Image-based Survival Analysis for Lung Cancer Patients using CNNs

TL;DR

This paper introduces a CNN-based method for lung cancer survival analysis using image data, simplifying the task to median survival classification to enable training with small batch sizes, outperforming previous methods.

Contribution

It proposes a novel CNN approach that simplifies survival prediction to median survival classification, allowing effective training on limited batch sizes for 3D medical images.

Findings

Outperforms previous state-of-the-art methods on lung cancer dataset

Enables CNN training with small batch sizes for 3D medical images

Predicts survival effectively using learned image features

Abstract

Traditional survival models such as the Cox proportional hazards model are typically based on scalar or categorical clinical features. With the advent of increasingly large image datasets, it has become feasible to incorporate quantitative image features into survival prediction. So far, this kind of analysis is mostly based on radiomics features, i.e. a fixed set of features that is mathematically defined a priori. To capture highly abstract information, it is desirable to learn the feature extraction using convolutional neural networks. However, for tomographic medical images, model training is difficult because on the one hand, only few samples of 3D image data fit into one batch at once and on the other hand, survival loss functions are essentially ordering measures that require large batch sizes. In this work, we show that by simplifying survival analysis to median survival classification, convolutional neural networks can be trained with small batch sizes and learn features that predict survival equally well as end-to-end hazard prediction networks. Our approach outperforms the previous state of the art in a publicly available lung cancer dataset.