Artera Shapes the Future of Cancer Treatment Using Machine Learning on AWS

Artera was founded in 2021 with the goal of using AI to globally personalize medical decisions and improve outcomes for patients. The company used AWS from the beginning to power its AI and ML solutions with high-performance, scalable infrastructure.

Before Artera’s solution, there were no AI-based tools to help physicians and cancer patients make personalized, timely treatment decisions. Instead, physicians submitted a patient’s biopsy tissue sample to a lab, where a chemical assay measured the expression levels of a small set of genes. The RNA expression of these genes was then used to assess a patient’s risk level. These tests have several limitations. First, the entire process can take 6 weeks—a long time to wait on making a high-stress decision about cancer therapy. Second, these tests only improve as fast as science can identify new genes linked to cancer risk. Third, these tests consume the original tissue samples, limiting the physician’s ability to order additional tests, as well as the patient’s ability to enroll in future clinical trials or participate in long-term monitoring.

Artera wanted to change that. “Artera was founded with the belief that there were a lot of signals in the histopathology image data that were not being used, but if an AI algorithm could be specifically developed with this mind, you could radically change cancer patient care,” says Nathan Silberman, chief technology officer of Artera. The company wanted to use that information to develop a solution that would make the biggest impact on patients. While it focused initially on intermediate-risk prostate cancer, the company’s ultimate goal has been a massive pan-tumor foundation model capable of assessing patient risk and therapy benefit across any cancer sample.

Artera’s clinical and ML teams collaborated to define the project’s infrastructure requirements. Because Artera’s solution would involve analyzing signals in biopsy slides, it knew it would need to process high-resolution digital images. However, these image files are enormous, sometimes reaching 8 GB, and would need to be broken down into tens of thousands of smaller patches for the model to handle—a time-consuming process. Furthermore, training Artera’s foundation models requires serving millions of images patches at high volume to AWS servers.

For this task, Artera needed a storage system solution that was simple to use but could effortlessly scale to manage its millions of image files and serve them at a moment’s notice. The company chose Amazon Elastic File System (Amazon EFS) for serverless, fully elastic file storage.

Amazon EFS’s ability to scale on demand makes it possible for Artera to rapidly develop software and AI products. Because EFS operates like a conventional file system, Artera’s software could be written agnostic of storage location, while its high throughput supported concurrent experiments. With no need for storage administration and seamless integration with Artera’s other tools, Amazon EFS provided the performance and consistency needed for ML workloads. “We wanted to provide an intuitive way for teams to navigate a shared cloud-based file system so that they can rapidly experiment,” says Silberman. “Amazon EFS was an excellent solution.”

To support large-scale data processing and ML training—and, in turn, to relay test results to patients more rapidly—Artera chose Amazon Elastic Kubernetes Service (Amazon EKS), a fully managed Kubernetes service. Amazon EKS automates the Kubernetes cluster management that powers Artera’s ML workflows, freeing up engineers to focus on developing algorithms rather than managing infrastructure.

With its solution concept solidified and the framework in place, Artera developed its first commercial product: the ArteraAI Prostate Test.

The ArteraAI Prostate Test is considered the standard of care for localized prostate cancer, being included in the 2024 National Comprehensive Cancer Network Clinical Practice Guidelines in Oncology. Powered by AWS, this test analyzes high-resolution digital slide images using Artera’s AI models, identifying biomarkers that it uses to predict patient risk and treatment outcomes.

Artera is able to scale its workloads with high throughput on Amazon EFS, using Amazon EKS for large-scale parallel data processing. This high performance and scalability make it possible for Artera to deliver results just 2–3 days after receiving the specimen—far faster than the 6-week turnaround of traditional genomic tests, which can return inconclusive results due to insufficient tissue. The ArteraAI Prostate Test results support clinical decision-making through patient risk stratification by estimating long-term outcomes and predicting different therapy benefits. “For other tests, physicians send the patient’s tissue to a lab that applies a chemical assay, which destroys the tissue,” says Silberman. “The fact that ArteraAI’s Prostate Test can be run using just digital imagery is a huge differentiator.”

Using AWS managed services and scalable solutions like Amazon EFS, Artera’s teams are freed up to focus on innovation. “Using AWS, we can focus on the histopathology problems rather than on maintenance and monitoring,” says Silberman. “There’s no question that using AWS has rapidly accelerated our time to market.”

Additionally, by using AWS, Artera stays compliant with data privacy and data residency requirements as it expands globally—helping to accelerate time to market in new regions. Amazon EFS allows users to mount file systems in the same AWS Region as their applications, so data stays close to where it’s processed. “One of the nice things about Amazon EFS is that it’s very simple to achieve data locality,” says Silberman.

In 2024, almost 300,000 Americans were diagnosed with prostate cancer. “Imagine a patient getting the worst news they’ve ever had and having to sit on that for 6 weeks to determine what the treatment plan is,” says Silberman. “Instead, Artera provides custom-tailored, personalized results within days.”

There are over 3.5 million prostate cancer survivors in the United States. By recommending personalized treatment plans, Artera is helping patients determine the best therapeutic options to achieve progression-free survival while minimizing unnecessary side effects. “We’ve heard from patients who have said that because of our test, they were able to avoid unnecessary treatments with a lot of side effects,” says Silberman. “That’s why all of us at Artera are here, giving clinicians as many data-backed insights as possible to inform the patient and make the best possible choice for their care.”

Using elastic, scalable solutions like Amazon EFS, Artera has a solid foundation for continuing to develop ML models for additional cancer tests. The company has announced plans for a breast cancer product, with several more products close behind. “What we have coming up is a rapid acceleration across different areas of cancer,” says Silberman. “As proud as we are of the work that we’ve done in the prostate cancer space, we’re just getting started.”