null Bridging the gap between AI and the clinic
Open source app helps predict brain tumour malignancy and patient survival
Jan. 30, 2020
Source: The Neuro. The power of artificial intelligence in medicine lies in its ability to find important statistical patterns in large datasets. A study published today is an important proof of concept for how AI can help doctors and brain tumour patients make better treatment decisions. Meningiomas – tumours that arise from the membranes that surround the brain and spinal cord – are the most common primary central nervous system tumour, with an incidence of 8.14 per 100,000 population.
In this study, researchers from The Neuro and the Montreal Children’s Hospital of the McGill University Health Centre trained machine learning algorithms on data from more than 62,000 patients with a meningioma. Their goal was to find statistical associations between malignancy, survival, and a series of basic clinical variables including tumour size, tumour location, and surgical procedure.
Jeremy Moreau, a PhD candidate and trainee at the Research Institute of the McGill University Health Centre (RI-MUHC), is the study’s first author. He is co-supervised by Sylvain Baillet, a neuroscientist at The Neuro, and Dr. Roy Dudley, a clinician-scientist at The Montreal Children’s Hospital and the RI-MUHC. Read more.