A Wide and Deep Neural Network for Survival Analysis from Anatomical Shape and Tabular Clinical Data

TL;DR

This paper presents a novel wide and deep neural network that integrates anatomical shape data and clinical information to predict Alzheimer's disease progression, handling censored data effectively.

Contribution

It introduces an invariant neural network architecture that combines shape and tabular data for survival analysis, improving prediction accuracy over existing models.

Findings

Outperforms shape-only deep networks and linear models on clinical data

Learns effective shape descriptors that enhance clinical biomarker predictions

Handles right censored survival data effectively

Abstract

We introduce a wide and deep neural network for prediction of progression from patients with mild cognitive impairment to Alzheimer's disease. Information from anatomical shape and tabular clinical data (demographics, biomarkers) are fused in a single neural network. The network is invariant to shape transformations and avoids the need to identify point correspondences between shapes. To account for right censored time-to-event data, i.e., when it is only known that a patient did not develop Alzheimer's disease up to a particular time point, we employ a loss commonly used in survival analysis. Our network is trained end-to-end to combine information from a patient's hippocampus shape and clinical biomarkers. Our experiments on data from the Alzheimer's Disease Neuroimaging Initiative demonstrate that our proposed model is able to learn a shape descriptor that augments clinical biomarkers and outperforms a deep neural network on shape alone and a linear model on common clinical biomarkers.