Bayesian Pseudo Labels: Expectation Maximization for Robust and Efficient Semi-Supervised Segmentation

TL;DR

This paper introduces a new EM-based formulation for pseudo-labeling in semi-supervised segmentation, proposing a simple, robust, and computationally efficient method called SegPL, with extensions for uncertainty estimation.

Contribution

It presents a novel EM formulation for pseudo-labeling, a simple semi-supervised segmentation method, and a probabilistic extension for uncertainty estimation.

Findings

SegPL is competitive with state-of-the-art methods on MRI and CT segmentation tasks.

SegPL demonstrates robustness against out-of-distribution noises and adversarial attacks.

The variational inference extension enables effective uncertainty estimation.

Abstract

This paper concerns pseudo labelling in segmentation. Our contribution is fourfold. Firstly, we present a new formulation of pseudo-labelling as an Expectation-Maximization (EM) algorithm for clear statistical interpretation. Secondly, we propose a semi-supervised medical image segmentation method purely based on the original pseudo labelling, namely SegPL. We demonstrate SegPL is a competitive approach against state-of-the-art consistency regularisation based methods on semi-supervised segmentation on a 2D multi-class MRI brain tumour segmentation task and a 3D binary CT lung vessel segmentation task. The simplicity of SegPL allows less computational cost comparing to prior methods. Thirdly, we demonstrate that the effectiveness of SegPL may originate from its robustness against out-of-distribution noises and adversarial attacks. Lastly, under the EM framework, we introduce a probabilistic generalisation of SegPL via variational inference, which learns a dynamic threshold for pseudo labelling during the training. We show that SegPL with variational inference can perform uncertainty estimation on par with the gold-standard method Deep Ensemble.