# SAMRI-3D: Adapting SAM2 for 3D MRI Segmentation with Global Volume Tokens

arXiv: https://arxiv.org/abs/2607.18014 · HTML: https://mlsanity.com/paper/2607.18014/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 35 references that are in the map.
- First posted on arXiv 20 Jul 2026.

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

- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [SAM 2: Segment Anything in Images and Videos](https://mlsanity.com/paper/2408.00714/) (2024)
- [Perceiver: General Perception with Iterative Attention](https://mlsanity.com/paper/2103.03206/) (2021)
- [Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images](https://mlsanity.com/paper/2201.01266/) (2022)
- [Medical SAM Adapter: Adapting Segment Anything Model for Medical Image Segmentation](https://mlsanity.com/paper/2304.12620/) (2023)
- [AMOS: A Large-Scale Abdominal Multi-Organ Benchmark for Versatile Medical Image Segmentation](https://mlsanity.com/paper/2206.08023/) (2022)
- [Medical SAM 2: Segment medical images as video via Segment Anything Model 2](https://mlsanity.com/paper/2408.00874/) (2024)
- [SAM-Med3D: Towards General-purpose Segmentation Models for Volumetric Medical Images](https://mlsanity.com/paper/2310.15161/) (2023)
- [MA-SAM: Modality-agnostic SAM Adaptation for 3D Medical Image Segmentation](https://mlsanity.com/paper/2309.08842/) (2023)
- [SAM2-UNet: Segment Anything 2 Makes Strong Encoder for Natural and Medical Image Segmentation](https://mlsanity.com/paper/2408.08870/) (2024)
- [TotalSegmentator MRI: Robust Sequence-independent Segmentation of Multiple Anatomic Structures in MRI](https://mlsanity.com/paper/2405.19492/) (2024)
- [Segment Anything in Medical Images and Videos: Benchmark and Deployment](https://mlsanity.com/paper/2408.03322/) (2024)
- [FlowSDF: Flow Matching for Medical Image Segmentation Using Distance Transforms](https://mlsanity.com/paper/2405.18087/) (2024)
- [Interactive 3D Medical Image Segmentation with SAM 2](https://mlsanity.com/paper/2408.02635/) (2024)
- [Novel adaptation of video segmentation to 3D MRI: efficient zero-shot knee segmentation with SAM2](https://mlsanity.com/paper/2408.04762/) (2024)
- [CAN3D: Fast 3D Medical Image Segmentation via Compact Context Aggregation](https://mlsanity.com/paper/2109.05443/) (2021)
- [Cross-Modality Domain Adaptation for Vestibular Schwannoma and Cochlea Segmentation](https://mlsanity.com/paper/2109.06274/) (2021)
- [TokenSeg: Efficient 3D Medical Image Segmentation via Hierarchical Visual Token Compression](https://mlsanity.com/paper/2601.04519/) (2026)
- [Transformer Compressed Sensing via Global Image Tokens](https://mlsanity.com/paper/2203.12861/) (2022)
- [SAMRI: Segment Any MRI](https://mlsanity.com/paper/2510.26635/) (2025)

## Abstract (arXiv)

Foundation models such as Segment Anything Model 2 (SAM2) have transformed natural-image and video segmentation, and recent work has begun adapting them to medical imaging. These adaptations, however, are largely general-purpose models that treat MRI as one modality among many; large-scale, MRI-specific modelling and benchmarking remain limited, even though MRI's low soft-tissue contrast leaves many boundaries effectively invisible on individual slices. We present SAMRI-3D, a benchmark and method for 3D MRI segmentation with SAM2. The SAMRI-3D benchmark is the largest MRI-only evaluation to date - 10,392 volumes from 34 datasets (27 public, 7 in-house) spanning 12 anatomical domains and 10+ sequences, with explicit seen/unseen splits. Freezing the image encoder and fine-tuning only the lightweight decoder and memory modules raises mean Dice from 0.58 (zero-shot SAM2) to 0.76, surpassing recent SAM-based medical models (SAMed-2 0.69, Medical-SAM2 0.49, SAM-Med3D 0.37) with strong statistical significance. To target invisible boundaries, we introduce Global Volume Tokens (GVT): persistent memory tokens trained with a Truncated Signed Distance Field (TSDF) reconstruction objective that is discarded at inference (zero added cost). This full model, SAMRI-3D, attains the best accuracy (0.78) and lowest variance across all 34 datasets and, uniquely, shows no drop on 8 held-out datasets (0.79 unseen vs. 0.78 seen); per-sequence analysis confirms the TSDF objective helps most where per-slice contrast is weakest. We will release the benchmark, code, and models in this paper.