profile_picture
Stanley Z. Hua
Data Scientist @ SickKids, Toronto, Ontario, CA
stanley.z.hua@gmail.com

Hi! My name is Stan, and I want to create equitable and robust machine learning systems that can meaningfully improve patient care, especially for those in under-served populations.

I care deeply about the usefulness of machine learning when deployed clinically, and thus much of my research interests revolve around methods to improve generalization when data is not representative, noisily labeled or completely unlabeled. I’ve worked on a variety of topics:

  1. Automatic classification of anatomical planes in pediatric renal ultrasound sequences for enabling community point-of-care ultrasound
  2. Longitudinal adaptation and evaluation of a deep ultrasound classifier of a severe pediatric hydronephrosis
  3. Curating a large dataset of openly available microscopy images for developing microscopy foundation models and transfer learning

I also believe in slow science, which is about the careful and meticulous art of advancing science.

Shoot me an email if you’d like to chat! And if you include the word “stupefy” in your email, I’ll know you read this :).

Interests

  • ML for Healthcare
  • Trustworthy ML
  • Data-Centric ML

Academia

University of Toronto
2019 - 2024
B.Sc. Computer Science Specialist

News

Publications

Machine Learning-Enabled Renal Ultrasound View Labeling to Expand Use of Point-Of-Care Imaging in Community Settings, 2024, Nature Conference on Precision Child Health
Stanley Hua , Lauren Erdman
Supervised Contrastive Learning for Improved View Labeling in Pediatric Renal Ultrasound Videos, 2023, 20th IEEE International Symposium on Biomedical Imaging (ISBI)
Stanley Hua , Irene Y. Chen , Alex X. Lu , Lauren Erdman
From Single-Visit to Multi-Visit Image-Based Models: Single-Visit Models are Enough to Predict Obstructive Hydronephrosis, 2022, 18th International Symposium on Medical Information Processing and Analysis (SIPAIM)
Stanley Hua , ... , Anna Goldenberg , Lauren Erdman
CytoImageNet: A large-scale pretraining dataset for bioimage transfer learning, 2021, NeurIPS Workshop on Learning Meaningful Representations of Life
Stanley Hua , Alex Lu , Alan Moses

Projects

Includes research projects, passion projects and hackathons.
[Passion Project] Alexandria: An AI Book Maker using ChatGPT
[Research] The Effect of PCA Dimensionality on K-Means Clustering of Medical Images (under small sample sizes)