Aizon_200x67

Aizon Improves Pharmaceutical Manufacturing With Artificial Intelligence Using AWS

2022

Pharmaceutical manufacturing is a complex and heavily regulated industry. There are hundreds of decisions and variables in each manufacturing process that must be calibrated correctly to deliver safe, effective drugs to patients. As today’s market scales, the demands of production management and process data analysis exceed human capabilities. Aizon, a San Francisco startup, is applying Amazon Web Services (AWS) products to help pharmaceutical manufacturers modernize and reap the benefits of digital manufacturing plants.

Aizon offers a software-as-a-service (SaaS) artificial intelligence (AI) platform aimed at helping pharmaceutical manufacturers optimize their processes and maximize efficiency. One of the biggest barriers for manufacturers attempting to perform the necessary analyses on their own is the inherent messiness of pharmaceutical process data.

“The reality nowadays is that the data scientists would have to spend roughly 80 percent of their time preparing and cleaning the data before they could run AI,” says Toni Manzano, Ph.D., co-founder and chief scientific officer of Aizon. He says Aizon addresses this inefficiency by building data-cleaning into their platform, freeing pharmaceutical customers to focus on their scientific areas of expertise.

AWS HCLS Symposium: Aizon
kr_quotemark

Thanks to AWS, we have greatly improved the scalability of our platform. We can deploy our full platform for customers of every size in just minutes, and we can replicate and escalate it without any problems any country.”

Toni Manzano, PhD
Chief Scientific Officer, Aizon

Applying Artificial Intelligence in a GxP Environment

 Aizon’s platform is uniquely suited to the pharmaceutical and biomanufacturing industries because it is designed to maintain compliance with the set of manufacturing best practices collectively known as GxP. Aizon is an early adopter of the AWS compliance and security framework and built its platform on top of this infrastructure. AWS previously published a whitepaper on GxP compliance and security, with participation from Aizon.

Aizon works with many biopharma companies, many of which perform their computing in-house with on-premises software; however, companies need to embrace cloud technologies to take advantage of “Pharma 4.0” advancements in the industry. Manzano estimates that before turning to AWS, one customer saw as much as 66 percent of the company completely occupied by computing tasks that are now outsourced to AWS.

“We know how to apply AI in GxP environments—that is our focus. Biopharma companies have the knowledge to improve the upstream or downstream processes for making a drug. Our expertise is not in the technology itself,” says Manzano. “Our customers don’t want to worry about encryption, storage, and background processes. AWS is doing all of that very well, so our customers can dedicate all their knowledge to the process itself, to validation, and with Aizon applying AI in this highly regulated industry.”

Manzano estimates that in addition to saving time and adding technical expertise by working with AWS, customers can accomplish their goals while saving approximately 80 percent of costs they would face without help from AWS.

Serverless Bioreactor Applications Built on AWS

One of Aizon’s most unique capabilities is the creation of “digital twins” for different components of biomanufacturing processes. The flagship application creates and operates twins of individual bioreactors, which are the large, fine-tuned machines used to grow cells and harvest biological products for certain pharmaceutical processes.

Using this digital application, individual bioreactors within a pharmaceutical manufacturing facility can be connected to one another and to the cloud via the AWS Internet of Things (AWS IoT), a serverless architecture that can collect, analyze, and store data for billions of devices. When conditions change within a bioreactor, this event triggers AWS Lambda, a serverless compute service, to execute a machine learning operation in Amazon SageMaker, ultimately feeding into process optimization analysis and output.

Manzano highlights AWS Lambda as a key tool that allows Aizon’s platform to function the way it does. Individual bioreactor conditions are already among the most complex problems in pharmaceuticals because they involve living, growing microorganisms; analyzing an entire process that contains multiple bioreactors increases the amount of data exponentially.

“Imagine how many bioreactors a large global pharmaceutical company has around the world—thousands upon thousands,” says Manzano. “For every one of those to be connected to the cloud, producing values, calculating predictions, detecting anomalies—all of this data is impossible to manage without a serverless architecture."

Aizon’s Platform in Action

A multinational pharmaceutical company producing promising drugs based on human blood plasma experienced an unexplained decrease in product yield coupled with an unwelcome increase in process variability over several years. To determine the root cause of these problems, the company turned to Aizon for help. Using its AI-enabled, GxP-compliant platform built on AWS cloud architecture, Aizon created an unsupervised learning model to identify patterns among plasma composition origins in thousands of the pharmaceutical company’s drug batches over the years. Aizon also created a supervised learning model to predict yield for each upcoming batch and coupled that with real-time dashboarding.

Aizon’s analyses revealed that the plasma composition origin explained over half of the process variance and that batch clusterization was critical to yield prediction. Furthermore, the company identified two critical process variables and found that optimizing just those variables was sufficient to improve the yield by double digits, resulting in potential gains of tens of millions of dollars for the pharmaceutical customer.

Aizon looks forward to helping more and more pharmaceutical companies realize similar benefits. “Thanks to AWS, we have greatly improved the scalability of our platform,” says Manzano. “We can deploy our full platform for customers of every size in just minutes, and we can replicate and escalate it without any problems in any country.” Aizon plans to continue expanding its software services while working with regulatory agencies around the world to establish standards for applying AI in pharmaceutical manufacturing.

Learn More

Learn more about how organizations like Aizon leverage AWS for Life Sciences to fuel innovations. 


About Aizon

Aizon provides platforms for managing and optimizing complex manufacturing processes in the pharmaceutical industry. Its solution enables more sophisticated control and analysis of bioprocesses all the way from research and development to drug production.

Benefits of AWS

● Delivered GxP-compliant data analytics platform
● Reduced compute costs by up to 80 percent
● Identified process improvements that could boost yield by over 10 percent
● Connected complex manufacturing equipment and processes to the cloud
● Enabled international platform deployment and scalability
● Performed process analysis and optimization to increase product yield
 


AWS Services Used

Amazon Lambda

AWS Lambda is a serverless, event-driven compute service that lets you run code for virtually any type of application or backend service without provisioning or managing servers. 

Learn more »

Amazon SageMaker

Build, train, and deploy machine learning (ML) models for any use case with fully managed infrastructure, tools, and workflows

Learn more »

AWS IoT

AWS offers Internet of Things (IoT) services and solutions to connect and manage billions of devices. Collect, store, and analyze IoT data for industrial, consumer, commercial, and automotive workloads.

Learn more »


Get Started

Organizations of all sizes across all industries are transforming their businesses and delivering on their missions every day using AWS. Contact our experts and start your own AWS journey today.