Project Mentor
Dr. Ji Yu
Department of Center for Cell Analysis and Modeling
Undergraduate Research Opportunity Description
Project Description | The summer project requires the student to develop and document test cases for a set of deep-learning models. These models are pre-trained to perform microscopy image analysis and deployed on our high-performance computing (HPC) cluster. The student will work closely with one or several experimental labs to implement image-analysis pipelines for the experimental data, e.g. quantifying the average cell division time of stem cells recorded by time-lapse microscopy, evaluate accuracy of the results, and document the analysis. |
Project Direction | The lab’s research focuses on developing AI algorithms for a variety of microscopy image analysis tasks, as well as infrastructures to integrate these algorithms to complex data analysis pipelines necessary for model biomedical research. The student’s work will specifically contribute to quality checking the AI models and also provide demos and documentations to the relevant models. |
Mentorship and Supervision | – Student will spend the first 2-3 weeks studying existing tutorial materials and replicate existing research results – Student will meet with the PI once per day to briefly discuss the work progress. – Student will make two presentations of their research throughout the summer semester – Student will attend RIP/Journal club seminars once a week. |
Student Qualifications | Junior/senior year student with an interest in deep learning / AI. |
Summer Schedule Options | Research Dates: May 19 to July 25, 2025 Schedule: Flexible schedule. Will be determined on a case-by-case basis. |
Project Continuation | Fall 2025 |
Academic Year Time Commitment | 6 hours/week |
Possible Thesis Project | Yes |
Application
Submit an online application for this research opportunity at https://quest.uconn.edu/prog/HRPSU25-37. The application deadline is Monday, February 17, 2025.
This application requires a resume or CV, GPA and science GPA. References should be available upon request.