Deloitte Cuts Deployment Times, Helps Clients Reduce Process Run Times Using AWS

Operating in more than 150 countries and territories, Deloitte is a network of independent firms providing audit and assurance, tax, legal, risk and financial advisory, and consulting services to a wide range of clients in 20 industry sectors. In the U.S., Deloitte LLP and Deloitte USA LLP serve more than 85 percent of the Fortune 500 and more than 6,000 private and middle-market companies.

Personalized medicine promises to revolutionize the treatment of cancer and other serious diseases. Increasingly, companies are using real-world data (RWD) to speed personalized drug discovery while improving patient safety. RWD includes electronic health records (EHRs), insurance claims, registries of products and diseases, and information collected from health-monitoring devices, which can be combined with data from clinical trials.

Organizations analyze RWD to create real-world evidence (RWE), which illuminates patterns of therapeutic usage, risks, and benefits. Additionally, the U.S. 21st Century Cures Act of 2016 directs the U.S. Food and Drug Administration to evaluate the potential of RWE to support the approval of new indications for previously approved drugs.

RWE can help researchers generate and test hypotheses about connections between genetics and disease, identify new therapeutic molecules, gain insights into disease mechanisms, and work together more effectively by sharing data.

Before these benefits can be achieved, vast and disparate data sets must be combined and analyzed. Deloitte helps solve these challenges through its ConvergeHEALTH services and software. Its ConvergeHEALTH Miner solution is a suite of tools and services that help accelerate data gathering, analysis, and management across the research lifecycle.

Deloitte clients have achieved significant results using Miner. One client dramatically improved time-to-market for new therapies and reduced RWD process run times by 99 percent while cutting related operating costs 70 percent. Another accelerated the identification of patient populations for clinical studies, resulting in getting medicine to patients faster, and was able to expand its safety and effectiveness analysis to improve patient outcomes.

Results like these require an agile, scalable platform. That’s why Deloitte uses Amazon Web Services (AWS). “Using AWS as our platform provides the large-scale, elastic storage and computing power required to develop RWE,” says Jinlei Liu, vice president of product development at Deloitte. “It enables clients to unify data in one place for advanced analysis and actionable insights into patient results. It also helps us meet HIPAA and other compliance requirements.”

Miner leverages a wide range of AWS services. Amazon Elastic Compute Cloud (Amazon EC2) provides compute scalability. For object storage, Miner uses Amazon Simple Storage Service (Amazon S3). Database functions are handled by Amazon Relational Database Service (Amazon RDS), with advanced extract, transform, and load (ETL) functionality handled by AWS Glue. Logical environments are provisioned and isolated using Amazon Virtual Private Cloud (Amazon VPC).

In addition to faster deployment, using AWS Service Catalog helps clients cut costs. “Customers can easily turn instances off when they are not being used and then turn them back on again instantly when they are needed,” says Kristin Feeney, senior data scientist at Deloitte. “Simply by turning them off on nights and weekends can save clients about a third of associated cloud compute costs.”

Deloitte can control who has access to the software. “With AWS Service Catalog, we have total control over the organizations who can access our software and which packages we offer them, enabling us to meet compliance requirements,” reports Liu.

Using infrastructure as code also benefits Deloitte experts. “We have a lot of internal data science teams who need RWD capabilities,” says Kristin Feeney, senior data scientist at Deloitte. “With AWS Service Catalog, we can deploy these advanced tools in hours or days instead of weeks, without incurring additional IT costs or waiting for approvals and provisioning as we would in a traditional data center. The simplicity of deploying Miner using AWS Service Catalog also reduces our costs and our reliance on third-party tools.”

Most importantly, deploying via AWS Service Catalog makes Miner more accessible to customers and enables them to use the platform more broadly. “Previously, data scientists were typically limited to the software they already had deployed,” says Liu. “Using AWS Service Catalog, they can now spin up an instance of Miner, run the data, and shut it down again to meet short-term analytics needs,” says Liu. “They don’t need an IT team to set it up. That helps them move faster and discover medical insights sooner.”