# Causal Attribution of Model Performance Gaps in Medical Imaging Under Distribution Shifts

arXiv: https://arxiv.org/abs/2512.09094 · HTML: https://mlsanity.com/paper/2512.09094/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 6 of its 16 references that are in the map.
- First posted on arXiv 9 Dec 2025.

## Builds on (most cited first, 6 of 6)

- [Improved Training of Wasserstein GANs](https://mlsanity.com/paper/1704.00028/) (2017)
- [Failing Loudly: An Empirical Study of Methods for Detecting Dataset Shift](https://mlsanity.com/paper/1810.11953/) (2018)
- [Causality matters in medical imaging](https://mlsanity.com/paper/1912.08142/) (2019)
- [Causal Machine Learning for Healthcare and Precision Medicine](https://mlsanity.com/paper/2205.11402/) (2022)
- [Why did the distribution change?](https://mlsanity.com/paper/2102.13384/) (2021)
- [A Framework for Interpretability in Machine Learning for Medical Imaging](https://mlsanity.com/paper/2310.01685/) (2023)

## Abstract (arXiv)

Deep learning models for medical image segmentation suffer significant performance drops due to distribution shifts, but the causal mechanisms behind these drops remain poorly understood. We extend causal attribution frameworks to high-dimensional segmentation tasks, quantifying how acquisition protocols and annotation variability independently contribute to performance degradation. We model the data-generating process through a causal graph and employ Shapley values to fairly attribute performance changes to individual mechanisms. Our framework addresses unique challenges in medical imaging: high-dimensional outputs, limited samples, and complex mechanism interactions. Validation on multiple sclerosis (MS) lesion segmentation across 4 centers and 7 annotators reveals context-dependent failure modes: annotation protocol shifts dominate when crossing annotators (7.4% ± 8.9% DSC attribution), while acquisition shifts dominate when crossing imaging centers (6.5% ± 9.1%). This mechanism-specific quantification enables practitioners to prioritize targeted interventions based on deployment context.