Jungkyu Park

Jungkyu Park

PhD Candidate

NYU School of Medicine

Biography

I am a PhD candidate in the Biomedical Imaging & Technology program at NYU School of Medicine, advised by Professor Krzysztof J. Geras and Professor Linda Moy.

I am working on the development and deployment of multimodal deep learning models for breast cancer screening, using 2D and 3D mammography images as well as ultrasound images.

Interests
  • Efficient Deep Learning
  • Multimodal Learning
  • Medical Imaging
Education

  • PhD in Biomedical Imaging & Technology, 2019 - Present

    NYU School of Medicine

  • MS in Data Science, 2019

    New York University

  • BS in Computer Science, 2017

    University of California, Irvine

News

Pre-doctoral TL1 Trainee ($92k in funding)
NYU Clinical & Translational Science Institute Aug 2023
Interviewed for the ‘CDS Incredible Alumni’ series
NYU Center for Data Science Mar 2021
#1 Winning Team at the DBTex Challenge
The DBTex Challenge Feb 2021
Top 10% Reviewer
NeurIPS 2020 Oct 2020
Best Capstone Project Award
NYU Center for Data Science Oct 2020

Recent Talks

Lightning talk at ML4H 2023
I presented our work on the Multi-modal Transformer (MMT), a deep neural network that integrates mammography and ultrasound to improve breast cancer detection and risk prediction.
Dec 10, 2023 ML4H 2023
Video
Lightning talk at ML4H 2023
Grand Challenge: Digital Breast Tomosynthesis Lesion Detection
The DBTex Challenge was the international competition focused on detecting breast lesions in Digital Breast Tomosynthesis (DBT) images using artificial intelligence. As part of the winning team, I presented our approach and key takeaways from the competition.
Feb 16, 2021 SPIE Medical Imaging Digital Forum
Video
Grand Challenge: Digital Breast Tomosynthesis Lesion Detection

Recent Posts

Data Prefetching in Deep Learning
In the previous post I explained that, after a certain point, improving efficiency of the data loaders or increasing the number of data loaders have a marginal impact on the training speed of deep learning.
Jungkyu Park
Last updated on Aug 10, 2023 5 min read
Visualizing data loaders to diagnose deep learning pipelines
Have you wondered why some of your training scripts halt every n batches where n is the number of loader processes? This likely means your pipeline is bottlenecked by data loading time, as shown in the following animation:
Jungkyu Park
Last updated on Aug 10, 2023 4 min read
ICS 46: Frequently made mistakes
These are some of the most frequent mistakes students made in the lab when I worked as a tutor for ICS 46, Data Structure Implementation and Analysis. Anyone who is taking the class could benefit by reviewing these examples and trying to figure out what is wrong with each of them.
Jungkyu Park
Last updated on Aug 10, 2023 3 min read

Projects

Breaking Numerical Reasoning in NLI
This project was one of four exemplary projects selected in Spring 2019 for the Natural Language Understanding and Computational Semantics course at NYU.
Breaking Numerical Reasoning in NLI
Web app: My UCI class is full
A web application which helped 1400+ students enroll in 5300+ courses without waitlists by sending 52k+ alerts.
Web app: My UCI class is full

Popular Topics

Algorithm Biomedical Imaging Breast Cancer Computer Vision COVID-19 Data Engineering Debugging Deep Learning Natural Language Inference Natural Language Processing Programming Simulation Teaching Visualization Web Development