Clinical documentation · Production

Clario enhances clinical trial documentation quality with Amazon Bedrock

The problem

Clario's process of generating multiple clinical trial documents from the imaging charter was time-consuming, taking weeks, and was prone to inadvertent manual errors, inconsistencies, and redundant information that could delay or negatively impact clinical trials.

Workflow diagram · grounded in source
1
On-premises document preprocessing
trigger
“Charter-derived documents are processed in an on-premises script in preparation for uploading.”
2
Secure transfer to AWS
integration
“Files are sent to AWS using AWS Direct Connect.”
3
Document chunking and embedding
ai_action
“The script chunks the documents and calls an embedding model to produce the document embeddings. It then stores the embeddings in an OpenSearch vector database for retrieval by our application. Clario uses an Amazon Titan Text Embeddings…”
4
BRS generation via Claude
ai_action
“Uses a global specification that stores the prompts to be used as input when calling the large language model. Queries OpenSearch for the relevant Imaging charter. Loops through every business requirement. Calls the Claude 3.7 Sonnet lar…”
5
Human review of generated BRS
human_review
“a business requirement writer can review the answers to produce a final document”
6
Document output to S3
output
“The final documents are written to Amazon S3 to be consumed and published by additional document workflows that will be built in the future.”
7
AI chat for document discovery
ai_action
“An as-needed AI chat agent to allow document-based discovery and enable users to converse with one or more documents.”
Reported outcome

The prototype solution significantly streamlined the BRS generation process, minimizing translation errors and inconsistencies and reducing the need for rework and study delays, with productionization planned during 2025.

Reported metrics
BRS generation process efficiencystreamlined the complicated BRS generation process significantly
Translation errors and inconsistenciesminimized the risk of translation errors and inconsistencies
Rework and study delaysreducing the need for rework and study delays
Clinical trial documentation generation time (baseline)can take weeks
Reported stack
Amazon BedrockAmazon S3AWS Direct ConnectAnthropic's Claude
Source
https://aws.amazon.com/blogs/machine-learning/clario-enhances-the-quality-of-the-clinical-trial-documentation-process-with-amazon-bedrock?tag=soumet-20
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

The prototype solution significantly streamlined the BRS generation process, minimizing translation errors and inconsistencies and reducing the need for rework and study delays, with productionization planned during 2…

What tools did this team use?

Amazon Bedrock, Amazon S3, AWS Direct Connect, Anthropic's Claude.

What results were reported?

BRS generation process efficiency: streamlined the complicated BRS generation process significantly; Translation errors and inconsistencies: minimized the risk of translation errors and inconsistencies; Rework and study delays: reducing the need for rework and study delays; Clinical trial documentation generation time (baseline): can take weeks (source-reported, not independently verified).

How is this clinical documentation AI workflow structured?

On-premises document preprocessing → Secure transfer to AWS → Document chunking and embedding → BRS generation via Claude → Human review of generated BRS → Document output to S3 → AI chat for document discovery.

Related clinical documentation cases