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

arXiv: https://arxiv.org/abs/2505.09564 · HTML: https://mlsanity.com/paper/2505.09564/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 34 references that are in the map.
- First posted on arXiv 14 May 2025.

## Builds on (most cited first, 20 of 20)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [SAM 2: Segment Anything in Images and Videos](https://mlsanity.com/paper/2408.00714/) (2024)
- [Self-training with Noisy Student improves ImageNet classification](https://mlsanity.com/paper/1911.04252/) (2019)
- [TotalSegmentator: robust segmentation of 104 anatomical structures in CT images](https://mlsanity.com/paper/2208.05868/) (2022)
- [Confidence Regularized Self-Training](https://mlsanity.com/paper/1908.09822/) (2019)
- [Uncertainty-aware Self-ensembling Model for Semi-supervised 3D Left Atrium Segmentation](https://mlsanity.com/paper/1907.07034/) (2019)
- [Embracing Imperfect Datasets: A Review of Deep Learning Solutions for Medical Image Segmentation](https://mlsanity.com/paper/1908.10454/) (2019)
- [Not-so-supervised: a survey of semi-supervised, multi-instance, and transfer learning in medical image analysis](https://mlsanity.com/paper/1804.06353/) (2018)
- [On the Challenges and Perspectives of Foundation Models for Medical Image Analysis](https://mlsanity.com/paper/2306.05705/) (2023)
- [Deep learning for cardiac image segmentation: A review](https://mlsanity.com/paper/1911.03723/) (2019)
- [Two-phase Pseudo Label Densification for Self-training based Domain Adaptation](https://mlsanity.com/paper/2012.04828/) (2020)
- [3D Semi-Supervised Learning with Uncertainty-Aware Multi-View Co-Training](https://mlsanity.com/paper/1811.12506/) (2018)
- [Collaborating Foundation Models for Domain Generalized Semantic Segmentation](https://mlsanity.com/paper/2312.09788/) (2023)
- [Improving the Generalization of Segmentation Foundation Model under Distribution Shift via Weakly Supervised Adaptation](https://mlsanity.com/paper/2312.03502/) (2023)
- [Echocardiography Segmentation with Enforced Temporal Consistency](https://mlsanity.com/paper/2112.02102/) (2021)
- [4D CNN for semantic segmentation of cardiac volumetric sequences](https://mlsanity.com/paper/1906.07295/) (2019)
- [Extraction of volumetric indices from echocardiography: which deep learning solution for clinical use?](https://mlsanity.com/paper/2305.01997/) (2023)
- [Fully Automatic Myocardial Segmentation of Contrast Echocardiography Sequence Using Random Forests Guided by Shape Model](https://mlsanity.com/paper/1806.07497/) (2018)
- [STRUDEL: Self-Training with Uncertainty Dependent Label Refinement across Domains](https://mlsanity.com/paper/2104.11596/) (2021)

## Built on by (most cited first, 1 of 1)

- [SAM3-Assisted Training of Lightweight YOLO Models for Precision Pig Farming](https://mlsanity.com/paper/2605.25860/) (2026)

## Abstract (arXiv)

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.