In this proposal, one of the two focuses is to leverage recent advances in large language models (e.g., BERT, T5, GPT-2) in Natural Language Processing (NLP) to extract valuable information from unstructured documents, which include clinical documents and user-generated content. The second is on machine learning using medical imaging data. Our aim is to develop robust privacy-preserving diagnostic and prognostic models that are explainable.

Ovarian cancer is one of the deadliest gynaecological conditions in the developed world, as it is often detected late into progression. Efforts to improve surgical and chemotherapeutic approaches have only made marginal improvements to patient outcomes over the past 30 years, but there remains potential within immunotherapeutic approaches. Understanding the tumor microenvironment, and particularly the immune response to cancers and pre-cancers, is critical to selecting appropriate therapies.

British Columbia’s Ministry of Mental Health and Addictions has identified healthy social and emotional development (SED) in early childhood as a priority action towards addressing the province’s mental health crisis. In response, ChildHealth BC (CHBC) is developing a multi-component provincial intervention to expand the capacity of caregivers to promote SED in young children- the first to be implemented at scale in Canada. While there is evidence on the efficacy of capacity-building SED strategies, research on how to bring these strategies to scale is lacking in the field.