Built an algorithm that automatically reconstructs a 3D vascular model of a patient’s brain from 2D MRI images to aid in safer planning of surgeries.
My interest in the application of medical technology involving the human brain was sparked by accident when I enrolled in an “Introduction to Computational Neuroscience” class. It dawned upon me that, despite its complexities, the human brain can be modeled by algorithms in an attempt to unlock insights into its mysterious workings. That “eureka” moment drove me to venture deeper into this emerging? field by conducting research at Tandon Lab at the John P. and Katherine G. McGovern Medical School at UT Health. There I worked on two projects under the guidance of Dr. Nitin Tandon. My first research project involved the creation of software that could synthesize a 3D vascular model of the human brain using 2D MRI images given that its widely accessible and utilized in hospitals. This would augment the neurosurgeon’s arsenal of tools during surgery by providing an accurate vasculature model, essential in mapping safe electrode trajectories in the patient’s brain. To achieve this, I converted the radiology images into readable code followed by applying image filters such as Frangi and Otsu to extract tubular structures and map them into a 3D matrix. The newly designed software was subsequently implemented as part of a wider suite of commercial tools that Tandon Lab proposes for hospital settings once FDA approval is received. This research experience was the start of an exciting endeavor for me in working with brain-related technologies. The most exciting aspect of this project was the ability to convert organic into tangible code that computers can interact with and prompted me to think of different possibilities in which we can interface with the brain.