# Imperfect Segmentation Labels: How Much Do They Matter?

arXiv: https://arxiv.org/abs/1806.04618 · HTML: https://mlsanity.com/paper/1806.04618/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Jun 2018.

## Built on by (most cited first, 14 of 14)

- [The KiTS21 Challenge: Automatic segmentation of kidneys, renal tumors, and renal cysts in corticomedullary-phase CT](https://mlsanity.com/paper/2307.01984/) (2023)
- [Medical Image Segmentation with Limited Supervision: A Review of Deep Network Models](https://mlsanity.com/paper/2103.00429/) (2021)
- [Automated Detection of Label Errors in Semantic Segmentation Datasets via Deep Learning and Uncertainty Quantification](https://mlsanity.com/paper/2207.06104/) (2022)
- [Estimating label quality and errors in semantic segmentation data via any model](https://mlsanity.com/paper/2307.05080/) (2023)
- [Rethinking Generalization: The Impact of Annotation Style on Medical Image Segmentation](https://mlsanity.com/paper/2210.17398/) (2022)
- [How to Efficiently Annotate Images for Best-Performing Deep Learning Based Segmentation Models: An Empirical Study with Weak and Noisy Annotations and Segment Anything Model](https://mlsanity.com/paper/2312.10600/) (2023)
- [Cohort Bias Adaptation in Aggregated Datasets for Lesion Segmentation](https://mlsanity.com/paper/2108.00713/) (2021)
- [Adversarial Robustness of Deep Learning Models for Inland Water Body Segmentation from SAR Images](https://mlsanity.com/paper/2505.01884/) (2025)
- [Towards Fairness under Label Bias in Image Segmentation: Impact, Measurement and Mitigation](https://mlsanity.com/paper/2605.06891/) (2026)
- [RootPainter3D: Interactive-machine-learning enables rapid and accurate contouring for radiotherapy](https://mlsanity.com/paper/2106.11942/) (2021)
- [Noisy Image Segmentation With Soft-Dice](https://mlsanity.com/paper/2304.00801/) (2023)
- [Marginal Thresholding in Noisy Image Segmentation](https://mlsanity.com/paper/2304.04116/) (2023)
- [Good Enough? An Investigation on the Impact of Label Quality in Large-Scale Medical Datasets](https://mlsanity.com/paper/2505.20928/) (2025)
- [Learning to Detect Label Errors by Making Them: A Method for Segmentation and Object Detection Datasets](https://mlsanity.com/paper/2508.17930/) (2025)

## Abstract (arXiv)

Labeled datasets for semantic segmentation are imperfect, especially in medical imaging where borders are often subtle or ill-defined. Little work has been done to analyze the effect that label errors have on the performance of segmentation methodologies. Here we present a large-scale study of model performance in the presence of varying types and degrees of error in training data. We trained U-Net, SegNet, and FCN32 several times for liver segmentation with 10 different modes of ground-truth perturbation. Our results show that for each architecture, performance steadily declines with boundary-localized errors, however, U-Net was significantly more robust to jagged boundary errors than the other architectures. We also found that each architecture was very robust to non-boundary-localized errors, suggesting that boundary-localized errors are fundamentally different and more challenging problem than random label errors in a classification setting.