Predicting individual traits from models of brain dynamics accurately and reliably using the Fisher kernel

Abstract

AbstractPredicting an individual’s cognitive traits or clinical condition using brain signals is a central goal in modern neuroscience. This is commonly done using either structural aspects, or aggregated measures of brain activity that average over time. But these approaches are missing what can be the most representative aspect of these complex human features: the uniquely individual ways in which brain activity unfolds over time, that is, the dynamic nature of the brain. The reason why these dynamic patterns are not usually taken into account is that they have to be described by complex, high-dimensional models; and it is unclear how best to use information from these models for a prediction. We here propose an approach that describes dynamic functional connectivity and amplitude patterns using a Hidden Markov model (HMM) and combines it with the Fisher kernel, which can be used to predict individual traits. The Fisher kernel is constructed from the HMM in a mathematically principled manner, thereby preserving the structure of the underlying HMM. In this way, the unique, individual signatures of brain dynamics can be explicitly leveraged for prediction. We here show in fMRI data that the HMM-Fisher kernel approach is not only more accurate, but also more reliable than other methods, including ones based on time-averaged functional connectivity. This is important because reliability is critical for many practical applications, especially if we want to be able to meaningfully interpret model errors, like for the concept of brain age. In summary, our approach makes it possible to leverage information about an individual’s brain dynamics for prediction in cognitive neuroscience and personalised medicine.