Scalable Multi-Task Gaussian Process Tensor Regression for Normative Modeling of Structured Variation in Neuroimaging Data
This work addresses the problem of capturing individual differences in brain function for clinical populations, offering a more sensitive tool for anomaly detection in neuroimaging, though it is incremental as it builds on existing probabilistic regression methods.
The authors tackled the challenge of modeling heterogeneous brain disorders by developing a scalable multi-task Gaussian process tensor regression framework for normative modeling of neuroimaging data, which substantially improved detection sensitivity over existing methods on a clinical fMRI dataset.
Most brain disorders are very heterogeneous in terms of their underlying biology and developing analysis methods to model such heterogeneity is a major challenge. A promising approach is to use probabilistic regression methods to estimate normative models of brain function using (f)MRI data then use these to map variation across individuals in clinical populations (e.g., via anomaly detection). To fully capture individual differences, it is crucial to statistically model the patterns of correlation across different brain regions and individuals. However, this is very challenging for neuroimaging data because of high-dimensionality and highly structured patterns of correlation across multiple axes. Here, we propose a general and flexible multi-task learning framework to address this problem. Our model uses a tensor-variate Gaussian process in a Bayesian mixed-effects model and makes use of Kronecker algebra and a low-rank approximation to scale efficiently to multi-way neuroimaging data at the whole brain level. On a publicly available clinical fMRI dataset, we show that our computationally affordable approach substantially improves detection sensitivity over both a mass-univariate normative model and a classifier that --unlike our approach-- has full access to the clinical labels.