SLIP: Segmentation with Low-latency Interactive Prompting for 3D Medical Images

2026-07-24

AI Summary

This work addresses critical limitations in existing interactive 3D medical image segmentation methods—namely high latency, poor responsiveness to continuous interactions, lack of reversible prompting, and absence of real-user validation. To overcome these challenges, the authors propose SLIP, the first end-to-end trainable interactive 3D segmentation framework. SLIP employs an image-prompt decoupling architecture, a lightweight block memory bank, and an interaction context propagation mechanism, enabling low-latency, reversible, and cross-modality generalizable interactions after a single forward pass of image encoding. Evaluated across 13 public datasets, SLIP achieves state-of-the-art performance and demonstrates significant advantages in response speed, interaction fluency, and user preference, as validated through a prospective user study involving six clinical experts.

Abstract

Interactive deep image segmentation enables efficient medical image annotation by iteratively refining predictions from user prompts, such as positive and negative clicks. Recent patch-based methods, including nnInteractive, achieve strong segmentation performance but remain limited in annotation workflows by high interaction latency, limited responsiveness to successive interactions, and the lack of support for reversible prompting. Furthermore, evaluation relies predominantly on simulated rather than controlled real-user interaction studies. We present SLIP, an end-to-end trainable framework for interactive 3D medical image segmentation that decouples image encoding from prompt-guided refinement. Image features are computed once and reused, while a lightweight patch memory bank maintains an interaction-aware segmentation state shared across patches. This representation enables prediction updates by propagating interaction context throughout the image, supports reversible prompting without recomputing image features, and substantially reduces interaction latency. By separating image representation from interactive reasoning, SLIP remains compatible with a wide range of image encoders. We train a single SLIP model for general interactive segmentation across diverse anatomical structures and imaging modalities. Beyond standard simulated evaluation, we conduct a controlled prospective user study comparing manual segmentation, nnInteractive, and SLIP across three clinical annotation tasks, six expert participants, and subjective usability measures, addressing the limited human validation of interactive segmentation methods. SLIP achieves SOTA interactive segmentation performance across 13 public datasets while providing lower interaction latency, greater responsiveness, support for reversible prompting, and higher user preference than existing approaches.

Problem

• interactive segmentation
• 3D medical images
• interaction latency
• reversible prompting
• user study

Innovation

• interactive segmentation
• low-latency prompting
• reversible prompting
• patch memory bank
• 3D medical imaging