeling & Systems

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Cardiac image segmentation is an important step in many cardiac image analysis and modeling tasks such as motion tracking or simulations of cardiac mechanics. While deep learning has greatly advanced segmentation in clinical settings, there is limited work on pre-clinical imaging, notably in porcine models, which are often used due to their anatomical and physiological similarity to humans. However, differences between species create a domain shift that complicates direct model transfer from human to pig data. Recently, foundation models trained on large human datasets have shown promise for robust medical image segmentation; yet their applicability to porcine data remains largely unexplored. In this work, we investigate whether foundation models can generate sufficiently accurate pseudo-labels for pig cardiac CT and propose a simple self-training approach to iteratively refine these labels. Our method requires no manually annotated pig data, relying instead on iterative updates to improve segmentation quality. We demonstrate that this self-training process not only enhances segmentation accuracy but also smooths out temporal inconsistencies across consecutive frames. Although our results are encouraging, there remains room for improvement, for example by incorporating more sophisticated self-training strategies and by exploring additional foundation models and other cardiac imaging technologies.

Problem

• Addressing domain shift in cardiac CT segmentation from human to porcine models
• Exploring foundation models for pseudo-label generation in pre-clinical imaging
• Improving segmentation accuracy and temporal consistency via self-training

Innovation

• Foundation models generate pseudo-labels for pig CT
• Self-training refines labels without manual annotation
• Iterative updates improve segmentation and temporal consistency