SAM-OCTA2: Layer Sequence OCTA Segmentation with Fine-tuned Segment Anything Model 2

TL;DR

SAM-OCTA2 fine-tunes a segmentation model for 3D OCTA data, enabling precise layer-wise vessel tracking and FAZ segmentation, surpassing previous methods in accuracy.

Contribution

This work introduces SAM-OCTA2, a novel approach that adapts the Segment Anything Model for 3D OCTA segmentation using low-rank adaptation and new prompt strategies.

Findings

Achieves state-of-the-art FAZ segmentation accuracy.

Effectively tracks vessels across OCTA layers.

Demonstrates superior performance on OCTA-500 dataset.

Abstract

Segmentation of indicated targets aids in the precise analysis of optical coherence tomography angiography (OCTA) samples. Existing segmentation methods typically perform on 2D projection targets, making it challenging to capture the variance of segmented objects through the 3D volume. To address this limitation, the low-rank adaptation technique is adopted to fine-tune the Segment Anything Model (SAM) version 2, enabling the tracking and segmentation of specified objects across the OCTA scanning layer sequence. To further this work, a prompt point generation strategy in frame sequence and a sparse annotation method to acquire retinal vessel (RV) layer masks are proposed. This method is named SAM-OCTA2 and has been experimented on the OCTA-500 dataset. It achieves state-of-the-art performance in segmenting the foveal avascular zone (FAZ) on regular 2D en-face and effectively tracks local vessels across scanning layer sequences. The code is available at: https://github.com/ShellRedia/SAM-OCTA2.