Student profile

Accepted into Duke University

GPA: 3.95

SAT: 1570

Extracurricular activities: Bioinformatics Research, Computational Pathology Research, Robotics Club, DECA, Math Team, Quiz Bowl Team, Community Service 



Some students have a background, identity, interest, or talent that is so meaningful they believe their application would be incomplete without it. If this sounds like you, then please share your story.



As I scanned an endless list of airfares in search of a good deal, one in particular caught my eye: $40. The price of a roundtrip ticket to St. Louis on Delta Airlines, more than ten times cheaper than any other available fare. Sensing a potential opportunity, I promptly changed the destination to Florida: $45 on Delta. California was $60. Kailua-Kona Hawaii, my dream vacation spot, was only $70. I had to act fast. Hunched over the keyboard typing furiously, I raced to click “Complete Reservation.” The appearance of the confirmation number on the screen elicited a sigh of relief as I finished in the nick of time. Moments later, the system was “briefly taken offline for maintenance.” My time was up, but this experience marked only the beginning of my journey into computational medicine.  


Finding good travel deals required a strong understanding of patterns in pricing data. These became clear after just a few weeks, as I noticed that airfares were cheapest on Tuesday mornings and hotel rates skyrocketed when booking within three months of travel. Regardless of the destination, my aim was invariant: to take trips of a lifetime at affordable prices.

This passion for maximizing value soon extended to writing artificial intelligence software. Completing my first machine learning program helped me realize the efficacy of large scale, computationally driven data analysis. The power of this algorithm lied in its ability to help understand patterns very similar to those that drove dynamic flight pricing.

My desire to work in medicine began to take shape just a few short months later, when I was misdiagnosed with Marfan’s Syndrome during an annual checkup in my freshman year. This event was just one of many unsuccessful diagnoses that my family had experienced. My grandfather suffered greatly due to a lately diagnosed aneurysm, while my grandmother lost her life to ovarian cancer detected only in its advanced stages. Why was diagnosis so difficult? Whether at a genetic level or in medical images, there had to have been patterns that aided in making definitive conclusions.

I embarked on a crusade, determined to find the patterns that the doctors had missed. Noticing the flaw in Delta’s system had been a good first step, but I was ready to take on a bigger challenge. As a research student at Harvard’s Biomedical Cybernetics Laboratory, my first project involved examining how intrinsic patterns in genetic mutations could be leveraged to compress large genomic data files. Scouring through endless strings of As, Cs, Ts, and Gs, I initially struggled to find meaning in the seemingly random sequences. However, my faith in patterns drove a continuous cycle of new ideas and experimentation. The breakthrough finally arrived when I discovered how to apply deep learning to systematically identify redundancies in the human genome. The result was magical, superhuman in a sense. I was immediately hooked and eager to learn more.

My intrigue with artificial intelligence quickly spread to other medical applications. Learning about developments in areas such as melanoma diagnosis, I turned my attention to medical image analysis. This past summer I interned at Harvard Medical School’s Heng Laboratory, using convolutional neural networks to predict breast cancer incidence from digital images of benign breast disease biopsies. As I continually worked to improve model performance, my efforts now had the potential to save a few lives in a hospital instead of just a few dollars on a flight.

I hope to utilize my passion for finding patterns with technology to help redefine modern medicine and render disease obsolete. While the inner workings of a prediction algorithm may still be abstract and obscure, the output can be something tangible like a novel drug treatment or early diagnosis. Being able to witness this impact on people’s lives drives me to pursue artificial intelligence’s applications to medicine, and I look forward to contributing to the next generation of lifesaving technologies.