# Is Exchangeability better than I.I.D to handle Data Distribution Shifts while Pooling Data for Data-scarce Medical image segmentation?

arXiv: https://arxiv.org/abs/2507.19575 · HTML: https://mlsanity.com/paper/2507.19575/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 44 of its 87 references that are in the map.
- First posted on arXiv 25 Jul 2025.

## Builds on (most cited first, 30 of 44)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [Fully Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1411.4038/) (2014)
- [TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation](https://mlsanity.com/paper/2102.04306/) (2021)
- [Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentation](https://mlsanity.com/paper/2105.05537/) (2021)
- [Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distribution](https://mlsanity.com/paper/2202.10054/) (2022)
- [Deep Double Descent: Where Bigger Models and More Data Hurt](https://mlsanity.com/paper/1912.02292/) (2019)
- [Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets](https://mlsanity.com/paper/2201.02177/) (2022)
- [Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs](https://mlsanity.com/paper/1412.7062/) (2014)
- [Train faster, generalize better: Stability of stochastic gradient descent](https://mlsanity.com/paper/1509.01240/) (2015)
- [Unlabeled Data Improves Adversarial Robustness](https://mlsanity.com/paper/1905.13736/) (2019)
- [HoVer-Net: Simultaneous Segmentation and Classification of Nuclei in Multi-Tissue Histology Images](https://mlsanity.com/paper/1812.06499/) (2018)
- [Medical Transformer: Gated Axial-Attention for Medical Image Segmentation](https://mlsanity.com/paper/2102.10662/) (2021)
- [Contrastive learning of global and local features for medical image segmentation with limited annotations](https://mlsanity.com/paper/2006.10511/) (2020)
- [Segment Anything Model for Medical Image Analysis: an Experimental Study](https://mlsanity.com/paper/2304.10517/) (2023)
- [CE-Net: Context Encoder Network for 2D Medical Image Segmentation](https://mlsanity.com/paper/1903.02740/) (2019)
- [A survey of loss functions for semantic segmentation](https://mlsanity.com/paper/2006.14822/) (2020)
- [Causality matters in medical imaging](https://mlsanity.com/paper/1912.08142/) (2019)
- [Synergistic Image and Feature Adaptation: Towards Cross-Modality Domain Adaptation for Medical Image Segmentation](https://mlsanity.com/paper/1901.08211/) (2019)
- [Invariant Representations without Adversarial Training](https://mlsanity.com/paper/1805.09458/) (2018)
- [TGANet: Text-guided attention for improved polyp segmentation](https://mlsanity.com/paper/2205.04280/) (2022)
- [PathLDM: Text conditioned Latent Diffusion Model for Histopathology](https://mlsanity.com/paper/2309.00748/) (2023)
- [Computer-Vision Benchmark Segment-Anything Model (SAM) in Medical Images: Accuracy in 12 Datasets](https://mlsanity.com/paper/2304.09324/) (2023)
- [3D Deeply Supervised Network for Automatic Liver Segmentation from CT Volumes](https://mlsanity.com/paper/1607.00582/) (2016)
- [CMU-Net: A Strong ConvMixer-based Medical Ultrasound Image Segmentation Network](https://mlsanity.com/paper/2210.13012/) (2022)
- [Scanner Invariant Representations for Diffusion MRI Harmonization](https://mlsanity.com/paper/1904.05375/) (2019)
- [On the Over-Memorization During Natural, Robust and Catastrophic Overfitting](https://mlsanity.com/paper/2310.08847/) (2023)
- [TransResU-Net: Transformer based ResU-Net for Real-Time Colonoscopy Polyp Segmentation](https://mlsanity.com/paper/2206.08985/) (2022)
- [HistoSeg : Quick attention with multi-loss function for multi-structure segmentation in digital histology images](https://mlsanity.com/paper/2209.00729/) (2022)
- [The Data Addition Dilemma](https://mlsanity.com/paper/2408.04154/) (2024)

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

- [ManifoldGD: Training-Free Hierarchical Manifold Guidance for Diffusion-Based Dataset Distillation](https://mlsanity.com/paper/2602.23295/) (2026)

## Abstract (arXiv)

Data scarcity is a major challenge in medical imaging, particularly for deep learning models. While data pooling (combining datasets from multiple sources) and data addition (adding more data from a new dataset) have been shown to enhance model performance, they are not without complications. Specifically, increasing the size of the training dataset through pooling or addition can induce distributional shifts, negatively affecting downstream model performance, a phenomenon known as the "Data Addition Dilemma". While the traditional i.i.d. assumption may not hold in multi-source contexts, assuming exchangeability across datasets provides a more practical framework for data pooling. In this work, we investigate medical image segmentation under these conditions, drawing insights from causal frameworks to propose a method for controlling foreground-background feature discrepancies across all layers of deep networks. This approach improves feature representations, which are crucial in data-addition scenarios. Our method achieves state-of-the-art segmentation performance on histopathology and ultrasound images across five datasets, including a novel ultrasound dataset that we have curated and contributed. Qualitative results demonstrate more refined and accurate segmentation maps compared to prominent baselines across three model architectures.