SAMSA 2.0: Prompting Segment Anything with Spectral Angles for Hyperspectral Interactive Medical Image Segmentation

TL;DR

SAMSA 2.0 enhances hyperspectral medical image segmentation by integrating spectral angle prompting with SAM, improving accuracy and robustness without retraining, especially in low-data and noisy conditions.

Contribution

Introduces spectral angle prompting for SAM, enabling spectral-spatial fusion in hyperspectral medical imaging without retraining, improving segmentation accuracy.

Findings

Up to +3.8% higher Dice scores over RGB-only models

Improved few-shot and zero-shot segmentation performance

Enhanced robustness in low-data and noisy scenarios

Abstract

We present SAMSA 2.0, an interactive segmentation framework for hyperspectral medical imaging that introduces spectral angle prompting to guide the Segment Anything Model (SAM) using spectral similarity alongside spatial cues. This early fusion of spectral information enables more accurate and robust segmentation across diverse spectral datasets. Without retraining, SAMSA 2.0 achieves up to +3.8% higher Dice scores compared to RGB-only models and up to +3.1% over prior spectral fusion methods. Our approach enhances few-shot and zero-shot performance, demonstrating strong generalization in challenging low-data and noisy scenarios common in clinical imaging.