Using FastTrack AI for a clinical indication in liver cirrhosis


Great new paper from Matthias Gunzer’s lab with another example using FastTrack AI for a clinical indication.

Link to paper in Liver International


We are using AI/Deep Learning in a new way: we apply it to microscopy images of living cells. Looking at living cells is a lost art. But with the advent of AI, we are now able to teach the computer to watch the behavior of tens of thousand of living cells over time.  This has opened the door wide open to new discoveries.

Previous implementations of AI have focused on big data applications. MetaVi Labs’ approach is different, we apply AI to the problem of learning cell morphology and behavior relative to other cells. 

Since a major function of immune cells is to travel to the site of the pathogens, Matthias Gunzer had the insight that immune cells migration behavior is an indication of health or illness in the system. In several clinical studies he has demonstrated this.

Matthias’ lab extracts immune cells from the patient’s blood samples, prepares culture samples treated with activator substances, and then places them under the microscope and records cells continuously at high speed over a period of hours. 

MetaVi Labs has provided a system where the lab staff can simply upload the recorded images to our on-line analysis system. A few minutes later the system provides the lab with the test results. Our staff and Matthias’ team worked together closely over a period of several years to customize the analysis and reports for his applications.

This is a very exciting new development as it opens an entirely new arena of new diagnostics based on peripheral blood.