AWS Partner solutions: AI-based insight transforms radiology workflows (PEX218)

Summary of the Video Transcription

Introduction

  • The presenters are Gavin Phillips, Worldwide Partner Leader at AWS covering Life Sciences partner Solutions, and Brian Warwick, Principal Solutions Architect for Healthcare and Life Sciences at AWS.
  • They will discuss how AI-based insights are transforming Radiology workflows.

Challenges in the Medical Imaging Space

  • File sizes and types vary based on different imaging modalities (e.g., MRI, ultrasound, PET scan).
  • Challenges around storage, transmission, and processing of large amounts of data.
  • Radiologists face the challenge of evaluating thousands of image slices per study, which can take a significant amount of time.

Adoption of AI-ML in Radiology

  • As of August 2022, the FDA had listed 950 AI/ML-enabled medical devices, with 76% of them falling into the Radiology and Medical Imaging category.
  • This indicates a significant amount of innovation and transformation happening in the Radiology space.

Radiologist Perspective on AI-ML

  • Based on a survey from the American Medical Association, clinicians (including radiologists) see the biggest opportunity for automation in administrative burdens and tasks.
  • Radiologists need the right orders, views, and artifacts to provide accurate diagnoses and improve patient care.

Typical Radiology Workflow

  1. Physician orders a scan for a patient, which is stored in the electronic medical record or radiology information system.
  2. The scan is performed, and the images are stored in a PACS (Picture Archiving and Communication System).
  3. Radiologists access the images and pull in additional information from the electronic medical record to evaluate the scans.
  4. Radiologists create a summary of their findings, which is then pushed back to the PACS or electronic medical record.

AI-ML Use Cases in Radiology

  • Examples include automating report generation, leveraging ancillary data from the electronic medical record, and creating comprehensive summaries for downstream specialists or patients.

AWS Services and Partner Solutions

  • AWS offers several high-trust certified services for healthcare and life sciences, such as Amazon Health Lake, Health Omics, Health Scribe, and Health Imaging.
  • Partner solutions built on AWS Bedrock demonstrate the use of prompt engineering, Retrieval-Augmented Generation (RAG), and custom large language models to address various radiology workflow challenges.

Next Steps

  • Attendees are encouraged to explore the AWS services mentioned, the generative AI partner playbook, and to reach out to the presenters for assistance in addressing their specific challenges.

Your Digital Journey deserves a great story.

Build one with us.

Cookies Icon

These cookies are used to collect information about how you interact with this website and allow us to remember you. We use this information to improve and customize your browsing experience, as well as for analytics.

If you decline, your information won’t be tracked when you visit this website. A single cookie will be used in your browser to remember your preference.

Talk to us