Summary of Autopet Challenge Iii: Testing the Robustness Of Generalized Dice Focal Loss Trained 3d Residual Unet For Fdg and Psma Lesion Segmentation From Whole-body Pet/ct Images, by Shadab Ahamed
AutoPET Challenge III: Testing the Robustness of Generalized Dice Focal Loss trained 3D Residual UNet for FDG and PSMA Lesion Segmentation from Whole-Body PET/CT Images
by Shadab Ahamed
First submitted to arxiv on: 16 Sep 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Medical Physics (physics.med-ph)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper presents a deep learning approach to segmenting cancerous lesions in PET/CT scans. The task is challenging due to variations in lesion size, shape, and radiotracer uptake, as well as the proximity of healthy organs that can exhibit similar uptake patterns. To overcome these challenges, the authors utilize a 3D Residual UNet model with Generalized Dice Focal Loss function and employ a 5-fold cross-validation and average ensembling technique. The model is trained on the AutoPET Challenge 2024 dataset and achieves a mean Dice Similarity Coefficient (DSC) of 0.6687, mean false negative volume (FNV) of 10.9522 ml, and mean false positive volume (FPV) of 2.9684 ml in the preliminary test phase for Task-1. The training code is shared via GitHub repository. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps doctors use special computers to better identify cancerous lesions on medical images. It’s hard because the lesions can look different and be in different places, making it tough for machines to figure out what’s normal and what’s not. To solve this problem, researchers created a special model that uses a combination of computer vision techniques and math to analyze the images. They tested the model on a big dataset and got good results, which could help doctors make more accurate diagnoses. |
Keywords
» Artificial intelligence » Deep learning » Loss function » Unet
