Case Study: Johns Hopkins University BIOS Division achieves faster, more accurate brain scan insights with Google Cloud Platform

Advancing intracerebral hemorrhage treatments through AI

Johns Hopkins University BIOS Division needed a faster, more accurate way to analyze brain CT scans and medical images for stroke and brain hemorrhage research. Working with Google Cloud Platform, the team moved beyond a manual, time-consuming review process to a cloud-based, data-science-driven approach.

Google Cloud Platform, with partner Quantiphi, implemented machine learning and cloud analytics using services such as Compute Engine, Cloud Dataflow, Cloud Healthcare API, AI Platform, and Google Kubernetes Engine. The result was a dramatic reduction in scan review time from five hours to 30 seconds per scan, with insights from about 500 patients generated in 90 minutes instead of 2,500 hours, improved accuracy with a dice coefficient of 0.93, and lower research and infrastructure costs.

Johns Hopkins University BIOS

Daniel F. Hanley

Junior Director