# Do Medical Foundation Models Generalize on the African Brain?

arXiv: https://arxiv.org/abs/2607.28771 · HTML: https://mlsanity.com/paper/2607.28771/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 19 references that are in the map.
- First posted on arXiv 30 Jul 2026.

## Builds on (most cited first, 11 of 11)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [UNETR: Transformers for 3D Medical Image Segmentation](https://mlsanity.com/paper/2103.10504/) (2021)
- [Medical SAM 2: Segment medical images as video via Segment Anything Model 2](https://mlsanity.com/paper/2408.00874/) (2024)
- [Automated brain extraction of multi-sequence MRI using artificial neural networks](https://mlsanity.com/paper/1901.11341/) (2019)
- [MedSAM2: Segment Anything in 3D Medical Images and Videos](https://mlsanity.com/paper/2504.03600/) (2025)
- [A generalizable 3D framework and model for self-supervised learning in medical imaging](https://mlsanity.com/paper/2501.11755/) (2025)
- [Foundation Models in Radiology: What, How, When, Why and Why Not](https://mlsanity.com/paper/2411.18730/) (2024)
- [SANSA: Unleashing the Hidden Semantics in SAM2 for Few-Shot Segmentation](https://mlsanity.com/paper/2505.21795/) (2025)
- [A labeled Clinical-MRI dataset of Nigerian brains](https://mlsanity.com/paper/2311.04425/) (2023)
- [State of Abdominal CT Datasets: A Critical Review of Bias, Clinical Relevance, and Real-world Applicability](https://mlsanity.com/paper/2508.13626/) (2025)
- [Feature Quality and Adaptability of Medical Foundation Models: A Comparative Evaluation for Radiographic Classification and Segmentation](https://mlsanity.com/paper/2511.09742/) (2025)

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

- [Dataset-Specific Encoding of Image Content Underlies Cross-Dataset Failure in Brain MRI Foundation Models](https://mlsanity.com/paper/2610.13027/) (2026)

## Abstract (arXiv)

Medical foundation models (FMs) are increasingly used for brain MRI analysis. However, their evaluation remains dominated by high-resource datasets, leaving generalization to African cohorts underexplored. We assess whether FMs generalize equally to African and non-African brain MRI data across two tasks: dementia classification using a Nigerian dataset and brain tumor segmentation using BraTS-Africa. We evaluate two generalist FMs (BrainIAC, 3DINO) and two segmentation-specific FMs (MedSAM2, Medical-SAM2) against a from-scratch baseline. For classification, FMs provide limited gains (highest ROC-AUC of 0.86 with BrainIAC), whereas for segmentation they consistently improve performance, reaching up to 0.86 Dice with MedSAM2. Performance differences between African and non-African cohorts are inconsistent and appear more related to dataset size than data origin. These results suggest that FMs do not exhibit an inherent bias against African cohorts, and highlight the limited availability and diversity of African neuroimaging datasets as the main barrier to robust evaluation and deployment.