Deep Learning for Digital Tissue Biomarker Discovery in Oncology

A talk by Günter Schmidt
Vice President, Image Data Sciences, AstraZeneca

Register to watch this content

By submitting your email you agree to the Terms of Service and Privacy Statement
Watch this content now

Stages covered by this talk

About this talk

Artificial Intelligence (AI) systems have surpassed human intelligence, fueled by the availability of dedicated high performance hardware, sophisticated deep learning architectures and algorithms, and Big Data. Deep convolutional neural networks have been successfully used to generate quantitative insights into the cancer immune contexture, and to replicate pathologists diagnostic scoring of tissue slides. However, the discovery of prognostic and predictive biomarkers for oncology using computational pathology remains challenging. Standard fully supervised deep learning methods require extensive training data from expert pathologists. Latest weakly supervised methods aim towards superhuman performance, but frequently lack interpretability and transparency. Although regulatory authorities such as the FDA are increasingly open to approve AI “software as a medical device” systems, no histopathology-based companion diagnostic test has been yet marketed. The recent CE marks for prostate and breast cancer detection by AI-based clinical grade systems are promising milestones towards a wider adoption of computational pathology in the clinical routine, for the benefit of our patients.

Have you got yours yet?

Our All-Access Passes are a must if you want to get the most out of this event.

Check them out

Learn from amazing companies like these

AstraZeneca

Proudly supported by

Want to sponsor? Contact us to find out more.