Sam Payabvash, MD

Stroke imaging: Devising prognostic model for stroke patients based on the location of brain parenchymal damage on initial scans. Application of deep neural networks for identification of acute infarct on head CT, localization of arterial occlusion, and grading of arterial collaterals on admission CT angiography. Outcome prediction based on imaging patterns of hemorrhagic stroke.

Head and neck cancers: Applying texture analysis, radiomics and machine-learning models for differentiation of neoplasms, prediction of molecular subtypes, and prognostication beyond current staging schemes based on CT, MRI, and PET scans in patients with tumors of brain and neck.

Brain connectivity in neurodevelopmental disorders: Examining the functional and microstructural connectivity of the brain in children at risk of autism and neurodevelopmental disorders based on diffusion tensor imaging and tractography. Using machine learning models to devise bioimaging biomarkers for neurodevelopmental disease based on quantitative metrics of brain connectivity.