# Automated Identification of Failure Cases in Organ at Risk Segmentation Using Distance Metrics: A Study on CT Data

arXiv: https://arxiv.org/abs/2308.10636 · HTML: https://mlsanity.com/paper/2308.10636/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 22 references that are in the map.
- First posted on arXiv 21 Aug 2023.

## Builds on (most cited first, 9 of 9)

- [A survey of loss functions for semantic segmentation](https://mlsanity.com/paper/2006.14822/) (2020)
- [Reducing the Hausdorff Distance in Medical Image Segmentation with Convolutional Neural Networks](https://mlsanity.com/paper/1904.10030/) (2019)
- [A New 2.5D Representation for Lymph Node Detection using Random Sets of Deep Convolutional Neural Network Observations](https://mlsanity.com/paper/1406.2639/) (2014)
- [Reconstructing Speech from Real-Time Articulatory MRI Using Neural Vocoders](https://mlsanity.com/paper/2104.11598/) (2021)
- [2D View Aggregation for Lymph Node Detection Using a Shallow Hierarchy of Linear Classifiers](https://mlsanity.com/paper/1408.3337/) (2014)
- [3D Convolutional Neural Networks for Ultrasound-Based Silent Speech Interfaces](https://mlsanity.com/paper/2104.11532/) (2021)
- [Neural Speaker Embeddings for Ultrasound-based Silent Speech Interfaces](https://mlsanity.com/paper/2106.04552/) (2021)
- [Improving Neural Silent Speech Interface Models by Adversarial Training](https://mlsanity.com/paper/2104.11601/) (2021)
- [Voice Activity Detection for Ultrasound-based Silent Speech Interfaces using Convolutional Neural Networks](https://mlsanity.com/paper/2105.13718/) (2021)

## Abstract (arXiv)

Automated organ at risk (OAR) segmentation is crucial for radiation therapy planning in CT scans, but the generated contours by automated models can be inaccurate, potentially leading to treatment planning issues. The reasons for these inaccuracies could be varied, such as unclear organ boundaries or inaccurate ground truth due to annotation errors. To improve the model's performance, it is necessary to identify these failure cases during the training process and to correct them with some potential post-processing techniques. However, this process can be time-consuming, as traditionally it requires manual inspection of the predicted output. This paper proposes a method to automatically identify failure cases by setting a threshold for the combination of Dice and Hausdorff distances. This approach reduces the time-consuming task of visually inspecting predicted outputs, allowing for faster identification of failure case candidates. The method was evaluated on 20 cases of six different organs in CT images from clinical expert curated datasets. By setting the thresholds for the Dice and Hausdorff distances, the study was able to differentiate between various states of failure cases and evaluate over 12 cases visually. This thresholding approach could be extended to other organs, leading to faster identification of failure cases and thereby improving the quality of radiation therapy planning.