# Full heart segmentation on TotalSegmentator (TS) (test)

Source: [Full heart segmentation on TotalSegmentator (TS) (test)](<https://www.sota2.com/research/sota/full-heart-segmentation-on-totalsegmentator-ts-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-12T16:25:14.305Z

Evaluation of full heart segmentation performance on the test split of the TotalSegmentator dataset, reporting Dice Similarity Coefficient (Dice) and Normalized Surface Distance (NSD) for specific cardiac chambers (Myo, LV, LA, RA, RV).

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Full heart segmentation |
| Dataset (including any stated split) | TotalSegmentator (TS) (test) |
| Metric | Dice (Myo) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Dice (Myo)](<https://www.sota2.com/research/sota/full-heart-segmentation-on-totalsegmentator-ts-test.md?metric=Dice%20(Myo)>): Higher is better; unit: PERCENT.

- [NSD (Myo)](<https://www.sota2.com/research/sota/full-heart-segmentation-on-totalsegmentator-ts-test.md?metric=NSD%20(Myo)>): Lower is better; unit: PERCENT.

- [Dice (LV)](<https://www.sota2.com/research/sota/full-heart-segmentation-on-totalsegmentator-ts-test.md?metric=Dice%20(LV)>): Higher is better; unit: PERCENT.

- [NSD (LV)](<https://www.sota2.com/research/sota/full-heart-segmentation-on-totalsegmentator-ts-test.md?metric=NSD%20(LV)>): Lower is better; unit: PERCENT.

- [Dice (LA)](<https://www.sota2.com/research/sota/full-heart-segmentation-on-totalsegmentator-ts-test.md?metric=Dice%20(LA)>): Higher is better; unit: PERCENT.

- [NSD (LA)](<https://www.sota2.com/research/sota/full-heart-segmentation-on-totalsegmentator-ts-test.md?metric=NSD%20(LA)>): Lower is better; unit: PERCENT.

- [Dice (RA)](<https://www.sota2.com/research/sota/full-heart-segmentation-on-totalsegmentator-ts-test.md?metric=Dice%20(RA)>): Higher is better; unit: PERCENT.

- [NSD (RA)](<https://www.sota2.com/research/sota/full-heart-segmentation-on-totalsegmentator-ts-test.md?metric=NSD%20(RA)>): Lower is better; unit: PERCENT.

- [Dice (RV)](<https://www.sota2.com/research/sota/full-heart-segmentation-on-totalsegmentator-ts-test.md?metric=Dice%20(RV)>): Higher is better; unit: PERCENT.

- [NSD (RV)](<https://www.sota2.com/research/sota/full-heart-segmentation-on-totalsegmentator-ts-test.md?metric=NSD%20(RV)>): Lower is better; unit: PERCENT.

- [Average Dice](<https://www.sota2.com/research/sota/full-heart-segmentation-on-totalsegmentator-ts-test.md?metric=Average%20Dice>): Higher is better; unit: PERCENT.

- [Average NSD](<https://www.sota2.com/research/sota/full-heart-segmentation-on-totalsegmentator-ts-test.md?metric=Average%20NSD>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Dice (Myo) | Result date | Report paper |
| --- | --- | --- | --- |
| GenMed-Full | 83.8 | Not recorded | 6a02b4a9dbc836b87d824ead |
| DiffAtlas | 83.52 | Not recorded | 68fa77551e209131cea72725 |
| GenMed-Naive | 83.52 | Not recorded | 6a02b4a9dbc836b87d824ead |
| SwinUNETR | 80.36 | Not recorded | 6a02b4a9dbc836b87d824ead |
| ICF | 80.22 | Not recorded | 68fa77551e209131cea72725 |
| nnU-Net | 80.22 | Not recorded | 6a02b4a9dbc836b87d824ead |
| ICMD | 77.8 | Not recorded | 68fa77551e209131cea72725 |
| Input-Cond Diffusion | 77.8 | Not recorded | 6a02b4a9dbc836b87d824ead |

### GenMed-Full

Dice (Myo): 83.8. Result ID: 6a03545247f34ce924e8c992.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b4a9dbc836b87d824ead.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b4a9dbc836b87d824ead.md>)

Recorded evaluation configuration: {"category":"our method"}. This record does not assert independent reproduction.

### DiffAtlas

Dice (Myo): 83.52. Result ID: 699480af3a6c3148491da76a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77551e209131cea72725.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77551e209131cea72725.md>)

[Implementation](<https://github.com/M3DV/DiffAtlas>)

Recorded evaluation configuration: {"Training Setting":"Full"}. This record does not assert independent reproduction.

### GenMed-Naive

Dice (Myo): 83.52. Result ID: 6a03545247f34ce924e8c991.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b4a9dbc836b87d824ead.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b4a9dbc836b87d824ead.md>)

Recorded evaluation configuration: {"category":"our method"}. This record does not assert independent reproduction.

### SwinUNETR

Dice (Myo): 80.36. Result ID: 6a03545247f34ce924e8c98f.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b4a9dbc836b87d824ead.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b831e209131cea2eb83.md>)

Recorded evaluation configuration: {"category":"discriminative method"}. This record does not assert independent reproduction.

### ICF

Dice (Myo): 80.22. Result ID: 699480af3a6c3148491da768.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77551e209131cea72725.md>)

Recorded evaluation configuration: {"Training Setting":"Full"}. This record does not assert independent reproduction.

### nnU-Net

Dice (Myo): 80.22. Result ID: 6a03545247f34ce924e8c98e.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b4a9dbc836b87d824ead.md>)

Recorded evaluation configuration: {"category":"discriminative method"}. This record does not assert independent reproduction.

### ICMD

Dice (Myo): 77.8. Result ID: 699480af3a6c3148491da769.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77551e209131cea72725.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71aa1e209131cea52693.md>)

[Implementation](<https://github.com/KidsWithTokens/MedSegDiff>)

Recorded evaluation configuration: {"Training Setting":"Full"}. This record does not assert independent reproduction.

### Input-Cond Diffusion

Dice (Myo): 77.8. Result ID: 6a03545247f34ce924e8c990.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b4a9dbc836b87d824ead.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71aa1e209131cea52693.md>)

[Implementation](<https://github.com/KidsWithTokens/MedSegDiff>)

Recorded evaluation configuration: {"category":"input-conditioned generative method"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
