# Skin Lesion Segmentation on ISIC 2016 (test) Source: [Skin Lesion Segmentation on ISIC 2016 (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:10:08.194Z Evaluation of skin lesion segmentation performance on the test set of the ISIC 2016 dataset, reporting metrics such as Dice (mDice) and Intersection over Union (mIoU). 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 | Skin Lesion Segmentation | | Dataset (including any stated split) | ISIC 2016 (test) | | Metric | mDice | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | 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 - [mDice](): Higher is better; unit: PERCENT. - [mIoU](): Higher is better; unit: PERCENT. ## Recorded results | Method | mDice | Result date | Report paper | | --- | --- | --- | --- | | Siamese-Diffusion | 94.31 | Not recorded | 68fa78821e209131cea78c41 | | Siamese-Diffusion | 94.14 | Not recorded | 68fa78821e209131cea78c41 | | T2I-Adapter | 93.71 | Not recorded | 68fa78821e209131cea78c41 | | Real Dataset | 93.57 | Not recorded | 68fa78821e209131cea78c41 | | ControlNet | 93.54 | Not recorded | 68fa78821e209131cea78c41 | | Copy-Paste | 93.53 | Not recorded | 68fa78821e209131cea78c41 | | Siamese-Diffusion | 90.06 | Not recorded | 68fa78821e209131cea78c41 | | Siamese-Diffusion | 89.91 | Not recorded | 68fa78821e209131cea78c41 | | Real Dataset | 89.58 | Not recorded | 68fa78821e209131cea78c41 | | T2I-Adapter | 89.48 | Not recorded | 68fa78821e209131cea78c41 | | Copy-Paste | 89.43 | Not recorded | 68fa78821e209131cea78c41 | | ControlNet | 89.42 | Not recorded | 68fa78821e209131cea78c41 | ### Siamese-Diffusion mDice: 94.31. Result ID: 69958ed83badfaffe688cbf9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Architecture":"SegFormer","Training Set":"Real + Synthetic","Mask Prior":"Random"}. This record does not assert independent reproduction. ### Siamese-Diffusion mDice: 94.14. Result ID: 69958ed83badfaffe688cbf7. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Architecture":"SegFormer","Training Set":"Real + Synthetic","Mask Prior":"Real"}. This record does not assert independent reproduction. ### T2I-Adapter mDice: 93.71. Result ID: 69958ed83badfaffe688cbf3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Architecture":"SegFormer","Training Set":"Real + Synthetic"}. This record does not assert independent reproduction. ### Real Dataset mDice: 93.57. Result ID: 69958ed83badfaffe688cbef. [Reporting paper]() Recorded evaluation configuration: {"Architecture":"SegFormer","Training Set":"Real Only"}. This record does not assert independent reproduction. ### ControlNet mDice: 93.54. Result ID: 69958ed83badfaffe688cbf5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Architecture":"SegFormer","Training Set":"Real + Synthetic"}. This record does not assert independent reproduction. ### Copy-Paste mDice: 93.53. Result ID: 69958ed83badfaffe688cbf1. [Reporting paper]() Recorded evaluation configuration: {"Architecture":"SegFormer","Training Set":"Real + Synthetic"}. This record does not assert independent reproduction. ### Siamese-Diffusion mDice: 90.06. Result ID: 69958ed83badfaffe688cbf8. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Architecture":"UNet","Training Set":"Real + Synthetic","Mask Prior":"Random"}. This record does not assert independent reproduction. ### Siamese-Diffusion mDice: 89.91. Result ID: 69958ed83badfaffe688cbf6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Architecture":"UNet","Training Set":"Real + Synthetic","Mask Prior":"Real"}. This record does not assert independent reproduction. ### Real Dataset mDice: 89.58. Result ID: 69958ed83badfaffe688cbee. [Reporting paper]() Recorded evaluation configuration: {"Architecture":"UNet","Training Set":"Real Only"}. This record does not assert independent reproduction. ### T2I-Adapter mDice: 89.48. Result ID: 69958ed83badfaffe688cbf2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Architecture":"UNet","Training Set":"Real + Synthetic"}. This record does not assert independent reproduction. ### Copy-Paste mDice: 89.43. Result ID: 69958ed83badfaffe688cbf0. [Reporting paper]() Recorded evaluation configuration: {"Architecture":"UNet","Training Set":"Real + Synthetic"}. This record does not assert independent reproduction. ### ControlNet mDice: 89.42. Result ID: 69958ed83badfaffe688cbf4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Architecture":"UNet","Training Set":"Real + Synthetic"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.