cv

Vice President Research

University of Toronto Machine Intelligence Student Team

Research Assistant

Shanghai Jiao Tong University (Dr. Bin Sheng)

Working on systematic reviews.

• Working on "Who Does Medical AI Serve? Hidden Institutional Logics Behind Clinical AI"

Student Researcher

University Health Network (Dr. Jun Ma)

Working on medical MLLMs.

• Delivered a mini course on multimodal large language models
• Applied C-RadioV4-style multi-teacher distillation on the 35TB SLC-PFM Pathology dataset from three encoders: H-optimus-1, UNI2-H, and Virchow2
• Utilized feature clustering with summary/patch token extraction to reduce the wall time from years to weeks on only 4 H100 GPUs
• Explored zero-shot feature upsampling (NAF) on pathology foundation models
• Managed the MICCAI FLARE 2026 challenge
• Fine-tuned and evaluated MedGemma 1 and 1.5 on various tasks across classification, multi-label classification, detection, instance detection, cell counting, regression and report generation, using Compute Canada (SLURM) clusters, as one of the baselines
• Made skills for agents to remotely access MFA-controlled SLURM clusters through shared ControlMaster SSH sessions

Research Team Lead

University of Toronto Machine Intelligence Student Team

Researching on downstream tasks in medical image processing.

• [UNSURE@MICCAI2026] The SegWithU Project (SegWithU: Uncertainty as Perturbation Energy for Single-Forward-Pass Risk-Aware Medical Image Segmentation)
• The AIP Project (Automated Iterative Pseudo-labeling)

Research Intern

Vector Institute (Dr. Bo Wang)

Competing in medical AI challenges presented by MICCAI.

• Ranked 4th in the PANORAMA Challenge
• Ranked 4th in the SegSTRONG-C Challenge
• Helped with Fast nnU-Net

Team Lead

Project Neura

Gathering researchers and developers to bring their ideas into reality. Developed projects like LEADS and MIP Candy.

• Gathering researchers and developers to bring their ideas into reality
• Developed MIP Candy (A Candy for Medical Image Processing), a fast-prototyping engine for complete experiment pipelines for medical image processing
• Developed LEADS (Lightweight Embedded Assisted Driving System), an end-to-end instrumentation, control, and analysis system for electric race cars
• Built Erbium, an infrastructure cloud compute platform with Docker

BASc

University of Toronto

Engineering Science (PEY Co-op)

• Machine Intelligence

OSSD and AP

St. Thomas of Villanova College

Compulsory Education

Shanghai Southwest Weiyu Middle School

Dean's Honour List

University of Toronto

Awarded for outstanding academic performance.

Digital Citizenship Graduation Award

STEM Fellowship, Royal Bank of Canada, LinkedIn Learning, and Villanova College

In recognition of exemplary digital leadership and the promotion of Canadian values.

Ontario Scholar

Ontario Ministry of Education

LEADS

Lightweight Embedded Assisted Driving System

• Developed an onboard instrumentation system that displays data like the wheel speed
• Developed a webpage dashboard that remotely monitors the vehicles’ status from the Pit crew
• Developed a multi-camera streaming and recording system
• Developed support for saving and replaying vehicle data
• Developed an efficient real-time data link using pure TCP/IP
• Developed a data analysis tool set that teaches the driver how to drive
• Customized the Ubuntu OS
• Developed support for multiple screens
• Developed GPS support

MIP Candy

A Candy for Medical Image Processing

• Developed a framework that brings ready-to-use training, inference, and evaluation pipelines together with aesthetics, so users can focus on their experiments, not boilerplate
• Developed a visualization system that helps the user see the complex data
• Accelerated training with a preloading mechanism
• Accelerated sliding window custom CUDA kernels

The AIP Project

Automated Iterative Pseudo-labeling

• Proposed an innovative way to make pseudo-labeling more efficient and effective

CIV102 Bridge Project

Best Winner Solution for Automated Calculation and Optimization with Extensive Docs

• Top 1 in the cohort and 2nd in the class of 2025
• Developed a complex cross-section composition and solving system
• Developed a bridge solver
• Calculated the optimal cross-section dimensions by turning the design into a COP (Convex Optimization Problem)

SegWithU

Uncertainty as Perturbation Energy for Single-Forward-Pass Risk-Aware Medical Image Segmentation

• Proposed a novel single-forward-pass uncertainty estimation method that models uncertainty as perturbation energy
• Conducted experiments that show SOTA performance on ACDC, BraTS2024, and LiTS datasets