Neural Processes Mixed-Effect Models for Deep Normative Modeling of Clinical Neuroimaging Data
This work addresses the need for more scalable and flexible normative modeling in clinical neuroimaging to derive biomarkers for psychiatric disorders, representing an incremental improvement over existing methods.
The authors tackled the problem of poor scalability and reliance on fixed parametric kernels in normative modeling of neuroimaging data by proposing a deep normative modeling framework based on neural processes, which showed substantial improvements in novelty detection performance for certain diagnostic problems compared to a baseline Gaussian process regression approach.
Normative modeling has recently been introduced as a promising approach for modeling variation of neuroimaging measures across individuals in order to derive biomarkers of psychiatric disorders. Current implementations rely on Gaussian process regression, which provides coherent estimates of uncertainty needed for the method but also suffers from drawbacks including poor scaling to large datasets and a reliance on fixed parametric kernels. In this paper, we propose a deep normative modeling framework based on neural processes (NPs) to solve these problems. To achieve this, we define a stochastic process formulation for mixed-effect models and show how NPs can be adopted for spatially structured mixed-effect modeling of neuroimaging data. This enables us to learn optimal feature representations and covariance structure for the random-effect and noise via global latent variables. In this scheme, predictive uncertainty can be approximated by sampling from the distribution of these global latent variables. On a publicly available clinical fMRI dataset, we compare the novelty detection performance of multivariate normative models estimated by the proposed NP approach to a baseline multi-task Gaussian process regression approach and show substantial improvements for certain diagnostic problems.