In the classroom
Teaching & curriculum
I design and teach graduate courses that bring modern machine learning and biomedical data science to engineering students. Both pair every lecture with a hands-on coding lab and a team project — so students leave having built and analysed real biomedical data, not just read about it.
Biomedical Data Fundamentals BMEG 591N
An overview of the cellular and clinical imaging modalities and omics technologies behind modern biomedical data — their principles, trade-offs, and how they combine across micro-to-macro scales, taught with weekly Python labs.
Module 1 · Omics
- Central dogma & sequencing
- Alignment & variant calling
- Expression & single-cell
- Cancer genomics & phylogeny
- TCGA & proteomics
Module 2 · Microscopy & digital pathology
- Fluorescence & confocal
- Electron microscopy
- Digital pathology (H&E, IHC)
- Spatial transcriptomics (Visium)
Module 3 · Medical imaging
- Radiography & CT
- PET / SPECT
- Ultrasound
- MRI
- DICOM / PACS
Team project: students write a grant proposal built on multi-scale, multimodal public datasets — identifying the data, justifying sample size, and planning for pitfalls, like a real proposal.
Machine Learning in Medicine BMEG 591T
Machine learning across the breadth of healthcare — from deep learning for medical imaging, to language models for clinical text, to genomics — taught through lectures, hands-on coding, and real-world case studies.
Module 1 · Deep learning & medical imaging
- Neural nets & CNNs
- Transfer learning
- U-Net & 3D U-Net
- GANs & diffusion
- Vision Transformers (ViT)
- Grad-CAM
Module 2 · Language models for medicine
- Embeddings (Word2Vec, BioWordVec)
- Transformers & attention
- BioBERT / ClinicalBERT
- Clinical text mining
Module 3 · Genomics
- PCA / t-SNE / UMAP
- DNA/RNA sequence modeling
- Variant calling (DeepVariant)
- GWAS
- AlphaFold
- Drug discovery
Capstone · Multimodal learning
- Fusion strategies
- CLIP / MedCLIP
- Cross-modal attention
- Genomics + pathology for prognosis
Team project: an end-to-end data-analysis project on real medical data, proposed and scoped one-on-one with the instructor.
Student Experience of Instruction (UBC SEI). Interpolated-median ratings of 4.6–4.8 / 5 on “Overall, I learned a great deal from this instructor,” across both graduate courses over four offerings (2024–2026).
Curriculum development & program leadership. Management-committee member of the NSERC CREATE MUSIC program (multi-scale, multimodal image and omics computing for health). Member of the SBME AI Curriculum Committee (2026–), designing two new undergraduate courses: Programming and Computational Thinking for Biomedical Engineering and Biomedical Data Fundamentals (undergraduate).