Use Case Intelligence RSS Search use cases Search
AWS Healthcare United States

Natera

Document Processing

In the case of Brent, a young husband and father sought care following a mountain bike accident that caused pain and discomfort.

Read the original story Explore all use cases
Natera nearly doubled data extraction, from 25 to 46 elements per test, enabling more accurate patient care, and it achieved a 150% cost reduction in data processing.
Natera’s identified over 200 potential use cases for AI implementation with personalized healthcare delivery.
The Signatera test is able to detect cancer recurrence 6-9 months earlier than standard imaging, so imagine what’s possible with further innovation.
The resulting chunks are analyzed in Amazon Bedrock using the Claude 3.5 Sonnet model to generate embeddings and perform document-level parsing and understanding.

Reported impact

Business outcome

Natera nearly doubled data extraction, from 25 to 46 elements per test, enabling more accurate patient care, and it achieved a 150% cost reduction in data processing. But, its work is far from over. Key reported metrics include Natera nearly doubled data extraction, from 25 to 46 elements per test, enabling more accurate patient care, and it achieved a 150% cost reduction in data processing.; Natera’s identified over 200 potential use cases for AI implementation with personalized healthcare delivery.; The Signatera test is able to detect cancer recurrence 6-9 months earlier than standard imaging, so imagine what’s possible with further innovation..

  • Cost Reduction
  • Customer Experience
  • Productivity Improvement
  • Speed to Market

Solution

Discover how Natera, a pioneer in cell-free DNA testing, used AWS to create a robust data and analytics platform and democratize data to deliver crucial tailored medical care for patients worldwide. Every year, over two million patients in the U.S. discover they have cancer for the first time, with over eighteen million patients suffering globally (source: National Institutes of Health World Health Organization ). In the case of Brent, a young husband and father sought care following a mountain bike accident that caused pain and discomfort. Unexpectedly, a test revealed atypical cells, and further examination found tumors. After their removal and biopsy, Brent was diagnosed with stage three bladder cancer. He started chemotherapy right away. As part of his treatment plan, Brent’s oncologist introduced him to Natera’s Signatera testing, which tracks cancer indicators over time.

Data flow

Source data is ingested through Amazon Textract, stored or indexed in Amazon RDS, processed with Amazon SageMaker, Amazon Bedrock, Claude 3, coordinated by AWS Step Functions, AWS HealthOmics, and delivered through Amazon Q.

Solution architecture

8 components · 5 layers
  1. Ingest
    • Amazon Textract Extracts structured information from business or clinical documents.
  2. Compute
    • Amazon SageMaker Supports model training, fine-tuning, evaluation, or managed inference for the customer workload.
    • Amazon Bedrock Hosts the foundation models and inference APIs that power the generative AI features.
    • Claude 3 Serves as the primary large language model for generation or analysis tasks.
  3. Storage
    • Amazon RDS Stores structured application or analytical data in a managed relational database.
  4. Serving
    • Amazon Q Provides conversational AI assistance for employees or end users.
  5. Orchestration
    • AWS Step Functions Coordinates multi-step application or data workflows.
    • AWS HealthOmics Runs large-scale biological workflows and pipeline execution for genomic or clinical data.

Architecture clues

  • AWS Step Functions ensures workflow reliability from dataset ingestion to output archival, with full traceability for troubleshooting—making the entire process automated and dependable.
  • Embeddings are stored in Amazon Relational Database Service (Amazon RDS) with vector search capabilities, optimized for semantic retrieval.
  • The platform handles compute, scalability, scheduling, and orchestration seamlessly, while Natera's scientists write their pipelines in Nextflow without worrying about underlying infrastructure.
  • The workflow management is handled by AWS Step Functions, which orchestrates the process flow from document upload.
  • This has enabled Natera to scale Signatera Oncology tests to serve over 250,000 patients, with HealthOmics now processing over 54,000 tests and 60,000 pipelines weekly across lab locations.
  • This pipeline enables more precise attribute extraction, increasing the number of extracted elements per test document from 25 to 46.

Evidence from the source

Natera nearly doubled data extraction, from 25 to 46 elements per test, enabling more accurate patient care, and it achieved a 150% cost reduction in data processing.
Natera’s identified over 200 potential use cases for AI implementation with personalized healthcare delivery.
The Signatera test is able to detect cancer recurrence 6-9 months earlier than standard imaging, so imagine what’s possible with further innovation.
The resulting chunks are analyzed in Amazon Bedrock using the Claude 3.5 Sonnet model to generate embeddings and perform document-level parsing and understanding.
This enables Natera to detect disease recurrence up to 9 months earlier than standard imaging—allowing patients to receive critical treatment sooner when it's most effective.
This pipeline enables more precise attribute extraction, increasing the number of extracted elements per test document from 25 to 46.
Transparency Evidence & methodology 95/100 reference value · 87% confidence · 1 source

Review how this reference was scored and trace each normalized claim back to its documented source.

01

Scoring methodology

How scores are calculated

Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.

Customer reference value95/100
  • Problem clarity20/20
    Measures whether the customer's operational problem is specific enough to compare.
  • Implementation depth20/25
    Rewards named components, data flow, integrations, and architecture detail.
  • Outcome specificity20/20
    Measures whether the reported benefit is concrete rather than a generic claim.
  • Quantified impact25/25
    3 metrics describe measurable impact.
  • Deployment scale10/10
    Rewards evidence of a live or scaled deployment and meaningful integration scope.
Evidence confidence87%
  • Source authority20/25
    The main evidence is an official vendor-published customer story.
  • Claim traceability21/25
    10 of 12 key fields link to a named source.
  • Direct evidence20/20
    6 direct source quotes retained.
  • Date provenance8/10
    The publication date comes from publisher-controlled metadata.
  • Source transparency8/10
    1 source reference document publisher, URL, authority, retrieval, and supported fields.
  • Implementation specificity10/10
    Rewards source-backed components, data flow, quantified impact, and a specific outcome.
02

Traceability

Sources & provenance

Field coverage shows which facts each source supports. Dates are shown only when documented by the source.

  1. AWS

    Natera customer story

    First-party
    https://aws.amazon.com/solutions/case-studies/natera-case-study/
    Source type
    Vendor-published customer story
    Published
    Jan 8, 2025
    Supports
    • Customer
    • Business problem
    • Solution
    • Products
    • Solution components
    • Data flow
    • Integrations
    • technical area
    • use case category
    • Business outcome
    • outcome category
    • Metrics
    • Architecture
    • evidence quotes
    • Publication date

Date evidence

Not separately sourced Industry, Region, Company size. Treat these normalized attributes as lower-confidence until a dedicated source is added.

Related

Similar use cases

See all like this
AWS Nov 14, 2024 Reveleer Healthcare United States To address these challenges, Reveleer is using Amazon Web Services (AWS) technologies. Value-based care is a healthcare model that prioritizes quality over quantity, focusing on improving patient outcomes and reducing… The Reveleer platform has a 100 percent uptime, and 90 percent of responses are served in under 8 seconds.… Document ProcessingRisk Management Amazon Bedrock · Amazon Comprehend Medical AWS Oct 1, 2024 Eka Care Healthcare India Eka Care needed to reduce the clinical time lost to administrative work, structure large volumes of unstructured medical data, and preserve compliance while scaling digital healthcare services for doctors and patients… just 3 minutes per patient Document Processing Amazon Bedrock · Amazon Transcribe AWS Aug 21, 2024 Pfizer Healthcare Global With Pfizer teams highly focused on ongoing technology projects, PACT addresses challenges around the bandwidth needed for prototyping. “With this collaboration, AWS brings in its prototyping teams to work with business… 16,000 hours Document ProcessingEnterprise Knowledge Search Amazon Bedrock · Amazon Kendra · Amazon SageMaker