Semi-Supervised Variational Autoencoder for Survival Prediction

TL;DR

This paper introduces a semi-supervised variational autoencoder that predicts survival groups from tumor segmentation masks, demonstrating high generalization and effectiveness with limited labeled data on a public brain tumor dataset.

Contribution

It presents a novel semi-supervised VAE model that can utilize any tumor segmentation output, improving survival prediction with fewer labeled samples.

Findings

Effective survival classification on BraTS 2019 dataset

Model generalizes across different scanning platforms and pulse sequences

Performs well with limited labeled data

Abstract

In this paper we propose a semi-supervised variational autoencoder for classification of overall survival groups from tumor segmentation masks. The model can use the output of any tumor segmentation algorithm, removing all assumptions on the scanning platform and the specific type of pulse sequences used, thereby increasing its generalization properties. Due to its semi-supervised nature, the method can learn to classify survival time by using a relatively small number of labeled subjects. We validate our model on the publicly available dataset from the Multimodal Brain Tumor Segmentation Challenge (BraTS) 2019.