Beyond Natural Images: A Dual-Stream DINOv3 Framework for PET/CT Segmentation

Yu-Nong Lin, Shansong Wang, Mojtaba Safari, Xiaofeng Yang

Keywords: Self-supervised learning, DINO, PET/CT, Foundation Model, Segmentation

TL;DR: Dual-DINO introduces the first PET-specific DINOv3 encoder and a dual-stream PET/CT framework that achieves superior tumor segmentation through separate, domain-aligned pretraining of each modality.

Abstract:

Self-supervised vision transformers like DINOv3 are strong universal feature extractors, yet their transferability to functional medical imaging remains limited when pretrained on misaligned natural-image domains. In this work, we introduce Dual-DINO, a dual-stream framework for PET/CT that addresses two key gaps in existing work: the absence of a public, PET-specific pretrained encoder and the reliance on fully paired PET/CT data for multimodal pretraining. First, we presented the first PET-specific DINOv3 encoder, pretrained exclusively on large-scale public FDG-PET datasets using the full three-stage DINOv3 self-distillation pipeline. Second, we proposed a modality-separated PET/CT framework in which PET- and CT-specific encoders are pretrained independently and fused during finetuning via multiscale cross-attention, enabling multimodal representation learning without requiring paired data during pretraining.

Evaluation on the HECKTOR tumor segmentation benchmark demonstrated three central findings: (1) misaligned natural-image pretraining degrades PET/CT performance relative to training from scratch, (2) domain-aligned CT pretraining substantially improves segmentation across all tumor sizes, and (3) dual-stream PET/CT pretraining achieves the best performance overall, highlighting the complementary contributions of functional and anatomical cues. Together, these results provide a fully public PET encoder and a scalable PET/CT foundation model that support domain-aligned representation learning under realistic clinical data constraints.

Primary Subject Area: Foundation Models

Secondary Subject Area: Segmentation

Registration Requirement: Yes

Visa & Travel: No

Read CFP & Author Instructions: Yes

Originality Policy: Yes

Single-blind & Not Under Review Elsewhere: Yes

LLM Policy: Yes

Midl Latex Submission Checklist: Ensure no LaTeX errors during compilation., Replace NNN with your OpenReview submission ID., Includes \documentclass{midl}, \jmlryear{2026}, \jmlrworkshop, \jmlrvolume, \editors, and correct \bibliography command., Did not override options of the hyperref package., Did not use the times package., Use the correct spelling and format, avoid Unicode characters, and use LaTeX equivalents instead., Any math in the title and abstract must be enclosed within $...$., Did not override the bibliography style defined in midl.cls and did not use \begin{thebibliography} directly to insert references., Avoid using \scalebox; use \resizebox when needed., Included all necessary figures and removed *unused* files in the zip archive., Removed special formatting, visual annotations, and highlights used during rebuttal., All special characters in the paper and .bib file use LaTeX commands (e.g., \'e for é)., No separate supplementary PDF uploads., Acknowledgements, references, and appendix must start after the main content.

Submission Number: 240