Drew Parker

PIGMENT is a deep learning framework that automates the segmentation and quantification of APP-positive pathology in porcine white matter histology, achieving high detection rates with limited expert annotations.

This paper develops a harmonization framework for structural connectomes to reduce site-related biases, showing that gamma-GLM improves generalizability and statistical power in multi-site analyses.

The paper investigates how different parcellation atlases affect deep learning tasks on structural connectomes for age and sex prediction, finding that atlas choice is generally not critical.

This study uses a novel connectomic measure to identify structural connectivity alterations in ASD and links network topology changes to autism severity.

This study introduces a novel connectomic measure, NAS, to quantify structural network disruptions in TBI patients and shows its superiority over existing graph theory measures.

This study evaluates variability in white matter bundle segmentation across 42 groups using the same dataset, finding protocol variability exceeds other sources of inconsistency.

This study compares probabilistic and deterministic diffusion tensor imaging tractography with stereotactic coordinates for targeting the ventral intermediate nucleus in DBS, finding probabilistic tractography most accurate.

This study systematically analyzes the consistency of structural and functional connectomes across different generation methods and populations, finding structural connectomes more consistent than functional ones.

The paper constructs a multifactorial model of white matter connectivity during adolescence using Bayesian networks and diffusion imaging data, showing that architectural specifications become more influential with age.

This paper shows that diffusion tensor imaging data are site-specific and demonstrates that ComBat effectively harmonizes them while preserving biological variability.