MTG Cuts Medical Research Times in Half, Secures Patient Data Using AWS
Portugal’s MTG Research and Development Lab developed a hybrid solution using AWS that supports healthcare providers conducting pioneering medical research. The solution helps hospitals and clinics process and analyze large amounts of data required for studies, while ensuring compliance with data protection laws for sensitive patient information. Using AWS Snowball Edge devices, MTG cut the time required to complete studies in half and lowered the total IT costs for hospitals conducting research.
Using AWS, we’ve sped up medical research dramatically. We’ve also freed up time for our teams to focus on deepening studies so they can impact game-changing public health policies.”
Founder and CEO, MTG Research and Development Lab
Medical research that improves disease prevention and treatment requires two elements: access to patient records and significant amounts of processing power to analyze and derive insights from this data.
Bringing these together is challenging for healthcare providers. Patient data is subject to strict privacy regulations that frequently require it to be stored on-site at a healthcare institution. The amount of processing power needed can be expensive and hard to achieve without a dedicated on-premises data center, which most healthcare providers don’t have the resources to run.
MTG Research and Development Lab, a Portugal-based contract research organization that specializes in real-world evidence generation, has developed a hybrid solution using Amazon Web Services (AWS) that supports healthcare providers conducting ground-breaking research.
The solution runs on AWS Snowball Edge devices, which provide petabyte-scale data transport with on-board storage and compute capabilities. This means AWS services and customer workloads run locally at hospitals’ premises, so researchers can analyze data while complying with regulations such as the General Data Protection Regulation (GDPR). Using MTG’s solution, researchers carry out large-scale scientific studies in half the amount of time and at a lower cost, compared to previous options.
Hybrid Solution Provides Power to Analyze Complex Medical Data
Founded in 2015, MTG develops data science pipelines and analytic code for healthcare providers. This provides medical researchers with a deep understanding of diseases and physical injuries that can improve healthcare quality. The approach requires multiple data sources, including ethnic and geographic origins, as well as information on patients’ pre-existing conditions.
MTG provides hospitals with affordable access to the compute power necessary to process all of this information. “Our studies involve heavy data processing requiring hardware that most healthcare providers do not have,” says Tiago Taveira-Gomes, founder and chief executive officer (CEO) at MTG. “Hospital IT staff are usually busy and even more overloaded because of the COVID-19 pandemic, so it is essential to provide a low-cost and easy-to-deploy solution that also does not increase the workload for IT staff.”
The hybrid solution combines analytic code developed by MTG and synthetically-generated data on AWS using Amazon Elastic Map Reduce (EMR), a cloud-based big data platform for running large-scale distributed data processing jobs. This code is then re-used on an Apache Spark cluster running on the Snowball Edge devices using Amazon Elastic Compute Cloud (Amazon EC2), which provides secure and resizable compute capacity to process the data on-premises.
Ensuring Compliance with Data Protection Laws
Healthcare data in Portugal is subject to the GDPR, a regulatory framework for the European Union (EU) that poses constraints on data access and the sharing of electronic medical information. The regulatory burden around GDPR has limited the ability of EU scientists to collaborate and conduct research.
To ensure hospitals are fully compliant with this and other regulations, data for all studies using MTG’s solution is analyzed within the healthcare provider’s private network. MTG aggregates and anonymizes the data and then exports it to AWS Regions. To visualize these aggregate results, MTG uses Amazon QuickSight, the serverless business intelligence service that supports a better understanding of data using interactive dashboards. Once the research study is complete, the hospitals can erase the data from the AWS Snowball Edge devices and return them to AWS.
On-Demand, Portable AWS Devices Cut Study Times in Half
Using MTG’s solution, healthcare providers can carry out research faster and at a lower cost compared to an on-premises data center. The cost savings come from using on-demand services and only paying for the compute power used—an ideal setup for studies into chronic diseases that are carried out only a few times a year. Running data analysis on the portable Snowball Edge devices means that MTG can support multiple studies at different hospital sites in parallel. This has cut the average time to complete research projects in half, from 4 months to 2.
The Snowball Edge devices provide hospitals with portable processing power to carry out complex data analysis and are easy for IT teams to manage. “It’s like having a mini data center in a box,” says Taveira-Gomes. “Once the devices arrive on-site, they can be set up and running in less than an hour, so it’s not a big job for our team nor for the healthcare provider. Within a few hours after setup, we have results ready to share with the researchers and clinicians at the healthcare institution.”
MTG offers a secure and efficient solution to process data needed for essential research. “It used to take around 6 months just to get project approval because of the lengthy data protection risk assessment healthcare providers had to carry out,” says Taveira-Gomes. “Now, because data is not handled by third parties, our customers skip this approval step and go right to carrying out research, so they can begin making discoveries to improve disease prevention and treatment.”
MTG’s hybrid approach was used for Portugal’s first large-scale observational study involving electronic patient records to estimate the prevalence of type 2 diabetes, heart failure, and chronic kidney disease.
MTG now derives insights into data that had been almost impossible to access before. “The deep analytics our customers produce contributes to the development of more effective health policies and treatment guidelines, improving quality of care and ultimately averting the progression of chronic disease,” says Taveira-Gomes.
A Scalable Solution That Supports Future Growth
Six institutions in Portugal and other European countries use MTG’s solutions. The company aims to expand its customer base and offer its solution to more hospitals. Using AWS, the company can quickly spin up compute resources for new customers. “The ability to grow is especially important for us,” says Taveira-Gomes. “With our solution, there are no fixed costs or big upfront investments needed—the only overhead is ordering the Snowball Edge devices.”
If MTG supports studies that require large amounts of compute capacity for extended periods of time, it plans to use AWS Outposts, which delivers AWS infrastructure and services to virtually any on-premises or edge location.
The company now supports federated studies that combine research from multiple healthcare institutions and can have a major impact on our understanding of disease. To help researchers carry out these studies, MTG builds secure machine-learning models. “We can easily train models to learn from multiple sources without data leaving a healthcare institution’s premises,” says Taveira-Gomes. “Using AWS, we’ve sped up medical research dramatically. We’ve also freed up time for our team to focus on deepening studies so they can impact game-changing public health policies.”
Based in Portugal, MTG Research and Development Lab is a contract research organization that specializes in real-world evidence generation in the healthcare domain. It aims to improve the understanding of diseases, trauma, and the quality of healthcare. Its research lab covers every aspect of the study lifecycle and works with worldwide institutions in federated research programs.
Benefits of AWS
- Reduced time to complete medical research studies by 50%
- Cut costs on compute resources by using on-demand services
- Supported medical research that improves disease prevention and treatment
- Complied with data privacy regulations such as the GDPR
AWS Services Used
AWS Snowball Edge
With an AWS Snowball Edge device, you can access the storage and compute power of the AWS Cloud locally and cost effectively in places where connecting to the internet might not be an option.
Amazon Elastic Compute Cloud (Amazon EC2)
Amazon Elastic Compute Cloud (Amazon EC2) offers the broadest and deepest compute platform, with over 500 instances and choice of the latest processor, storage, networking, operating system, and purchase model to help you best match the needs of your workload.
Amazon EMR is a cloud big data platform for running large-scale distributed data processing jobs, interactive SQL queries, and machine learning (ML) applications using open-source analytics frameworks.
Amazon QuickSight allows everyone in your organization to understand your data by asking questions in natural language, exploring through interactive dashboards, or automatically looking for patterns and outliers powered by machine learning.