Automated Lesion Segmentation in Whole-Body PET/CT in a multitracer setting
This work addresses lesion segmentation for diagnostic workflows in medical imaging, but it appears incremental as it adapts existing methods to a multitracer setting.
The study tackled automated lesion segmentation in whole-body PET/CT images for FDG and PSMA tracers by developing a workflow with specialized preprocessing and YOLOv8 classification, aiming to improve accuracy, but no concrete performance numbers were provided in the abstract.
This study explores a workflow for automated segmentation of lesions in FDG and PSMA PET/CT images. Due to the substantial differences in image characteristics between FDG and PSMA, specialized preprocessing steps are required. Utilizing YOLOv8 for data classification, the FDG and PSMA images are preprocessed separately before feeding them into the segmentation models, aiming to improve lesion segmentation accuracy. The study focuses on evaluating the performance of automated segmentation workflow for multitracer PET images. The findings are expected to provide critical insights for enhancing diagnostic workflows and patient-specific treatment plans. Our code will be open-sourced and available at https://github.com/jiayiliu-pku/AP2024.