I am a sophomore at Stanford University pursuing a career at the crossroads of mathematics, electrical engineering, and software. I'm interested in embedded systems, product design, philosophy, and technology applications in high-impact verticals (construction, space, climate change, healthcare, and security, to name a few). Long story short, I like to build things!
My current focus is split between automating circuit design with machine learning and democratizing mobile health. Reach out if you would like to grab a cup of coffee and chat!
Developing machine learning methods to optimize hardware.
March 2020 - Present
Collaborating with Stanford Radiology to develop a web tool for automating radiology tasks utilizing deep learning and computer vision.
August 2018 - May 2020
Applying computer vision to medical imaging. Designing deep neural networks for object detection in medical images and researching domain adaptation techniques for medical image synthesis.
Currently collaborating with medical professionals to develop an automated method for assessing postoperative wounds using artificial intelligence.
June 2017 - June 2019
Building mobile applications for the App Store to solve a wide variety of problems. Currently have over 20,000 downloads for all my apps combined from users all over the world. Attended WWDC in 2016, 2017, and 2018 on scholarship.
August 2015 - Present
Summit is an iOS app that summarizes the daily news. It helps busy people read the news in a matter of seconds. It has over 17,000 downloads and won me an Apple Scholarship to the Worldwide Developers Conference in 2016. It also won the 2017 Congressional App Challenge in the CA-17 (Silicon Valley) district.
Biosnap enables you to capture medical monitor data with a picture and store it automatically in your Health app swiftly without the usage of the mobile keyboard. With Biosnap, there's no reason to buy an expensive internet-connected medical monitor or maintain a log of measured biomarkers.
During the month of August in 2018, I developed a web tool for bone x-ray annotation as well as the backend infrastructure for image masking and cropping. It will fit into the full application pipeline which includes a machine learning system to clasify bone features and tumors using the masked regions.
I spent the summer of 2018 at the UC Berkeley Artificial Intelligence Research (BAIR) Lab building 3D and 2D renditions of segmentation architectures for automated brain tumor semantic segmentation in MRI scans. I set up a 3D Unet for training with patch normalization and percentile thresholding. I developed a variety of different 2D segmentation algorithms to evaluate axial brain slices one at a time, including CycleGAN generated data. Finally, I implemented a masking strategy that leverages whole tumor segmentation maps to bolster dice scores on enhancing tumor and tumor core segmentations.
The incidence of postoperative wound infections after lower extremity bypass can be as high as 10%-20%. An automated method of diagnosing wound complications would serve to limit the expense of time and money from hospitals, doctors, insurers, and patients. The algorithmic classification of wound images, due to variability in the appearance of wound sites, is a challenge. Deep convolutional neural networks (CNNs), a subgroup of artificial neural networks that exhibit great promise in the analysis of of visual imagery, may be leveraged to categorize surgical site wounds. We present Deepwound, a multi-label CNN trained to classify wound images with image pixels and labels as the sole inputs.
Biometric measurements captured from medical devices, such as blood pressure gauges, glucose monitors, and weighing scales, are essential to tracking a patient's health. Trends in these measurements can accurately track diabetes, cardiovascular issues, and assist medication management for patients. Currently, patients record their results and date of measurement in a physical notebook. It may be weeks before a doctor sees a patient’s records and can assess the health of the patient. This research presents a mobile application that enables users to capture medical monitor data and send it to their doctor swiftly using HealthKit. A key contribution of this paper is a robust engine that can recognized digits from medical monitors with an accuracy of 98.2%.
CS 106B Programming Abstractions in C++
CS 107E Computer Systems from the Ground Up
PHYSICS 61 Mechanics and Special Relativity
PHYSICS 63 Electricity, Magnetism, and Waves
MATH 51 Linear Algebra and Multivariable Calculus
MATH 52 Integral Calculus of Several Variables
MATH 53 Ordinary Differential Equations
MATH 115 Functions of a Real Variable
POLISCI 131L Modern Political Thought
ESF 15 College and the Good Life
MSE&E 178 + 472 The Spirit of Entrepreneurship
HUMBIO 51 Big Data for Biologists
WWDC Scholarship 2018
Recognized as a Young Innovator to Watch
Consumer Electronics Show 2018
$1,000 Academic Scholarship
WWDC Scholarship 2017
2017 Congressional App Challenge
California's 17th District (Silicon Valley)
Global Oncology Grand Prize
WWDC Scholarship 2016
Google Science Fair 2019 Regional Finalist
Top 100 Submissions Worldwide
2019 Cutler-Bell Prize in High School Computing
ACM & CTSA
$10,000 Academic Scholarship
Lead Author of Publication
IEEE International Conference on Bioinformatics and Biomedicine 2018
Top 10 Abstracts
Surgical Infection Society Annual Meeting 2018
Northern California + Nevada JSHS 2018
$1,000 Academic Scholarship
Lead Author of Publication
American Medical Informatics Association Annual Symposium 2018