AACR Highlights: Focus on AI

The 110th Annual Meeting of the American Association for Cancer Research launched on March 30th, 2019 in Atlanta, GA. Over 22,000 delegates attended including scientists, clinicians, and patient advocates from around the globe. A team from Reveal Biosciences was present to highlight our recent advances in the application of artificial intelligence (AI) in pathology for preclinical research, clinical trials, and decision support tools. Our team was also interested to learn how AI is being applied to other areas of cancer research including patient stratification, prognostics, and clinical trial facilitation. Immunotherapy also continued to be a hot topic as researchers delve into combination therapies and how to convert non-responders into responders.

Dr. John Carpten, Chair of the Department of Translational Genomics at USC, chaired a compelling opening plenary session “Achieving equitable patient care through precision and convergent cancer science” to a packed house. Concluding the session with a discussion about opportunities for the future, he discussed how he is investigating tumor heterogeneity by combining histopathological analysis with spatial transcriptomics to assess the tumor microenvironment. Collaborating with 10X Genomics for whole transcriptome sequencing of cell clusters in a way that preserves spatial information, he profiled immune clusters, tumor clusters, and immune effector cells, looking at tumor inflammation at the single gene level and revealing the differences between hot and cold regions of a tumor.

AI was an emerging topic. Three sessions were devoted exclusively to AI along with many posters. Dr. Regina Barzilay of MIT spoke about “Deep learning for reading mammograms: state-of-the-art and clinical implementation” during the session on “Convergence of Artificial Intelligence, Big Data, and Prediction in Cancer”. She discussed the accuracy of current state of the art AI models and examined some of the current challenges such as clinical site-to-site image variation and enriching image data with additional patient data. Posters showcased applications of AI to genomic analysis, biomarker discovery, patient stratification, and histopathology image analysis.

The Reveal Biosciences team was excited to be in the middle of the action with a booth where we announced our Series A investment led by Intel Capital and unveiled our new logo and Data Powered Pathology branding. We met with many new clients and collaborators and attracted significant interest in our new service offerings including ImageDx, a cloud-based image management platform, automated image analysis for quantitative tumor profiling, and cancer prognostics. Contact us to learn more: [email protected]

Looking forward to AACR 2020 in San Diego, CA!