Fine-tuning Segment Anything for Real-Time Tumor Tracking in Cine-MRI
Valentin Boussot, C\'edric H\'emon, Jean-Claude Nunes, Jean-Louis Dillenseger

TL;DR
This paper presents a real-time tumor tracking method in cine-MRI using fine-tuned foundation models, achieving high accuracy under data scarcity and strict time constraints, with potential for MRI-guided radiotherapy.
Contribution
The work demonstrates the effective fine-tuning of SAM 2.1 for real-time tumor segmentation in cine-MRI with minimal data and strict runtime limits.
Findings
Achieved a Dice score of 0.8794 on the test set.
Selected SAM-based segmentation for its real-time performance.
Ranked 6th in the TrackRAD2025 challenge.
Abstract
In this work, we address the TrackRAD2025 challenge of real-time tumor tracking in cine-MRI sequences of the thoracic and abdominal regions under strong data scarcity constraints. Two complementary strategies were explored: (i) unsupervised registration with the IMPACT similarity metric and (ii) foundation model-based segmentation leveraging SAM 2.1 and its recent variants through prompt-based interaction. Due to the one-second runtime constraint, the SAM-based method was ultimately selected. The final configuration used SAM2.1 b+ with mask-based prompts from the first annotated slice, fine-tuned solely on the small labeled subset from TrackRAD2025. Training was configured to minimize overfitting, using 1024x1024 patches (batch size 1), standard augmentations, and a balanced Dice + IoU loss. A low uniform learning rate (0.0001) was applied to all modules (prompt encoder, decoder, Hiera…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
