Artificial intelligence tools for population-based brain metastasis surveillance post cancer diagnosis 

While brain metastasis (BM) exceeds primary brain tumours and comprises the majority workload in neuro-oncology, there is a lack of population-based estimate of BM incidence rate post cancer diagnosis. In our recent study, we developed and validated artificial intelligence tools, specifically natural language processing algorithms (NLP) based on the BioBERT model, to label radiology reports for BM presence. The NLP algorithms were used to predict the BM labels in a population-based cancer cohort in Alberta who were diagnosed between 2012 and 2020 with follow up till 2022. We then estimated the cumulative incidence rate during a 10-year time frame post-diagnosis. In this webinar, I will present the process of training and testing of our NLP algorithms and their prediction performance. The resulting incidence rate estimates obtained by applying our NLP algorithms will be discussed. 

Speaker bio:

Dr. Yuan, professor at the University of Alberta’s School of Public Health (2011- present) received an undergraduate degree from the Nanjing University, China in 1996, followed by a MSc. from the Michigan State University, USA, and a MMath in Biostatistics & a PhD in Statistics both from the University of Waterloo, Canada in 2008.

Dr. Yuan’s academic career focuses on cancer research. She led and collaborated in projects ranging from prevention, surveillance, diagnostic and screening, health service and survivorship with over 70 publications, of which fifteen are on the brain tumours surveillance in Canada. A recipient of the Early Career Award from the Canadian Institute of Cancer Research in 2017, Dr. Yuan received major research grants, including Canadian cancer society, Canadian Institutes of Health Research, Natural Sciences and Engineering Research Council of Canada and National institutes of Health (US).

Register