Vijay Viswanathan's Personal Page

Links to Specific Projects:       Platelet Modelling Project      Computational Diagnosis (20 Under 20)

Platelet Modelling Project

During the summer of 2012, I worked at Stony Brook University as a Simons Summer Research Fellow. I worked on developing a computational model of blood platelet morphology. I wrote detailed mathematical expressions to parametrize the shape of a blood platelet in both activated and unactivated states. I then wrote C++ code to match a platelet image to the relevant mathematical expressions and to use this model to determine an approximation of the drag coefficient for this platelet. Both of these results were novel, and the drag coefficient measurement for individual platelet (the value was calculated specially for each input image) had never been done before.

With my work I hoped to pave the way for better, more personalized computational fluid dynamics simulations for blood diseases. I worked under Seetha Pothapragada and Dr. Yuefan Deng in the Center for Excellence in Wireless and information technology. With the results of my work and the help given to me, I was selected as a Regional Finalist in the Siemens Competition, which is a highly competitive science research contest. I got to present my work to a panel of 10 esteemed researchers at MIT.

As a result of my research and the Siemens Competition, I was featured as the subject of three local newspaper articles. I have displayed links to those three articles below.

As my first effort in structured computer science research, I learned a lot about what to do and what not to do when working for somebody else in a research setting. I also learned a lot about proper programming practices and graphics in C++. I learned more through this one project than any class in high school.

Below is a video of me explaining my work. This video was meant to be part of a submission for a different fellowship.

.

Diagnosis Diseases with Machine Learning

By the end of my senior year of high school, I heard about an opportunity where aspiring young entrepreneurs could apply with an idea to Peter Thiel, founder of Pay-Pal. If chosen, I would earn $100,000 to start the business with one major caveat: I would have to delay college for 2 years or drop out entirely.

Since I was interested in computational biology from my previous research, I had been thinking about different approaches to using machine learning to develop an image recognition system to diagnose diseases. This was a classic problem in image recognition and by no means was it a novel idea. However, the implementation of such an idea is quite difficult.

With the understanding that many fine pattern recognition algorithms exist, I decided to focus on developing the database of training data needed to power such a diagnosis system. I presented my idea in the application, got an interview with a business leader, and after 4 rounds of application reviews, soon found myself getting an invitation to be a finalist in the program. I was flown to San Francisco in April of 2012 and had an amazing time meeting other young minds and connecting with established Silicon Valley visionaries. I also received tips and advice on my project from wiser people.

Ultimately out of us 40 finalists (out of over 1000 total applicants) only 24 of us were chosen to be fellows, and I was not one of them. But this really sparked my interest in entrepreneurship and inventing cool things.

Another nice perk was the media coverage we got! I got a couple of minutes of speaking time on CNBC's televised documentary, 20 Under 20: Transforming Tomorrow. I also was briefly shown on CBS's 60 Minutes. Below is a picture of me from the show.

Smiley face