Exclusive Content:

Haiper steps out of stealth mode, secures $13.8 million seed funding for video-generative AI

Haiper Emerges from Stealth Mode with $13.8 Million Seed...

Running Your ML Notebook on Databricks: A Step-by-Step Guide

A Step-by-Step Guide to Hosting Machine Learning Notebooks in...

“Revealing Weak Infosec Practices that Open the Door for Cyber Criminals in Your Organization” • The Register

Warning: Stolen ChatGPT Credentials a Hot Commodity on the...

Enhancing Medical Content Review at Flo Health with Amazon Bedrock (Part 1)

Revolutionizing Medical Content Management: Flo Health’s Use of Generative AI

Introduction

In collaboration with Flo Health, we delve into the rapidly advancing field of healthcare science, highlighting the crucial role of accurate medical information in well-being.

The Challenge of Medical Content Accuracy

Maintain medical accuracy is vital, especially when individuals seek health information. With millions of users relying on Flo Health for women’s health articles, the need for an efficient verification process becomes paramount.

Introducing MACROS: A Game-Changer for Medical Content

Through partnership with the AWS Generative AI Innovation Center, Flo Health introduces the Medical Automated Content Review and Revision Optimization Solution (MACROS), leveraging AI to ensure content remains up-to-date and accurate.

Objectives and Success Criteria of the Proof of Concept

Key objectives include validating generative AI’s feasibility for content verification, improving efficiency and accuracy compared to manual reviews.

Overview of the MACROS Solution

Exploring the components and capabilities of MACROS, highlighting its seamless integration with Flo Health’s existing systems.

Architecture and Workflow Description

An inside look at MACROS’ architecture, detailing the flow of data and processes involved in content review and revision.

Future Enhancements for Improved Efficiency

Considerations for future iterations of MACROS focused on streamlining AI-driven workflows.

Content Review and Revision Mechanics

A detailed examination of how the system reviews and revises medical content efficiently.

Operating Modes for Diverse Content Management Needs

Explaining the different operating modes that MACROS offers for both detailed and bulk content processing.

Extracting Actionable Guidelines from Unstructured Data

Introducing the Rule Optimizer feature that refines guidelines from unstructured medical documents.

Implementation Considerations and Lessons Learned

Insights from the PoC development process, covering data preparation, cost management, and regulatory compliance.

Preliminary Results of the PoC

Reviewing key metrics that illustrate the substantial benefits of AI-assisted medical content review.

Key Takeaways for Future Projects

Summarizing important insights for optimizing AI in medical content management.

Conclusion and Next Steps

Highlighting the potential of generative AI for improving medical content review processes, with a teaser for Part 2 focusing on real-world implementation challenges.

Meet the Team Behind the Innovation

Introducing the experts from AWS Generative AI Innovation Center who contributed to this groundbreaking project.

Transforming Medical Content Accuracy with Generative AI: A Journey with Flo Health

This blog post is based on work co-developed with Flo Health.

Healthcare science is rapidly advancing, and the need for accurate, up-to-date medical information has never been more critical. When individuals seek health information, they often do so during vulnerable moments, making the accuracy of the data not just important but potentially life-saving.

Flo Health stands at the forefront of this mission, creating thousands of medically credible articles each year that serve millions of users worldwide with valuable information on women’s health. However, ensuring the accuracy and relevance of such a vast content library poses a significant challenge. Given the continual evolution of medical knowledge, manually reviewing each article is both time-consuming and susceptible to human error.

To tackle this challenge, Flo Health has partnered with the AWS Generative AI Innovation Center to develop the Medical Automated Content Review and Revision Optimization Solution (MACROS). This innovative AI-driven approach aims to streamline the verification and maintenance of medical information accuracy at scale.

The MACROS Solution: An Overview

MACROS is designed to automate the processing of large volumes of medical content from credible scientific sources. Its capabilities include:

  • Identifying Potential Inaccuracies: Verifying the relevance of existing content based on the latest medical research and guidelines.
  • Proposing Updates: Incorporating user feedback while ensuring adherence to current medical standards.
  • Efficient Content Evaluations: Enabling rapid reviews that significantly reduce the burden on human experts.

Using Amazon Bedrock, MACROS allows Flo Health to conduct thorough content assessments efficiently, ensuring that the information remains accurate and facilitates informed healthcare decisions.

The Journey Ahead: A Two-Part Series

This blog series will explore Flo Health’s exciting journey with generative AI for medical content verification.

  • Part 1: Delving into the proof of concept (PoC), we’ll highlight initial solutions, their capabilities, and the early results achieved.
  • Part 2: We will shift our focus to the challenges of scaling and real-world implementation.

Each part of the series can stand alone while together illustrating the transformative potential of AI in medical content management.

Setting Success Metrics for Proof of Concept

Before proceeding with the technical solution, clear objectives were established for the PoC:

Key Objectives:

  • Validate the feasibility of using generative AI for medical content verification.
  • Compare accuracy levels with traditional manual reviews.
  • Assess improvements in processing time and cost.

Success Metrics:

  • Accuracy: Achieving a content piece recall of 90%.
  • Efficiency: Reducing detection time from hours to minutes per guideline.
  • Cost Reduction: Decreasing expert review workload.
  • Quality: Maintaining Flo’s editorial standards and ensuring medical accuracy.
  • Speed: Aiming for a process that is 10 times faster than manual reviews.

The collaboration among Flo Health’s medical experts, content teams, and AWS specialists was crucial in fine-tuning the AI model’s performance. This partnership led to the development of MACROS, a custom-built solution for AI-assisted medical content verification.

Content Review and Revision: The Heart of MACROS

The MACROS solution comprises two main processes:

  1. Content Review and Revision: Allows for adherence to medical standards and style guidelines at scale.
  2. Rule Optimization: Accelerates the extraction of new medical guidelines from research, ensuring quality and relevance.

Both processes can be conducted through a user-friendly interface or directly via API calls, allowing for integration with Flo Health’s existing tech infrastructure for seamless updates.

Challenges and Lessons Learned

As Flo Health progressed through the PoC, several challenges emerged:

  • Data Preparation: Standardizing input formats was essential for effective validation across diverse medical topics.
  • Cost Management: Optimizing token usage and prompt design helped balance performance with cost efficiency.
  • Regulatory Compliance: Maintaining strict oversight and documentation practices was vital due to the sensitive nature of medical content.
  • Model Optimization: Leveraging Amazon Bedrock’s capabilities for diverse model selection proved beneficial in achieving cost efficiency without sacrificing accuracy.

Key Takeaways

The implementation of MACROS not only streamlined the medical content review process but reinforced the importance of human-AI collaboration. Regular expert feedback was crucial for refining the system, highlighting that AI works best as an augmentation tool rather than a replacement for human expertise.

Conclusion and Next Steps

As we conclude the first part of our series, we showcase how generative AI can revolutionize the medical content review process, enhancing speed and efficiency while maintaining high standards of accuracy.

Stay tuned for Part 2, where we will explore the production journey, addressing challenges and strategies for effective scaling. Are you ready to move your AI initiatives into production?


About the Authors

Liza (Elizaveta) Zinovyeva, Ph.D. – Applied Scientist at AWS Generative AI Innovation Center, based in Berlin.

Callum Macpherson – Data Scientist at the AWS Generative AI Innovation Center, specializing in AI-based business transformations.

Arefeh Ghahvechi – Senior AI Strategist at the AWS GenAI Innovation Center, focusing on rapidly realizing generative AI value.

Nuno Castro – Sr. Applied Science Manager with extensive experience in AI and ML.

Dmitrii Ryzhov – Senior Account Manager at AWS enabling digital-native companies through AI and cloud technologies.

Nikita Kozodoi, Ph.D. – Senior Applied Scientist at AWS Generative AI Innovation Center, driving impactful AI solutions.

Aiham Taleb, Ph.D. – Senior Applied Scientist specializing in generative AI applications across industries.

Latest

Identify and Redact Personally Identifiable Information with Amazon Bedrock Data Automation and Guardrails

Automated PII Detection and Redaction Solution with Amazon Bedrock Overview In...

OpenAI Introduces ChatGPT Health for Analyzing Medical Records in the U.S.

OpenAI Launches ChatGPT Health: A New Era in Personalized...

Making Vision in Robotics Mainstream

The Evolution and Impact of Vision Technology in Robotics:...

Revitalizing Rural Education for China’s Aging Communities

Transforming Vacant Rural Schools into Age-Friendly Facilities: Addressing Demographic...

Don't miss

Haiper steps out of stealth mode, secures $13.8 million seed funding for video-generative AI

Haiper Emerges from Stealth Mode with $13.8 Million Seed...

Running Your ML Notebook on Databricks: A Step-by-Step Guide

A Step-by-Step Guide to Hosting Machine Learning Notebooks in...

VOXI UK Launches First AI Chatbot to Support Customers

VOXI Launches AI Chatbot to Revolutionize Customer Services in...

Investing in digital infrastructure key to realizing generative AI’s potential for driving economic growth | articles

Challenges Hindering the Widescale Deployment of Generative AI: Legal,...

Create an AI-Driven Website Assistant Using Amazon Bedrock

Building an AI-Powered Website Assistant with Amazon Bedrock Introduction Businesses face a growing challenge: customers need answers fast, but support teams are overwhelmed. Support documentation like...

Migrate MLflow Tracking Servers to Amazon SageMaker AI Using Serverless MLflow

Streamlining Your MLflow Migration: From Self-Managed Tracking Server to Amazon SageMaker's Serverless MLflow A Comprehensive Guide to Optimizing MLflow with Amazon SageMaker AI Migrating Your Self-Managed...

Introducing Visa Intelligent Commerce on AWS: Empowering Agentic Commerce with Amazon...

Revolutionizing Commerce: The Emergence of Agentic AI with Visa and AWS Transforming Commerce with Agentic AI: A New Frontier Co-written with Sangeetha Bharath and Seemal Zaman...