# Automatic identification of segmentation errors for radiotherapy using geometric learning

arXiv: https://arxiv.org/abs/2206.13317 · HTML: https://mlsanity.com/paper/2206.13317/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 24 references that are in the map.
- First posted on arXiv 27 Jun 2022.

## Builds on (most cited first, 10 of 10)

- [How Powerful are Graph Neural Networks?](https://mlsanity.com/paper/1810.00826/) (2018)
- [Fast Graph Representation Learning with PyTorch Geometric](https://mlsanity.com/paper/1903.02428/) (2019)
- [Unsupervised Learning of Visual Representations by Solving Jigsaw Puzzles](https://mlsanity.com/paper/1603.09246/) (2016)
- [Learning Mesh-Based Simulation with Graph Networks](https://mlsanity.com/paper/2010.03409/) (2020)
- [Open3D: A Modern Library for 3D Data Processing](https://mlsanity.com/paper/1801.09847/) (2018)
- [Confidence Calibration and Predictive Uncertainty Estimation for Deep Medical Image Segmentation](https://mlsanity.com/paper/1911.13273/) (2019)
- [SplineCNN: Fast Geometric Deep Learning with Continuous B-Spline Kernels](https://mlsanity.com/paper/1711.08920/) (2017)
- [Deep learning to achieve clinically applicable segmentation of head and neck anatomy for radiotherapy](https://mlsanity.com/paper/1809.04430/) (2018)
- [Reverse Classification Accuracy: Predicting Segmentation Performance in the Absence of Ground Truth](https://mlsanity.com/paper/1702.03407/) (2017)
- [Automatic segmentation with detection of local segmentation failures in cardiac MRI](https://mlsanity.com/paper/2011.07025/) (2020)

## Built on by (most cited first, 3 of 3)

- [S3M: Scalable Statistical Shape Modeling through Unsupervised Correspondences](https://mlsanity.com/paper/2304.07515/) (2023)
- [Unsupervised correspondence with combined geometric learning and imaging for radiotherapy applications](https://mlsanity.com/paper/2309.14269/) (2023)
- [Geometric Learning-Based Transformer Network for Estimation of Segmentation Errors](https://mlsanity.com/paper/2308.05068/) (2023)

## Abstract (arXiv)

Automatic segmentation of organs-at-risk (OARs) in CT scans using convolutional neural networks (CNNs) is being introduced into the radiotherapy workflow. However, these segmentations still require manual editing and approval by clinicians prior to clinical use, which can be time consuming. The aim of this work was to develop a tool to automatically identify errors in 3D OAR segmentations without a ground truth. Our tool uses a novel architecture combining a CNN and graph neural network (GNN) to leverage the segmentation's appearance and shape. The proposed model is trained using self-supervised learning using a synthetically-generated dataset of segmentations of the parotid and with realistic contouring errors. The effectiveness of our model is assessed with ablation tests, evaluating the efficacy of different portions of the architecture as well as the use of transfer learning from an unsupervised pretext task. Our best performing model predicted errors on the parotid gland with a precision of 85.0% & 89.7% for internal and external errors respectively, and recall of 66.5% & 68.6%. This offline QA tool could be used in the clinical pathway, potentially decreasing the time clinicians spend correcting contours by detecting regions which require their attention. All our code is publicly available at https://github.com/rrr-uom-projects/contour_auto_QATool.