Extending SEEDS to a Supervoxel Algorithm for Medical Image Analysis

TL;DR

This paper introduces 3D SEEDS, an extension of the SEEDS superpixel algorithm to 3D volumes, offering faster processing, improved segmentation accuracy, and open-source availability for medical image analysis.

Contribution

It presents the novel 3D SEEDS algorithm, extending superpixel segmentation to 3D medical images with significant speed and accuracy improvements.

Findings

10x faster supervoxel generation

6.5% higher Dice score

Reduced under-segmentation error

Abstract

In this work, we extend the SEEDS superpixel algorithm from 2D images to 3D volumes, resulting in 3D SEEDS, a faster, better, and open-source supervoxel algorithm for medical image analysis. We compare 3D SEEDS with the widely used supervoxel algorithm SLIC on 13 segmentation tasks across 10 organs. 3D SEEDS accelerates supervoxel generation by a factor of 10, improves the achievable Dice score by +6.5%, and reduces the under-segmentation error by -0.16%. The code is available at https://github.com/Zch0414/3d_seeds