Lessons Learned in Genomics and Imaging Machine Learning – Forum, October 15, 2021

150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Join the Machine Learning in Medical Imaging Consortium (MaLMIC) for a networking opportunity

Monthly Virtual Forum Series on Zoom

Friday, October 15, 2021
3:30 to 5:30 p.m. EDT

Phedias Diamandis, a neuropathologist and researcher at UHN, discussed the clinical challenges of classifying brain malignancies and the promise of combining ‘omics with digital morphology to help untangle the substantial heterogeneity in these tumours. This was followed by a talk by Wail Ba-Alawi, an affiliate scientist at Princess Margaret Cancer Centre, on using multi-omics techniques for cancer biomarker discovery, and by Anglin Dent, an MSc student at University of Toronto, who expanded on Phedias’ talk by discussing progress in applying deep learning approaches to define biologically distinct subpopulations in glioblastoma. The talks were followed by discussion focused on lessons learned, opportunities to collaborate and sharing of resources.

Monthly Virtual Forum Series on Zoom

Friday, October 15, 2021
3:30 to 5:30 p.m. EDT

Phedias Diamandis, a neuropathologist and researcher at UHN, discussed the clinical challenges of classifying brain malignancies and the promise of combining ‘omics with digital morphology to help untangle the substantial heterogeneity in these tumours. This was followed by a talk by Wail Ba-Alawi, an affiliate scientist at Princess Margaret Cancer Centre, on using multi-omics techniques for cancer biomarker discovery, and by Anglin Dent, an MSc student at University of Toronto, who expanded on Phedias’ talk by discussing progress in applying deep learning approaches to define biologically distinct subpopulations in glioblastoma. The talks were followed by discussion focused on lessons learned, opportunities to collaborate and sharing of resources.

Phedias Diamandis

Integrating morphologic and molecular histopathological features through whole slide image registration and deep learning

Phedias Diamandis, MD, PhD
Neuropathologist, University Health Network
Talk summary: Modern molecular pathology workflows rely on integration of morphologic and immunohistochemical patterns for analysis, but automation is challenged by large files sizes of whole slide images and shifts/rotations in tissue sections introduced during slide preparation. In this talk, Phedias introduced a workflow that couples scale-invariant feature transform and deep learning to efficiently align and integrate histopathological information found across multiple independent studies and highlight utility in molecular subclassification of diffuse gliomas. This efficient pathologist-inspired workflow provides a generalizable approach to help automate a variety of advanced immunohistochemically compatible diagnostic and discovery exercises in surgical pathology

Wail Ba-Alawi

Multi-omics in cancer: biomarker discovery

Wail Ba-Alawi, PhD
Affiliate Scientist, Princess Margaret Cancer Centre
Talk summary: Identifying biomarkers predictive of cancer cells’ response to drug treatment is one of the main challenges in precision oncology. Recent large-scale cancer pharmacogenomic studies have boosted the research for finding predictive biomarkers by profiling thousands of human cancer cell lines at the molecular level and screening them with hundreds of approved drugs and experimental chemical compounds. In this talk, Wail highlighted the virtues and challenges of combining features from different data types in order to find better predictive biomarkers that can be used in the clinic.