Hands-Free Segmentation of Medical Volumes via Binary Inputs

TL;DR

This paper introduces a hands-free, interactive segmentation method for 3D medical images that uses binary questions to efficiently identify structures, leveraging probabilistic sampling and shape priors.

Contribution

The novel approach combines binary questioning with probabilistic sampling and shape priors for efficient, interactive 3D medical volume segmentation.

Findings

Outperforms baseline methods in prostate MRI segmentation

Effective in synthetic failure case analysis

Demonstrates adaptation potential in various scenarios

Abstract

We propose a novel hands-free method to interactively segment 3D medical volumes. In our scenario, a human user progressively segments an organ by answering a series of questions of the form "Is this voxel inside the object to segment?". At each iteration, the chosen question is defined as the one halving a set of candidate segmentations given the answered questions. For a quick and efficient exploration, these segmentations are sampled according to the Metropolis-Hastings algorithm. Our sampling technique relies on a combination of relaxed shape prior, learnt probability map and consistency with previous answers. We demonstrate the potential of our strategy on a prostate segmentation MRI dataset. Through the study of failure cases with synthetic examples, we demonstrate the adaptation potential of our method. We also show that our method outperforms two intuitive baselines: one based on random questions, the other one being the thresholded probability map.