Current Research

Machine Learning Processes for use in Surgical Intervention and Diagnoses

Developed convolutional neural networks for automated detection, segmentation, and volumetric
assessment of hematoma and peripheral edema in intracerebral hemorrhage patients, based on T2w MR
images. Tested various machine learning data augmentation techniques for automated recognition of the
placenta in pregnant Rhesus Macaques infected with ZIKV, for further use in AI-assisted detection of viral
complications during fetal development. Qualitatively identified regions of cartilage in the knee via
multimodal MR images for assessment of recovery following knee injury and subsequent reconstructive
surgery