---
title: 'nnsam2: No-New SAM2 Segmentation Framework'
url: https://www.emergentmind.com/topics/no-new-sam2-nnsam2
type: topic
---

# nnsam2: No-New SAM2 Segmentation Framework

No-New SAM2, abbreviated **nnsam2**, is a few-shot framework for multi-modality segmentation of lumbar paraspinal muscles that uses **SAM2 without fine-tuning or architectural modification** and then refines the resulting pseudo-labels with three sequential, independent nnU-Net models [2510.05555]. It was introduced for segmentation of the combined **multifidus (MF)** and **lumbar erector spinae (LES)** in the **L4–L5 region**, operationalized as slices spanning **L3/L4 to L5/S1 disc levels**, across multi-sequence MRI and multi-protocol CT [2510.05555]. In the reported minimal-supervision regime, **one annotated slice per dataset** was used to seed SAM2, and the framework was evaluated not only as a segmentation system but also as a pipeline for **muscle volume**, **Dixon fat ratio**, and **CT attenuation** measurement [2510.05555].

## 1. Task definition and supervision regime

nnsam2 was developed for **lumbar paraspinal muscle (LPM)** analysis in heterogeneous cross-domain imaging cohorts. The study used **1,219 scans (19,439 slices) from 762 participants across six datasets**, including **943 MRI scans from 486 participants** and **276 CT scans from 276 participants** [2510.05555]. The supervision budget was deliberately minimal: **six slices, one per dataset, served as labeled examples**, and the remaining **19,433 slices** were used for testing [2510.05555].

The target structures were the left and right lumbar paraspinal muscles in axial MRI or CT around the lower lumbar region. **MF and LES were segmented jointly** as one class per side, because they cannot be reliably distinguished on CT; the same convention was then used on MRI for consistency [2510.05555]. This design tied the method directly to downstream composition analysis rather than to class-rich anatomical parsing.

The six source datasets spanned markedly different acquisition settings. MRI cohorts included **AFL T2-weighted MRI**, **back-pain T1-weighted MRI**, **back-pain T2-weighted MRI**, and **AGBRESA Dixon MRI**; CT cohorts included **TotalSegmentator CT** and **WORD CT** [2510.05555]. This heterogeneity is central to the method’s intended role as a few-shot, cross-dataset, cross-modality framework rather than a sequence-specific or site-specific segmenter.

## 2. Meaning of “No-New SAM2”

In nnsam2, **“No-New SAM2” means that SAM2 itself is not fine-tuned and not architecturally modified** [2510.05555]. The contribution is therefore not a new SAM2 checkpoint, adapter stack, or prompt encoder, but a pipeline that uses **frozen SAM2 for prompt-based pseudo-label generation** and **nnU-Net for 3D contextual refinement** [2510.05555]. A plausible implication is that the phrase “No-New” applies specifically to the SAM2 component rather than to the absence of all additional learning stages.

This places nnsam2 at a conservative point within the broader SAM2 adaptation literature. Other SAM2-based systems add trainable adapters while freezing only part of the backbone, as in **SAM2-Adapter** [2408.04579], repackage the SAM2 encoder inside a U-shaped segmentation model, as in **SAM2-UNet** [2408.08870], wrap SAM2 in a training-free detector-guided tracking system, as in **Seg2Track-SAM2** [2509.11772], or use SAM2 directly in a zero-shot pseudo-video medical pipeline, as in **3D knee MRI segmentation with SAM2** [2408.04762]. Relative to those variants, nnsam2 preserves the original SAM2 model more strictly while moving the main learning burden into a downstream refinement cascade.

## 3. Pseudo-label generation with frozen SAM2

The first stage of nnsam2 is **single-slice SAM2 pseudo-label generation** [2510.05555]. Before SAM2 inference, **MRI intensities were clipped to the 0.5th–99.5th percentiles**, **CT intensities were clipped to \([-30,150]\) HU**, CT volumes were downsampled along the head-foot axis by **3× for TotalSegmentator CT** and **2× for WORD CT**, and all MR and CT images were **rescaled to \([0,255]\)**, **converted to JPEG**, and **resized to \(256\times256\)** [2510.05555].

For each dataset, a **reference volume** was selected as the volume with the **smallest average feature distance** across the **L3/L4, L4/L5, and L5/S1** disc slices using **DINOv2 features** [2510.05555]. A single manually annotated slice from that reference volume—the **top slice**—served as the SAM2 prompt seed [2510.05555]. SAM2 was then applied through an **interleaved sequence** that alternated slices from the reference volume and each target inference volume, proceeding from superior to inferior slices across the lumbar region [2510.05555].

This design allowed SAM2’s sequential propagation mechanism to transfer one prompted annotation into pseudo-labels for the remainder of a dataset [2510.05555]. For each predicted mask, SAM2 also produced an **IoU score**, which was treated as a **confidence estimate** and later used for pseudo-label selection [2510.05555]. The output of this stage was therefore a confidence-ranked set of SAM2 pseudo-labels rather than a final segmentation.

## 4. Three-stage nnU-Net refinement cascade

The second stage of nnsam2 refines SAM2 pseudo-labels through **three sequential, independent nnU-Net models** [2510.05555]. The purpose of this cascade is to suppress pseudo-label noise while introducing 3D anatomical consistency that SAM2 alone does not impose in this application [2510.05555].

In **Refinement Step 1**, the method retained the **top 10%** of SAM2 masks ranked by IoU at the dataset level and the **top 2%** at the slice level, then pooled the selected pseudo-labels from all datasets to train **nnU-Net 1** [2510.05555]. The reported rationale was empirical: **5% gave too few masks**, whereas **20% introduced too much noise**; the slice-level **2%** preserved anatomically difficult slices that might otherwise be lost [2510.05555].

In **Refinement Step 2**, nnU-Net 1 predictions were kept only if their **mean DSC with the SAM2 prediction was \(>0.90\)** and their **cross-sectional area was no larger than \(1.5\times\)** that of the superior slice [2510.05555]. From these validated predictions, the **top 10%** were used to train **nnU-Net 2** [2510.05555]. In **Refinement Step 3**, nnU-Net 2 predictions were compared against nnU-Net 1 predictions, retained only if **DSC \(>0.90\)** and area was **\(\le 1.25\times\)** the superior slice area, and the **top 20%** were used to train **nnU-Net 3**, which served as the final predictor [2510.05555]. The threshold logic was again empirical: **0.85 admitted noisy masks**, **0.95 excluded too many valid masks**, **\(2.0\times\)** area growth was too loose, and **\(1.25\times\)** was too strict too early [2510.05555].

The nnU-Net implementation used **nnU-Net version 1.7.1**, trained for **1,000 epochs** with initial learning rate **0.01** and **hybrid Dice-cross-entropy loss** [2510.05555]. **Mirroring was disabled** to avoid anatomical mislabeling, and postprocessing kept only the **largest connected region per muscle class** [2510.05555]. This suggests that nnsam2 relies on progressively cleaner supervision and explicit anatomical smoothness constraints rather than a single pseudo-labeling pass.

## 5. Datasets, baselines, and segmentation accuracy

nnsam2 was evaluated against **vanilla SAM2**, **MedSAM2**, **Medical SAM2**, **FAMNet**, and **TotalSegmentator** [2510.05555]. The abstract reports that nnsam2 **outperformed vanilla SAM2, its medical variants, TotalSegmentator, and the leading few-shot method**, achieving **DSCs of 0.94–0.96 on MR images and 0.92–0.93 on CT** [2510.05555].

The per-dataset numbers reported in the study were consistent with that summary. On MRI, nnsam2 reached **0.95/0.95** left/right DSC on **AFL T2W**, **0.96/0.96** on **back-pain T2W**, **0.96/0.96** on **back-pain T1W**, and **0.94/0.94** on **AGBRESA Dixon** [2510.05555]. On CT, it reached **0.92/0.93** on **TotalSegmentator CT** and **0.92/0.92** on **WORD CT** [2510.05555]. The paper summary further notes that improvements over vanilla SAM2 were generally around **0.02–0.03 DSC**, whereas improvements over **FAMNet** were much larger and often exceeded **0.20** [2510.05555].

The baseline behavior is also informative. Vanilla SAM2 was already a strong pseudo-label generator in this setting, with DSCs around **0.90–0.94** across the six datasets, but it did not match the refined nnsam2 outputs [2510.05555]. **Medical SAM2** and **MedSAM2** were less stable across the mixed cohorts, and **TotalSegmentator**—used off the shelf rather than fine-tuned—did not reach nnsam2 performance on the target muscles [2510.05555]. This suggests that the main gain of nnsam2 came from converting sparse prompt-based pseudo-labels into a pooled, denoised, anatomically regularized volumetric training signal.

## 6. Composition analysis and statistical agreement

A defining characteristic of nnsam2 is that it was evaluated not only by segmentation overlap but also by the statistical comparability of derived muscle measurements [2510.05555]. The reported endpoints were **muscle volume** from MRI and CT, **fat ratio** from Dixon MRI, and **CT attenuation** in Hounsfield units [2510.05555]. According to the abstract, **automated and expert measurements were statistically equivalent for muscle volume (MRI/CT), CT attenuation, and Dixon fat ratio**, with **TOST \(P < 0.05\)** and **ICCs ranging from 0.86 to 1.00** [2510.05555].

The statistical framework combined **two one-sided tests (TOST)**, **intraclass correlation coefficients (ICC)**, and model-based **Bland–Altman** analysis [2510.05555]. The paper gives the ICC as
\[
ICC = \frac{\sigma^2_{between}}{\sigma^2_{between} + \sigma^2_{within}},
\]
where \(\sigma^2_{between}\) is variance across individuals and \(\sigma^2_{within}\) is residual or within-subject variance [2510.05555]. For Bland–Altman analysis using linear mixed-effects models, the limits of agreement were
\[
LoA = \text{mean bias} \pm 1.96 \times \text{overall standard deviation}.
\]
Equivalence in TOST was concluded when the **90% confidence interval** for the estimated method effect lay entirely within the predefined equivalence bounds, with equivalence accepted if **\(p \le 0.05\)** [2510.05555].

The reported measurement agreement was strongest on MRI, especially in the back-pain cohorts, where ICCs for volume reached **1.00** on both sides for both **T1W** and **T2W** MRI [2510.05555]. On AGBRESA Dixon MRI, fat-ratio ICCs were **0.97** and **0.96** for left and right sides, respectively [2510.05555]. CT remained slightly harder, but volume and attenuation still reached high agreement, with CT attenuation ICCs up to **0.99** on TotalSegmentator CT [2510.05555]. This measurement layer is central to the clinical framing of nnsam2: segmentation quality was treated as a means to reproducible quantitative phenotype extraction rather than as the sole endpoint.

## 7. Limitations and position within SAM2 adaptation research

The study identifies several limitations. Manual references were produced by a **trained annotator plus one clinical professor reviewer**, rather than multiple independent expert raters [2510.05555]. **MF and LES were grouped into a single class**, which constrains finer-grained anatomical analysis [2510.05555]. The framework also depends on **nnU-Net refinement**, and the paper notes that if data are insufficient, one may need to use **SAM2 directly without nnU-Net enhancement** [2510.05555]. Reported failure cases occurred in scans with **poor image resolution** or **indistinct muscle boundaries** [2510.05555].

Within the broader SAM2 ecosystem, nnsam2 occupies a distinct methodological niche. Some medical SAM2 systems reinterpret volumes as videos and run zero-shot inference with no additional training, as in **3D knee MRI segmentation with SAM2** [2408.04762]. Others remove manual prompting through support sets and LoRA, as in **SAM2-SGP** [2506.19658], or introduce explicit auxiliary modules for volumetric continuity and boundary supervision, as in **SAM2-3dMed** [2510.08967] and **UniUltra** [2511.15771]. By contrast, nnsam2 keeps SAM2 unchanged and shifts adaptation into a downstream pseudo-label refinement regime [2510.05555]. This suggests that nnsam2 is best understood not as a new SAM2 model, but as a **few-shot bootstrap strategy** that exploits frozen SAM2 generalization and then regularizes it with sequential nnU-Net refinement.

The study also released **open code and data**, including manual annotations for **826 MRI scans (13,633 slices)** and **276 CT scans (3,832 slices)**, via the project repository reported in the paper [2510.05555]. That release reinforces the framework’s emphasis on annotation efficiency, cross-dataset reproducibility, and multi-modality evaluation under a supervision budget of only **one labeled slice per dataset**.

Source: https://www.emergentmind.com/topics/no-new-sam2-nnsam2