What challenges did Keck School of Medicine of USC face before AI implementation?

Manual Medicare Coverage Analysis (MCA) in clinical trial budgeting was slow, prone to errors, and a bottleneck delaying trial activation and completion. Administrative burdens, disparate systems, and lack of process standards impaired timely activation of clinical trials.

What was the impact of the AI solution?

Reduced MCA process time from days to milliseconds, significantly accelerating clinical trial activation. Cut clinical trial activation times by 50%, enabling faster patient enrollment and drug development. Enhanced operational efficiency allowed staff and budget to be reallocated to other research priorities. Supports approximately 200 clinical trials annually with improved activation speed and data-driven analytics for future optimization.

Source

GCPProductionEvidence: Medium65/100

The Keck School of Medicine of USC Accelerates Clinical Trials with Google Cloud Machine Learning

Use case typeHealthcare workflow automationUpdated Jun 13, 2026

The Keck School of Medicine of USC, part of the University of Southern California, faced slow, manual, and error-prone Medicare Coverage Analysis (MCA) processes causing delays in clinical trial activation and completion. The school collaborated with Google Cloud and partner Pluto7 to implement machine learning models that automated complex decision-making workflows in clinical trial budgeting and billing. The ML system reads standard care guidelines and predicts billing designations with 70-90% accuracy, accelerating the MCA budgeting process from days to milliseconds. This automation shortened clinical trial activation times by 50% and improved efficiency in managing approximately 200 annual clinical trials, freeing up staff and budget. The use of Google Cloud serverless infrastructure and BigQuery enables scalable analytics and ongoing enhancements with ML.

Organization: Keck School of Medicine of USC
Industry: Healthcare
Location: United States
Published: May 2026

Reported outcomes

−50%

timeTime & speed

Strategic outcomes

Speed & agilityAccelerated clinical trial activationNew product / capabilityAutomated trial budgeting decisionsScale & capacitySupported higher trial volumeCost efficiencyFreed staff and budget resources

Catalog median for time & speed deployments: −50% across 312 reported metrics. Compare benchmarks →

Why do we believe this?Outcome claims, sources, and evidence checks

Normalized claim

Time: 50% decrease

Google Cloud Customer StoriesMay 10, 2026Customer storyInferred claimMedium evidence strength

Cut clinical trial activation times by 50%, enabling faster patient enrollment and drug development.

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: Keck School of Medicine of USC
Provider: GCP
Maturity: Production
Linked source: Google Cloud Customer Stories

Deployed solution on Google Cloud serverless infrastructure and leveraged BigQuery for large-scale data analytics

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 2 of 2

1Workflow Automation
2Machine Learning

Manual Medicare Coverage Analysis (MCA) in clinical trial budgeting was slow, prone to errors, and a bottleneck delaying trial activation and completion.
Administrative burdens, disparate systems, and lack of process standards impaired timely activation of clinical trials.

Implemented machine learning models using Google Cloud AI and analytics to automate decision processes in clinical trial budgeting and billing designations.
Built algorithms to interpret standard care guidelines and accurately assign billing categories for study procedures within milliseconds.
Deployed solution on Google Cloud serverless infrastructure and leveraged BigQuery for large-scale data analytics.

Technologies

Google Cloud AI Machine Learning BigQuery Serverless infrastructure

Reduced MCA process time from days to milliseconds, significantly accelerating clinical trial activation.
Cut clinical trial activation times by 50%, enabling faster patient enrollment and drug development.
Enhanced operational efficiency allowed staff and budget to be reallocated to other research priorities.
Supports approximately 200 clinical trials annually with improved activation speed and data-driven analytics for future optimization.

Architecture

The architecture includes machine learning models trained collaboratively with USC MCA experts, deployed on Google Cloud serverless infrastructure, utilizing BigQuery for data analytics and workflow automation.

Sources & evidence1

Evidence: Medium65/100Evidence strength