A. Hawkins-Daarud

This paper develops BioNet, a biologically-informed deep learning model, to predict intratumoral heterogeneity in post-treatment glioblastoma using MRI and biopsy data, showing improved performance over existing methods.

The paper maps MRI-localized glioma biopsies to MRI habitats, identifying transcriptional tissue states and their association with cellular subpopulations and MRI features.

This paper uses mathematical modeling to differentiate between drug insensitivity and poor drug delivery in glioblastoma PDX models, validated by experimental results.

This study extends a GBM mechanistic model to investigate how ischemia influences distal tumor recurrence by incorporating vascular efficiency and hypoxia.

This study uses a computational model to show that hypoxic cell migration drives glioblastoma growth and distant recurrence.

The paper uses biopsy data and a mechanistic model to study interactions between EGFR and PDGFRA amplified sub-populations in glioblastomas.

The paper uses a machine learning model to predict seizure risk in glioma patients at presentation, finding sex, tumor size, and age as significant predictors.

The paper uses image-based metrics to predict TMZ response in GBM patients, finding that less invasive tumors show greater volumetric response and improved survival.

This study identifies sex-specific predictors of survival in glioblastoma patients using MR imaging and clinical data.