Anatomical Predictions using Subject-Specific Medical Data

TL;DR

This paper introduces a neural network-based method to predict individual brain MRI changes over time using diffeomorphic deformation fields, aiding in treatment planning and scientific understanding.

Contribution

The novel approach models brain changes with a CNN-predicted deformation field, incorporating subject-specific data to improve prediction accuracy.

Findings

The method achieves accurate brain change predictions.

External clinical data enhances prediction performance.

Model variants influence prediction quality.

Abstract

Changes over time in brain anatomy can provide important insight for treatment design or scientific analyses. We present a method that predicts how a brain MRI for an individual will change over time. We model changes using a diffeomorphic deformation field that we predict using function using convolutional neural networks. Given a predicted deformation field, a baseline scan can be warped to give a prediction of the brain scan at a future time. We demonstrate the method using the ADNI cohort, and analyze how performance is affected by model variants and the subject-specific information provided. We show that the model provides good predictions and that external clinical data can improve predictions.