# The impact of using voxel-level segmentation metrics on evaluating multifocal prostate cancer localisation

arXiv: https://arxiv.org/abs/2203.16415 · HTML: https://mlsanity.com/paper/2203.16415/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 2 of its 25 references that are in the map.
- First posted on arXiv 30 Mar 2022.

## Builds on (most cited first, 2 of 2)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Harnessing Uncertainty in Domain Adaptation for MRI Prostate Lesion Segmentation](https://mlsanity.com/paper/2010.07411/) (2020)

## Built on by (most cited first, 6 of 6)

- [Prototypical few-shot segmentation for cross-institution male pelvic structures with spatial registration](https://mlsanity.com/paper/2209.05160/) (2022)
- [Combiner and HyperCombiner Networks: Rules to Combine Multimodality MR Images for Prostate Cancer Localisation](https://mlsanity.com/paper/2307.08279/) (2023)
- [Anomaly Detection and Generation with Diffusion Models: A Survey](https://mlsanity.com/paper/2506.09368/) (2025)
- [Weakly supervised localisation of prostate cancer using reinforcement learning for bi-parametric MR images](https://mlsanity.com/paper/2402.13778/) (2024)
- [Promptable cancer segmentation using minimal expert-curated data](https://mlsanity.com/paper/2505.17915/) (2025)
- [Automated Sperm Morphology Analysis Based on Instance-Aware Part Segmentation](https://mlsanity.com/paper/2408.00112/) (2024)

## Abstract (arXiv)

Dice similarity coefficient (DSC) and Hausdorff distance (HD) are widely used for evaluating medical image segmentation. They have also been criticised, when reported alone, for their unclear or even misleading clinical interpretation. DSCs may also differ substantially from HDs, due to boundary smoothness or multiple regions of interest (ROIs) within a subject. More importantly, either metric can also have a nonlinear, non-monotonic relationship with outcomes based on Type 1 and 2 errors, designed for specific clinical decisions that use the resulting segmentation. Whilst cases causing disagreement between these metrics are not difficult to postulate. This work first proposes a new asymmetric detection metric, adapting those used in object detection, for planning prostate cancer procedures. The lesion-level metrics is then compared with the voxel-level DSC and HD, whereas a 3D UNet is used for segmenting lesions from multiparametric MR (mpMR) images. Based on experimental results we report pairwise agreement and correlation 1) between DSC and HD, and 2) between voxel-level DSC and recall-controlled precision at lesion-level, with Cohen's [0.49, 0.61] and Pearson's [0.66, 0.76] (p-values}<0.001) at varying cut-offs. However, the differences in false-positives and false-negatives, between the actual errors and the perceived counterparts if DSC is used, can be as high as 152 and 154, respectively, out of the 357 test set lesions. We therefore carefully conclude that, despite of the significant correlations, voxel-level metrics such as DSC can misrepresent lesion-level detection accuracy for evaluating localisation of multifocal prostate cancer and should be interpreted with caution.