Penghui Du ☕️

Hi! I’m Penghui, a first-year master’s student in Neuro-X at EPFL, with a strong passion for computational neuroimaging and neuromodulation techniques. My long-term goal is to pursue a PhD and contribute to advancements in this fascinating research field.

Experience

 
 
 
 
 
École Polytechnique Fédérale de Lausanne (EPFL)
Master Student in Neuro-X
September 2024 – Present Ecublens, Switzerland
 
 
 
 
 
Martinos Center for Biomedical Imaging, Harvard Medical School
Undergraduate Research Assistant
July 2023 – December 2023 Charlestown, Unites States
  • Supervisor: Dr. Jingyuan Chen (https://jechenlab.com/)
  • Research Focus: Human Cerebral Cortex Organization Estimated by Functional PET-FDG “Metabolic Connectivity”
 
 
 
 
 
University of Zurich
Regular Visiting Student in Neuroinformatics
February 2023 – June 2023 Zurich, Switzerland
 
 
 
 
 
Southern University of Science and Technology
BSc in Intelligent Medical Engineering
August 2020 – June 2024 Shenzhen, China
  • Academic Supervisor: Dr. Quanying Liu
  • GPA: 3.84 / 4 (92.79), Ranking 2 / 22
  • 2024 Outstanding Graduate
  • 2022 BME “Fortunatt” Scholarship
  • 2022 Outstanding Student Scholarships (First Prize)

Accomplish­ments

Neuromatch Academy
2022 Neuromatch Computational Neuroscience Summer School
I studied computational neuroscience fundamentals such as reinforcement learning, leaky Integrate-and-Fire models, Hodgkin-Huxley models with my teammates. We then conducted an project on RNN and working memory, and presented our results to other teams.
See certificate
Tsinghua University and Peking University
Merit Student of CLS-CIBR-IDG Summer School in Neuroscience
I attended various neuroscience lectures in the summer school, followed by our teams presentation on a chosen paper. I was recognized with a Merit Student Award.
See certificate
Guangdong Biomedical Engineering Association
First Prize in 2022 Guangdong Undergraduate Biomedical Engineering Innovation Design Competition
We designed a deep learning model, combining Transformer and UNet, for labeling the key organs involved in radiotherapy in CT images. Our unique pre-training approach ensured high segmentation accuracy and reduced computational cost, earning us first prize in the competition.
See certificate
Department of Education of Guangdong Province
First prize in 13th “Challenge Cup” Entrepreneurship Competition.
We designed a business plan for manufactoring seizure monitor devices for severely ill newborns, and won first prize in the competition. I am team captain in this competition, and I am responsible for proposing technical ideas and designing business plan.
See certificate

Projects

[OHBM 2024] Human Cerebral Cortex Organization Estimated by Functional PET-FDG Metabolic Connectivity
The recent development of high-temporal resolution functional PET (fPET) introduces an emerging focus on “metabolic connectivity (MC)” , providing a complementary perspective to the hemodynamic-based “functional connectivity (FC)” assessed by fMRI. In this study, we applied a connectivity gradient-based analytical scheme on a resting-state simultaneous fPET-fMRI dataset, aiming to characterize the detailed cortical organization of fPET-derived MC and understand how it differs from the fMRI-derived functional network structures.
[OHBM 2024] Human Cerebral Cortex Organization Estimated by Functional PET-FDG Metabolic Connectivity
In Prep | Exploring the effect of psychotropic drugs on zebrafish brain dynamics by high-throughput calcium imaging
In this research project, we employed a novel high-throughput screening system to collect a vast amount of neural activity data (via calcium imaging) from zebrafish exposed to psychotropic drugs. Subsequently, we analyzed the data to explore the influence of psychotropic drugs on the activity across various brain regions, as well as the functional connectivity within the zebrafish’s brain.
In Prep | Exploring the effect of psychotropic drugs on zebrafish brain dynamics by high-throughput calcium imaging
Assessing Generalization of Cognitive Tasks Using Multi-regional Modular Recurrent Neural Networks with Transfer Learning
In this study, we proposed a multi-regional modular recurrent neural network to simulate the cognitive processes. The model is structured into three different modules: perception, information integration, and decision. Here a transfer learning approach is adopted to investigate generalizability across tasks. After training models on source tasks, we fixed the information integration layers, transferred the models to target tasks, and tested their performance. By comparing the performance of different source-target task pairs, we assessed the similarity between different cognitive tasks.
Assessing Generalization of Cognitive Tasks Using Multi-regional Modular Recurrent Neural Networks with Transfer Learning
CT image segmentation of key organs for nasopharyngeal cancer radiation therapy
Nasopharyngeal cancer is a serious tumor in the upper pharynx, requiring careful outlining of key organs for radiotherapy planning. Traditionally, this is done manually on CT images, a slow, labor-intensive process reliant on the physician’s experience. To improve this, we use a pre-trained 3D model and an enhanced TransUNet network to automate segmentation of four key organs with impressive results.
CT image segmentation of key organs for nasopharyngeal cancer radiation therapy

Publications

(2024). Integration of cognitive tasks into artificial general intelligence test for large models. iScience.

DOI

(2023). Promoting interactions between cognitive science and large language models. The Innovation.

DOI

(2022). Transfer learning to decode brain states reflecting the relationship between cognitive tasks. In International Workshop on Human Brain and Artificial Intelligence.

PDF DOI

Contact

  • penghui-du@outlook.com / penghui.du@epfl.ch
  • +41 77 211 89 07 / +86 158 8937 2606
  • 4 Rue Favre-Louis, Ecublens, Vaud 1024