PHM5009 – Precision Oncology
Course Overview
This course will provide an in-depth exploration of the molecular and genomic foundations of cancer, emphasizing how these insights inform individualized diagnosis, prognosis, and treatment strategies. Key topics include cancer hallmarks and genomics, identification of novel targetable vulnerabilities, biomarker discovery and validation. The course also introduces core artificial intelligence and machine learning concepts, progressing from foundational mathematical principles to advanced deep learning and transformer-based models, with a focus on biomedical and medical imaging applications relevant to precision oncology. It also addresses model optimization, evaluation, and the ethical considerations necessary for responsible AI deployment in healthcare. We will also explore clinical applications of AI in pathological diagnosis and surgical oncology as well as precision oncology applications in biotech, pharma and patient care.
Learning Outcomes
- Explain what precision oncology is and how it differs from traditional cancer care.
- Describe the molecular and genomic basis of cancer: Key genetic, epigenetic, and signaling alterations involved in tumor initiation, progression, and metastasis.
- Understand how genomic and other “omics” data are generated and interpreted and how these data can inform diagnosis, prognosis, and treatment.
- Evaluate the role of advanced molecular diagnostics in the identification of therapeutic vulnerabilities in cancer.
- Evaluate surgical strategies in relation to molecular profiles, critically appraise the use of genomics in surgical decision-making, including risk-reduction surgeries.
- Understand the principles of digital pathology and use of AI tools in modern oncologic pathology workflows.
- Explain fundamental AI and machine learning principles, including regression, loss functions, and optimization.
- Design, optimize, and evaluate machine learning models. Understand and apply deep learning architectures such as CNNs and RNNs for image and biomedical data.
- Analyze and compare classical ML methods with modern deep learning approaches.
- Develop proficiency in the use of publicly available and custom-designed bioinformatics tools for precision oncology applications.
- Recognize the ethical, safety, and regulatory challenges involved in precision oncology.