Elizabeth M. Sweeney

The paper uses multi-compartment T2 relaxometry to show that parenchymal cerebrospinal fluid fraction increases quadratically with age in brain tissue.

QSMRim-Net is a deep learning model that identifies chronic active multiple sclerosis lesions on QSM images by addressing class imbalance, achieving high accuracy and correlation with expert annotations.

This study uses graph matching to analyze functional connectivity reorganization in stroke patients, showing it correlates with motor recovery and structural disconnection.

The paper uses a three-compartment T2 relaxometry model to show that CSFF increases with age in non-active MS patients, indicating higher perivascular space load in cerebral cortical regions.

This paper develops an automated method to detect paramagnetic rim lesions in MS using MRI data and radiomic features with a random forest classifier.

This paper evaluates MS lesion segmentation algorithms using a new open-science infrastructure, finding that automatic methods lag behind human experts in detection and delineation.