Functional Connectome Fingerprinting Using Shallow Feedforward Neural Networks

Abstract

Although individual subjects can be identified with high accuracy using correlation matrices computed from resting-state functional magnetic resonance imaging (rsfMRI) data, the performance significantly degrades as the scan duration is decreased. Recurrent neural networks can achieve high accuracy with short duration (72s) data segments but are designed to use temporal features not present in the correlation matrices. Here we show that shallow feedforward neural networks that rely solely on the information in rsfMRI correlation matrices can achieve state-of-the-art identification accuracies ([≥] 99.5%) with data segments as short as 20s and across a range of input data-size combinations when the total number of data points (# regions x # time points) is on the order of 10,000.