Sliding Window FastEdit: A Framework for Lesion Annotation in Whole-body PET Images

November 24, 2023 · Entered Twilight · 🏛 IEEE International Symposium on Biomedical Imaging

Repo contents: .dockerignore, .gitattributes, .gitignore, .pre-commit-config.yaml, Dockerfile, LICENSE, README.md, build.sh, export.sh, guidance.md, ipynb, monailabel, pyproject.toml, requirements.txt, requirements_docker.txt, scripts, setup.cfg, src, test.sh, test_autopet.sh, test_autopet_ensemble.sh

Authors Matthias Hadlich, Zdravko Marinov, Moon Kim, Enrico Nasca, Jens Kleesiek, Rainer Stiefelhagen arXiv ID 2311.14482 Category eess.IV: Image & Video Processing Cross-listed cs.AI, cs.CV, cs.HC Citations 4 Venue IEEE International Symposium on Biomedical Imaging Repository https://github.com/matt3o/AutoPET2-Submission/ ⭐ 10 Last Checked 3 months ago

Abstract

Deep learning has revolutionized the accurate segmentation of diseases in medical imaging. However, achieving such results requires training with numerous manual voxel annotations. This requirement presents a challenge for whole-body Positron Emission Tomography (PET) imaging, where lesions are scattered throughout the body. To tackle this problem, we introduce SW-FastEdit - an interactive segmentation framework that accelerates the labeling by utilizing only a few user clicks instead of voxelwise annotations. While prior interactive models crop or resize PET volumes due to memory constraints, we use the complete volume with our sliding window-based interactive scheme. Our model outperforms existing non-sliding window interactive models on the AutoPET dataset and generalizes to the previously unseen HECKTOR dataset. A user study revealed that annotators achieve high-quality predictions with only 10 click iterations and a low perceived NASA-TLX workload. Our framework is implemented using MONAI Label and is available: https://github.com/matt3o/AutoPET2-Submission/