Skip to content

README.md

Latest commit

 

History

History
24 lines (16 loc) · 963 Bytes

README.md

File metadata and controls

24 lines (16 loc) · 963 Bytes
Raw

ISUS: Interactive Segmentation Utilities for Spheres

Official preview for the MICCAI 2026 paper Shape-Adaptive Guidance Signal for Interactive Cortical Sulcal Labeling in the ISUS repository.

We present a spherical CNN-based interactive segmentation framwork that leverages a shape-adaptive guidance signal designed to label cortical sulci. The user interactions are encoded on spherical surfaces by solving eikonal equation with a curvature-based speed function.

Overview

Description

This repository will provide functions that can be used in training and evaluating interactive segmentation networks on spherical surfaces.

User Interaction (Click) Sampler

  • Rule-based Sampler
  • Weighted Random Sampler
  • Initial Seeds Sampler

Guidance Signal Encoder

  • Angular Distance Transform (ADT)
  • Disk
  • Weighted Geodesic Distance Transform (WGDT)

Status

Code release coming soon.