a Ph.D. student of Bioengineering at University California, Los Angeles,
and research student of Biomedical Imaging Research Institute at Cedars-Sinai Medical Center.
My area includes deep learning, magnetic resonance imaging(MRI) and medical image analysis.
I am a researcher/developer/technology enthusiast, eager to bring latest techniques to biomedical industry, aiming to let people have easily accessible, more accurate, personalized, and affordable healthcare service.
With a strong background in both biomedical engineering and artificial intelligence, I am doing my best to crack down the most unsolved and challenging issues in magnetic resonance imaging(MRI) by translating deep artificial neural networks a.k.a deep learning into biomedical science.
I am also a coding zealot. I found my passion for programming when I was in high school. Ever since then, I designed and built many cool pieces of stuff, from small and simple android apps to complicated parallel computing.
I have a love of clean and elegant design, not only for how the product shows apparently but also for the inner structure of codes and styling. As a computer science graduate, I have made a lot of different application on variant platforms. And for a research student, I also have insights of scientific algorithm developing and implementation.
School of Engineering and Applied Science, University California, Los Angeles, CA, USA
School of Engineering and Applied Science, University of Pennsylvania, Philadelphia, PA, USA
School of Engineering and Applied Science, University of Pennsylvania, Philadelphia, PA, USA
Sino-Dutch Biomedical and Information Engineering School, Northeastern University, Liaoning, China
Implemented a fast Hough transformation algorithm in c++ program to create Brain MRI Tractography from MRI diffusion image (DWI)
Big data analysis on over 15 million hours data from 150 thousand ICU patients to predict comorbidities Designed and built neural networks with CNN and RNN (LSTM) in Keras, scikit-learn python packages Preliminary results had an 88.5 AUROC in predicting congestive heart failure patients, which is excellent for clinical application
2016 Spring term, Data Structure CIS Department 2015 Fall term, Programming Languages & Techniques, CIS Department Jobs includes recitation, course reviews, office hour, and assignment/exam grading.
Developed the atlas-based image segmentation algorithm to analyze the cardiac function from MRI. Applied image registration for MOCO to improve imaging quality and cardiac function assessment Relevant abstracts published on SCMR 2016 and ISMRM 2016
Leading researcher in algorithm development at Whole-Heart Medical Image Segmentation group. Applied nonrigid registration and atlas-based segmentation to goal mean error distance of 0.6 mm. GPU Parallel Computing (CUDA) is used to accelerate metric computing in segmentation algorithm