# A Systematic Benchmark of Intensity Normalisation Methods for 3D Knee MRI Segmentation and Cross-Domain Generalisability

arXiv: https://arxiv.org/abs/2607.20028 · HTML: https://mlsanity.com/paper/2607.20028/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 46 references that are in the map.
- First posted on arXiv 22 Jul 2026.

## Builds on (most cited first, 10 of 10)

- [A Survey on Deep Learning in Medical Image Analysis](https://mlsanity.com/paper/1702.05747/) (2017)
- [nnU-Net Revisited: A Call for Rigorous Validation in 3D Medical Image Segmentation](https://mlsanity.com/paper/2404.09556/) (2024)
- [TorchIO: A Python library for efficient loading, preprocessing, augmentation and patch-based sampling of medical images in deep learning](https://mlsanity.com/paper/2003.04696/) (2020)
- [SKM-TEA: A Dataset for Accelerated MRI Reconstruction with Dense Image Labels for Quantitative Clinical Evaluation](https://mlsanity.com/paper/2203.06823/) (2022)
- [Domain Shift in Computer Vision models for MRI data analysis: An Overview](https://mlsanity.com/paper/2010.07222/) (2020)
- [The International Workshop on Osteoarthritis Imaging Knee MRI Segmentation Challenge: A Multi-Institute Evaluation and Analysis Framework on a Standardized Dataset](https://mlsanity.com/paper/2004.14003/) (2020)
- [Evaluation of Preprocessing Techniques for U-Net Based Automated Liver Segmentation](https://mlsanity.com/paper/2103.14301/) (2021)
- [Negligible effect of brain MRI data preprocessing for tumor segmentation](https://mlsanity.com/paper/2204.05278/) (2022)
- [Effect of Intensity Standardization on Deep Learning for WML Segmentation in Multi-Centre FLAIR MRI](https://mlsanity.com/paper/2307.03827/) (2023)
- [Putting the Segment Anything Model to the Test with 3D Knee MRI - A Comparison with State-of-the-Art Performance](https://mlsanity.com/paper/2504.13340/) (2025)

## Abstract (arXiv)

Robust out-of-the-box performance is essential for the clinical deployment of deep learning models in medical imaging. An important but underexplored factor affecting model generalisability is intensity normalisation, particularly for magnetic resonance imaging (MRI), where image intensities vary across scanners and protocols. In this study, we systematically compared seven normalisation methods and their impact on the performance of a 3D U-Net model for meniscus segmentation from knee MRI. The methods included standard scaling approaches, histogram-based techniques, and a Gaussian Mixture Model (GMM)-based method. Models were trained on the IWOAI 2019 dataset and evaluated on both internal and external test sets (SKM-TEA) to assess generalisability. Performance was similar internally but differences were significant on external data, with Z-score, Ny'ul histogram matching, and CLAHE showing greater robustness than other methods. However, these differences were small compared to the significant performance drop observed between datasets. Overall, while intensity normalisation had a measurable effect on model generalisability, its impact was limited relative to the effects of domain shift, highlighting the need for complementary strategies for robust deployment.