Bringing Clinical Data to GCP for Automated AI Discoveries (Cloud Next '18)

Опубликовано 25 июля 2018, 21:36

In this session, we will present how the Imagia Deep Radiomics platform contributes to the discovery of imaging biomarkers for anomaly detection, disease progression and treatment response. We will outline preliminary steps such as integration, data ingestion and storage, as well as specific regulatory compliance mechanisms ensuring patient information traceability, controlled accessibility and revocability using a combination of Google products including Google PubSub, Google Cloud Storage and BigTable.

Further, we will illustrate how Imagia Deep Radiomics can make outcome-oriented predictions using a combination of automated image feature learning through models in ML Engine and Kubernetes Engine, and statistical selection in Datalab backed by BigQuery. We will show how we develop, test, deploy and scale our platform with the help of Kubernetes.

Finally, we will demonstrate how clinical researchers can undertake different discovery processes leading to the design of robust radiomics biomarkers for personalized healthcare, without the need of AI expertise.

Event schedule → g.co/next18

Watch more Machine Learning & AI sessions here → bit.ly/2zGKfcg
Next ‘18 All Sessions playlist → bit.ly/Allsessions

Subscribe to the Google Cloud channel! → bit.ly/NextSub


re_ty: Publish; product: Cloud - General; event: Google Cloud Next 2018;