I am a data scientist and ML researcher based in Vancouver, BC, currently in my MDS at UBC while contributing to three research labs across biomedical engineering, neuropathology, and addiction psychiatry.
My background spans electronics engineering, health informatics, and data science. That combination gives me the rigour to build robust ML pipelines and the domain knowledge to make them clinically meaningful.
Whether it is 6DOF VR sensor streams from Alzheimer's patients, post-mortem IHC brain slides, or large-scale health claims, I am drawn to data that sits at the edge of what medicine can currently see.
Outside the lab: hiking the North Shore mountains, lifting heavy things.
Three concurrent positions at UBC spanning cognitive neuroscience, neuropathology, and addiction psychiatry.
Supervisor: Dr. Manu Madhav · UBC Biomedical Engineering
Supervisor: Dr. Veronica Hirsch-Reinshagen · UBC Pathology / VGH
Supervisor: Dr. Michael Krausz · UBC Psychiatry
Hover a cluster to illuminate its connections. Each node is a skill.
Independent builds, coursework, and capstone work — separate from lab research.
Reproducing Stanford's INSPECT benchmark — a late-fusion pipeline combining CTPA deep learning (LRCN: ResNet encoder + GRU sequence model) with structured EHR data (LightGBM) to predict PE and 8 downstream clinical outcomes. Trained on GCP with GCS storage.
Predicts pChEMBL binding affinity against EGFR (cancer-relevant kinase) using 13k ChEMBL molecules. Compares Morgan fingerprint MLP vs Graph Attention Network (3-feature and 7-feature). MC Dropout uncertainty quantification on both models. Scaffold-split evaluation.
3-class MRI classification (Cognitively Normal, MCI, Alzheimer's) on the OASIS-1 dataset. CNN pipeline with Grad-CAM interpretability for radiologist-friendly visualisation. Subject-level splits to prevent data leakage.
Conference presentations, in-progress papers, and technical writing on Medium.
Co-authored under Dr. Michael Krausz (UBC Psychiatry). Systematic review, data extraction, and full manuscript writing on ABI prevalence in opioid-overdose populations.
Presented at the PATHS To A Cure Neuroscience Research Conference, UBC. 🏆 Best Oral Presentation.
Accessible, in-depth content on ML, clinical data science, and neuroscience for a technical audience.