Saige E. Rutherford

This paper updates the normative modelling protocol for computational psychiatry, including new methods, software tools, and models for neuroimaging data analysis.

The paper establishes a normative reference for large-scale human brain dynamics using resting-state fMRI data from over 10,000 individuals, revealing systematic reorganization across development and ageing and disorder-specific deviations in mental health conditions.

The paper uses normative modeling to assess cerebellar morphology differences in 15q11.2 CNV carriers compared to non-carriers, finding modest and individualized effects.

The paper maps cerebellar anatomical variations across the lifespan and clinical illnesses, revealing heterogeneity in conditions like autism, schizophrenia, and Alzheimer's.

This paper extends hierarchical Bayesian regression to model non-Gaussian data with skewness and kurtosis, showing improved performance on neuroimaging datasets.

This paper expands a normative model repository to include lifespan trajectories of brain structure and function, showing their advantage over raw data in neuroimaging tasks.

This work develops normative models of brain growth and aging using neuroimaging data to quantify variability across disorders and improve case-control inferences.

This paper introduces a standardized protocol for normative modeling in computational psychiatry using the PCNtoolkit, enabling analysis of biological heterogeneity without relying on case-control comparisons.

This paper introduces a federated probabilistic framework using hierarchical Bayesian regression for multi-site normative modeling of neuroimaging data, showing improved accuracy and adaptability.

This study shows predictive models applied to resting state fMRI data achieve higher test-retest reliability than individual features across various scanning and processing conditions.

This study uses large neuroimaging datasets to develop a predictive model classifying children who stutter from fluent peers, achieving 72% accuracy.

This study builds predictive neuromarkers of children's neurocognitive abilities from brain connectivity data, showing reliable generalization across sites and explaining over 14% of variance.

This study identifies a connectome-wide neurosignature of the P factor, a general risk factor for psychopathology, and shows its generalizability across adolescent brain imaging data.

The paper develops a CNN for automated fetal brain masking in rs-fMRI data, achieving good performance across different scanners and populations.

The paper identifies FPN-DMN network separation as a mediator of general intelligence, showing it predicts intelligence scores and reverses in different task conditions.

The paper identifies a low-rank structure in resting state connectomes, showing that a small number of components account for most inter-individual variation and can predict neurocognitive and clinical variables.

This study identifies resting state fMRI connectivity patterns predicting neurocognitive profiles in youth across three domains, showing cross-site generalizability for General Ability.

The paper predicts intelligence from whole-brain task activation patterns, achieving a 0.68 correlation, better than other modalities.