Bio-Rad Explores AWS IoT Greengrass and Amazon SageMaker for Next-Gen Medical Devices

The Clinical Diagnostics Group at Bio-Rad offers a portfolio of solutions for the diagnosis, monitoring, and treatment of diseases ranging from diabetes to infectious diseases. The team’s new solution tests and analyzes patient samples.
 
The solution effectively supports the transfer of massive data payloads ranging up to multiple gigabytes, which required the establishment of a centralized data storage system. Additionally, the team needed Internet of Things (IoT) capabilities to facilitate the automatic and programmatic upload of data—including high-volume instrument data and low-volume system telemetry data—to a centralized hub in the cloud. The solution had to adhere to high levels of security mandated by HIPAA and other privacy regulations. “It was essential to centralize data collection in the cloud securely from all connected labs,” says Jasmina Desai, the associate director of Bio-Rad’s Clinical Diagnostics Group. “This approach makes it possible for everyone to log in, compare, and analyze data in a single location, promoting global collaboration across labs.”
 
In addition, Bio-Rad also sought to use ML and automation to reduce manual intervention and expedite data processing. “Our current process methods of data analytics were time consuming and lacked standardization,” says Jasmina. The company wanted a scalable and secure solution to automate the whole process while maintaining data integrity.

Using a range of AWS services, Bio-Rad has built a cutting-edge solution for secure and seamless transfer of instrument data to the cloud. Built on the foundation of AWS IoT Greengrass, the robust and scalable infrastructure deftly manages the upload of high-volume data, without inadvertently affecting the functionality of Bio-Rad’s edge devices. “The solution virtually guarantees data security so that no data is lost during the transfer, even if there’s a connectivity drop in the lab,” says Jasmina. “AWS IoT Greengrass has been a true blessing to us. It has a small footprint, and we know the incoming data is safe.”

The incoming data undergoes minimal transformation using AWS Glue, a serverless data integration service that makes it easier to discover, prepare, move, and integrate data from multiple sources for analytics and ML. Next, the data flows through an automated pipeline of ML models for data processing using Amazon SageMaker Notebooks, fully managed notebooks for exploring data and building ML models. This approach significantly reduces the need for manual intervention, leading to enhanced efficiency in data analysis.

A single ML model might analyze up to 50 GB of data, making it impractical to run on a data scientist’s laptop. To address this, Bio-Rad uses the Amazon SageMaker console to automatically calculate needed resources, virtually eliminating the need for manual setup and maintenance of powerful servers. “Using AWS, we can concentrate on enhancing the model without worrying about notebook or server requirements,” says Jasmina. “This has simplified our lives significantly. If we had to manage this independently, it would entail substantial effort to support the cloud deployment.”

AWS IoT Greengrass is also used to reduce potential errors in the deployment of ML models, such as inadvertently uploading the wrong model. Instead, data scientists know that they are always using the most up-to-date version of ML models.

AWS also provides Bio-Rad with flexible storage, adequate compute, and robust security with which it can store, analyze, and protect the sensitive data. The data is securely stored in Amazon Simple Storage Service (Amazon S3), an object storage service offering industry-leading scalability, data availability, security, and performance. The solution is on par with the highest security standards using AWS Key Management Service (AWS KMS), which organizations can use to create, manage, and control cryptographic keys across applications and AWS services. Using AWS KMS, data scientists focus on analysis without having to maintain their own keys.

Because AWS IoT Greengrass and Amazon SageMaker were new to Bio-Rad, the Clinical Diagnostics Group turned to AWS for additional support as it developed the smart device. AWS conducted demonstrations, workshops, and AWS Immersion Days, sessions delivered by AWS solutions architects and subject matter experts in AWS solutions. “We wouldn’t be able to do it this fast without the support that was provided by the AWS team,” says Jasmina. “They have gone over and beyond as we use AWS services to make this project a success.”

Bio-Rad is pushing forward and bringing a paradigm shift in the industry while upholding the utmost standards of security and compliance. “In the realm of a connected world, we embrace the transformative potential of AWS,” says Jasmina. “Scaling up independently can be time consuming, but using AWS, we can scale up efficiently and effectively.”