PUBLICATIONS
Embedded Files
Foundations of Computer Vision & Machine Learning
Imbalance Classification using Regrouping
Le Peng, Yash Travadi , Ju Sun. Accepted by NeurIPS’22 Workshop: When Medical Imaging Meets NeurIPS, 2022.
Proposed an imbalanced learning framework that can:
can transform an imbalanced classification problem into a balanced classification problem
can easily plug into the existing imbalanced learning framework
can improve the imbalanced learning performance
Rethink Transfer Learning in Medical Imaging
Le Peng, Hengyue Liang, Taihui Li, Ju Sun. Submitted to IEEE Transaction on Medical Imaging, 2022. [paper][project website]
Proposed a novel transfer learning strategy targeting boosting the model performance and shortening the inference time. The method has been tested on a variety of medical applications including classification (2D & 3D) and segmentation.
Early Stopping beyond Supervised Learning: Self-Validation
Taihui Li, Zhong Zhuang, Hengyue Liang, Le Peng, Hengkang Wang, Ju Sun. Accepted by British Machine Vision Conference (BMVC), 2021. [paper]
Proposed a novel early stopping mechanism for deep image prior based methods.
Learning Disentanglement Representations with Federated Learning
Le Peng, Gaoxiang Luo, Ju sun. In preparation for IEEE Transactions on Medical Imaging.}, 2022
Proposed a novel scheme in coporate federated learning with disentanglment learning.
AI Models and System for Healthcare
Evaluation of federated learning variations for COVID-19 diagnosis using chest radiographs from 42 US and European hospitals
Le Peng, Gaoxiang Luo, Andrew Walker, Zachary Zaiman, Emma K Jones, Hemant Gupta, Kristopher Kersten, John L Burns, Christopher A Harle, Tanja Magoc, Benjamin Shickel, Scott D Steenburg, Tyler Loftus, Genevieve B Melton, Judy Wawira Gichoya, Ju Sun, Christopher J Tignanelli Accepted by Journal of the American Medical Informatics Association (JAMIA), 2022. [paper]
Collaborated work with Indiana University, Emory University Fairview, and Nvidia
Using Nvidia Clara Train framework to implement a real-world federated learning system. Work was highlighted in NVIDIA Clara white paper
Ju Sun, Le Peng, Taihui Li, Dyah Adila, Zach Zaiman, Genevieve Melton, Nicholas E Ingraham, Eric Murray, Daniel Boley, Sean Switzer, John L Burns, Kun Huang, Tadashi Allen, Scott D Steenburg, Judy Wawira Gichoya, Erich Kummerfeld, Christopher J Tignanelli. Accepted by Radiology: Artificial Intelligence, 2021. [medRxiv]
A real-world implement of covid-19 chest Xray diagnoisis model. Press coverage can be found in StarTribune and UMN News
On Going Projects
Robustness in Computer Vision
Designed algorithms to detect early stopping for unsupervised leaning
Improving the robustness of deep learning to natural corruptions in object recognition and detection, and in inverse problems
Fair and Incentivized Federated Learning
designing an incentive scheme to encourage and retain participation of federated learning
Incorporating practical factors such as participants’ strategic behaviors and short- and long-term payoffs.
Google Sites
Report abuse