# Less is More: Sample Selection and Label Conditioning Improve Skin Lesion Segmentation

arXiv: https://arxiv.org/abs/2004.13856 · HTML: https://mlsanity.com/paper/2004.13856/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Apr 2020.

## Built on by (most cited first, 8 of 8)

- [A Survey on Deep Learning for Skin Lesion Segmentation](https://mlsanity.com/paper/2206.00356/) (2022)
- [Towards Single Stage Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2106.10309/) (2021)
- [Segmentation Style Discovery: Application to Skin Lesion Images](https://mlsanity.com/paper/2408.02787/) (2024)
- [What Can We Learn from Inter-Annotator Variability in Skin Lesion Segmentation?](https://mlsanity.com/paper/2508.09381/) (2025)
- [Test-Time Selection for Robust Skin Lesion Analysis](https://mlsanity.com/paper/2308.05595/) (2023)
- [MSE-Nets: Multi-annotated Semi-supervised Ensemble Networks for Improving Segmentation of Medical Image with Ambiguous Boundaries](https://mlsanity.com/paper/2311.10380/) (2023)
- [Exploring Intrinsic Properties of Medical Images for Self-Supervised Binary Semantic Segmentation](https://mlsanity.com/paper/2402.02367/) (2024)
- [Federated Medical Image Segmentation under Real-World Label Noise: A Benchmark Suite for Noisy Label Learning Method Selection](https://mlsanity.com/paper/2606.16868/) (2026)

## Abstract (arXiv)

Segmenting skin lesions images is relevant both for itself and for assisting in lesion classification, but suffers from the challenge in obtaining annotated data. In this work, we show that segmentation may improve with less data, by selecting the training samples with best inter-annotator agreement, and conditioning the ground-truth masks to remove excessive detail. We perform an exhaustive experimental design considering several sources of variation, including three different test sets, two different deep-learning architectures, and several replications, for a total of 540 experimental runs. We found that sample selection and detail removal may have impacts corresponding, respectively, to 12% and 16% of the one obtained by picking a better deep-learning model.