# AutoPET Challenge 2022: Step-by-Step Lesion Segmentation in Whole-body FDG-PET/CT

arXiv: https://arxiv.org/abs/2209.09199 · HTML: https://mlsanity.com/paper/2209.09199/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 3 of its 3 references that are in the map.
- First posted on arXiv 4 Sep 2022.

## Builds on (most cited first, 3 of 3)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [Auxiliary Tasks in Multi-task Learning](https://mlsanity.com/paper/1805.06334/) (2018)

## Built on by (most cited first, 2 of 2)

- [Mirror U-Net: Marrying Multimodal Fission with Multi-task Learning for Semantic Segmentation in Medical Imaging](https://mlsanity.com/paper/2303.07126/) (2023)
- [AutoPET Challenge 2023: Sliding Window-based Optimization of U-Net](https://mlsanity.com/paper/2309.12114/) (2023)

## Abstract (arXiv)

Automatic segmentation of tumor lesions is a critical initial processing step for quantitative PET/CT analysis. However, numerous tumor lesions with different shapes, sizes, and uptake intensity may be distributed in different anatomical contexts throughout the body, and there is also significant uptake in healthy organs. Therefore, building a systemic PET/CT tumor lesion segmentation model is a challenging task. In this paper, we propose a novel step-by-step 3D segmentation method to address this problem. We achieved Dice score of 0.92, false positive volume of 0.89 and false negative volume of 0.53 on preliminary test set.The code of our work is available on the following link: https://github.com/rightl/autopet.