PulseTrendingJournal ClubResearchersJournalsExplore
Instagram
HomeTrendingJournal ClubExplore
Synapse
⌘+K
Synapse
May 18, 2026The EANM JournalOpen Access

Proof of Concept of Anatomical Independence in Total-Body PET Using AI-Based CT-Free Multi-Organ Segmentation

View Full Paper
Ask AI
Bookmark
Share

Authors

SXSong XueCCChristoph ClémentYMYing Miao

Discussion

Loading...

Member takes

Overview

Retrospective study investigates AI-based PET-only segmentation for organs, suggesting enhanced imaging methods.

Key Points

  • This study aims to evaluate the feasibility of segmenting multiple organs using PET imaging alone, powered by AI techniques.
  • Retrospective study utilizing non-corrected PET images from two PET/CT scanners.
  • A U-Net-like deep learning model was trained on a dataset of 938 PET scans.
  • Validation performed on internal (68) and external (382) datasets across various scanners and tracers.
  • Mean Dice score of 0.828 (95% CI, 0.816–0.839) for internal tests using Biograph Vision Quadra and uExplorer scanners.
  • Achieved a mean Dice score of 0.787 (95% CI, 0.782–0.791) on external datasets across five cross-scanner cohorts.
  • Mean Dice score of 0.726 (95% CI, 0.698–0.755) across four cross-tracer cohorts.

Cite This Study

Xue et al. (2026) studied this question.

synapsesocial.com/papers/6a0aac2b5ba8ef6d83b6fc15https://doi.org/10.1016/j.eanmj.2026.100221
View Full Paper
Ask AI
Bookmark
Share