AutoPET Challenge 2023: Sliding Window-based Optimization of U-Net
This work addresses tumor segmentation for cancer staging in medical imaging, but it is incremental as it optimizes an existing U-Net architecture.
The paper tackled the problem of tumor segmentation in FDG-PET/CT scans, which is challenging due to misinterpretations of glucose uptake, and resulted in a method that achieved competitive performance on the AutoPET challenge dataset of 1014 studies.
Tumor segmentation in medical imaging is crucial and relies on precise delineation. Fluorodeoxyglucose Positron-Emission Tomography (FDG-PET) is widely used in clinical practice to detect metabolically active tumors. However, FDG-PET scans may misinterpret irregular glucose consumption in healthy or benign tissues as cancer. Combining PET with Computed Tomography (CT) can enhance tumor segmentation by integrating metabolic and anatomic information. FDG-PET/CT scans are pivotal for cancer staging and reassessment, utilizing radiolabeled fluorodeoxyglucose to highlight metabolically active regions. Accurately distinguishing tumor-specific uptake from physiological uptake in normal tissues is a challenging aspect of precise tumor segmentation. The AutoPET challenge addresses this by providing a dataset of 1014 FDG-PET/CT studies, encouraging advancements in accurate tumor segmentation and analysis within the FDG-PET/CT domain. Code: https://github.com/matt3o/AutoPET2-Submission/